Slow Delayed Rectifier Potassium Current (I_Ks) and the Repolarization Reserve

Abstract

The aim of this review is to present the properties of the slow component of the delayed rectifier potassium current (I_Ks) in the human ventricle. The review gives a detailed description of the physiology, molecular biology and pharmacology of the I_Ks current, including kinetic properties, genetic structures, agonists and antagonists. The authors also present the role of the I_Ks current in the human cardiac repolarization focusing on several pathophysiological situations, such as the LQT syndrome and the Torsade de Pointes arrhythmia.

INTRODUCTION

Action potential of the cardiac ventricular muscle (Fig. 1), which normally shows a more than a hundred millisecond plateau phase, provides the opportunity of the heart for rhythmic contraction to pump blood through the vessels to supply the different organs. This unique electrophysiological property of the ventricle, unlike other excitable tissues like nerve or skeletal muscle lacking plateau phase, is due to the simultaneously opening and closing transmembrane ion channels providing the possibility of positively and negatively charged ions to pass, and thereby carrying inward and outward currents through the cell membrane. Inward depolarizing currents in the heart are mainly carried by sodium and calcium, whose ions move from the extracellular space to the cytosol. Outward repolarizing currents mainly but not exclusively are carried by potassium ions that flow from the inside of the myocytes to the extracellular space. In the cardiac muscle, there are several potassium channels, which distinctly differ from each other in respect of activation range, channel gating kinetic properties, and pharmacological response to various drugs. Changes or modulation of the potassium channels may lead to both antiarrhythmic and proarrhythmic consequences, therefore, better understanding of the function of these channels has great importance; it is a necessity in the therapy or prevention of life‐threatening cardiac arrhythmias.

Figure 1.

Action potential waveforms and underlying ionic currents in human atrial (left) and ventricular (right) myocytes. I_to= transient outward current; I_Kur= ultrarapid delayed rectifierpotassium current; I_Kr= rapid delayed rectifierpotassium current; I_Ks= slow delayed rectifierpotassium current; I_K1= inward rectifierpotassium current; I_K,ACh= acetylcholine‐sensitive potassium current; I_K(ATP)= ATP‐dependent potassium current. Structural channel protein components are given for all currents. Cartoons show schematic current profileof the corresponding potassium current.

ELECTROPHYSIOLOGICAL PROPERTIES OF THE I_Ks CURRENT

The delayed rectifier potassium current (I_K) is a major outward current responsible for ventricular muscle action potential repolarization. ¹ , ² This current was first described by Noble and Tsien using the two‐microelectrode voltage‐clamp technique in sheep cardiac Purkinje fibre strands. ³ Since its discovery, it has been examined in single isolated myocytes obtained from various regions of the heart in several mammalian species. ⁴ , ⁵ , ⁶ , ⁷ In most species, I_K can be separated into rapid (I_Kr) and slow (I_Ks) components that differ from one another in terms of their sensitivity to drugs, rectification characteristics, and kinetic properties. ² , ⁴ , ⁶ , ⁸ , ⁹ , ¹⁰

In dog cardiac ventricular cells, it was shown that I_Ks activated slowly (τ≈800 ms) at voltage more positive than 0 mV, and deactivated rapidly and monoexponentially (τ≈150 ms). ⁶ In the guinea pig, where it was first described, I_Ks activated very slowly, not saturating even after 5–7 s at +50 mV, and deactivated slowly, within 500–1000 ms. ¹¹ First, it was not observed in the rabbit, ² but later studies revealed a large and consistent I_Ks in rabbit ventricle as well. ⁷ , ¹²

Because of the species differences regarding the existence and properties of I_Ks, it would only be proper to ask as to how these findings can be extrapolated to humans. In view of the known difficulties in obtaining human tissue in general and undiseased human ventricular tissue for research in particular, few studies have so far characterized I_Ks in human ventricle. Some of the available data on I_Ks have been obtained in ventricular myocytes dissociated from diseased human hearts. Beuckelmann et al. ¹³ reported that I_Ks was absent or hardly detectable in human left ventricular myocytes. Li et al. ¹⁴ described both I_Kr and I_Ks in right ventricular myocytes obtained from explanted diseased human hearts. In this study, I_Ks was examined using CdCl₂ to eliminate the inward Ca current (I_Ca), and BaCl₂ to block the inward rectifier potassium current (I_k1). ¹⁴ Since Cd²⁺ substantially changes the kinetic properties and amplitude of I_Ks, ⁵ and Ba²⁺, due to the voltage dependent block and unblock of I_K1 channels, ¹⁵ , ¹⁶ elicited a tail current similar to I_Ks tail current, ¹⁶ the results of the above study by Li et al. regarding the properties of I_Ks ¹⁴ must be interpreted with caution.

In cardiac ventricular myocytes isolated from undiseased human donor hearts, I_Ks exhibited slow and voltage independent activation at more positive than 0 mV (τ≈ 900 ms, Fig. 2A,B) and had fast and monoexponential deactivation kinetics (Fig. 2D). ¹⁶ This deactivation was voltage‐dependent, being faster at more negative (at ‐50 mV, τ≈ 90 ms), and slower at more positive voltages (at 0 mV, τ≈ 350 ms) (Fig. 2D). ¹⁶

Figure 2.

The properties of slow delayed rectifierpotassium current I_Ks in undiseased human ventricular myocytes. Panel A. The cell sized normalized current–voltage relation of I_Ks tail currents. Vertical axis represents peak tail current normalized to cell size measured in pA/pF, while horizontal axis gives the membrane potential measured in millivolts. Inset shows the applied voltage protocols, n indicates the number of experiments. Panels B–D. The voltage dependence of activation and deactivation kinetics of human I_Ks. Panel B shows original I_Ks tail current activation kinetic traces measured at depolarizing voltages to +30 mV and after returning to −40 mV. Panels C and D show the voltage dependence of the activation (C) and deactivation (D) time constants. Vertical axis: The activation and deactivation time constants (τ) in milliseconds; Horizontal axis: The membrane potential measured in millivolts. Number in parentheses represents the number of experiments at each voltages (from Ref. 16, used with permission).

I_Ks is known to be enhanced by elevation of the intracellular cAMP level, therefore the current is augmented during increased sympathetic tone. ¹⁷ , ¹⁸

Considering the kinetic properties, the human cardiac slow delayed rectifier potassium current ¹⁶ best resembles those measured in the dog ⁶ , ⁸ and rabbit ⁷ , ¹² ventricle, but significantly differs from those found in the guinea pig heart, ⁴ , ⁹ , ¹⁰ where the density of I_Ks is very large. The kinetic properties of I_Ks also differ from each other in the rabbit, dog, or human ventricle (Fig. 3, own unpublished results). Nevertheless, in this respect, the dog and rabbit seem to be more suitable species for preclinical cardiac electrophysiological evaluation of new drugs than the guinea pig.

Figure 3.

The activation (upper panels) and deactivation (bottom panels) kinetics of I_Ks in human, dog, rabbit, and guinea pig ventricular myocytes. Currents were measured with patch‐clamp technique as described in 12, 16, 49, and 59. Activation kinetics was studied by applying the envelope of tails protocols (see inset) from the holding potential of −40 mV. Insets show the value of activation or deactivation time constants (τ) determined by fitting the activation and deactivation curves with exponential functions. n = number of experiments. (Unpublished results from the present authors.)

MOLECULAR BIOLOGY OF THE I_Ks CURRENT

The structural organization of the ion channel mediating I_Ks was subject of extensive studies during the last decade, and at the very beginning it was controversial. Initially, it was suggested that this current was created by MinK. ¹⁹ , ²⁰ The name itself means that minimum sequence required for a K‐current. ²⁰ This gene (KCNE1) encodes a protein (MinK or I_sK) that contains only 130 amino acids and has only one single membrane‐spanning domain with an extracellular N‐terminus. ¹⁹ MinK mRNA has been found in the mouse ¹⁹ and neonatal rat ²¹ heart, and the protein has been detected immunohistochemically in guinea pig ventricular myocytes. ²² This hypothesis was supported when expression of MinK protein in oocytes resulted in a current that resembled the slow delayed rectifier I_Ks. ²³ Moreover, altered ionic selectivity and regulation of the expressed minK current after mutagenesis of the transmembrane segment and the PKC consensus site seemed to support the idea that minK encoded the I_Ks channel. ²⁴ , ²⁵ Later, positional cloning identified a novel gene KCNQ1 responsible for the chromosome 11‐associated form of the congenital LQT1 syndrome. ²⁶ , ²⁷ However, the expressed KvLQT1 protein forms a channel conducting a current unlike any current previously identified in cardiac preparations. This was surprising, since mutations in KvLQT1 cause the LQT1 syndrome known to be related to defects in the I_Ks channels. ²⁶ , ²⁷ Parallel two groups led by Sanguinetti ²⁷ and Lazdunski ²⁸ found and published in Nature that coexpression of KvLQT1 and MinK yielded a current that corresponds to the native slow I_Ks component. These observations, together with the biochemical data demonstrating that heterologously expressed KvLQT1 and MinK proteins can associate, have been interpreted as suggesting that MinK coassembles with KvLQT1 to form functional cardiac I_Ks channels (Fig. 4). In addition, the finding that mutations in the transmembrane domain of MinK alter the properties of the KCNQ1‐encoded Kv channels was interpreted as suggesting that the transmembrane segment of MinK contributes to the channel pore. ²⁴ , ²⁹ , ³⁰ Thus, MinK controversy was ended by the demonstration that it acts only as β‐subunit that alters the intrinsic gating of KvLQT1; the I_Ks like MinK current in oocytes was due to its association with an endogenous Xenopus XKvLQT1 subunit. ²⁷

Figure 4.

Transient expression of KvLQT1, IsK/MinK, and KvLQT1 + I_sK/MinK in transfected chinese hamster ovary (CHO, Panel a) or african green monkey kidney (COS, Panel b) cells induce a current nearly identical to cardiac I_Ks. KvLQT1is the pore forming, while MinK/I_sK is the auxiliary subunit of the I_Ks current. [from Ref. 27 (Panel a) and Ref. 28 (Panel b), used with permission].

KvLQT1 protein (KCNQ1 gene) is the pore‐forming subunit of the I_Ks channel complex with cytoplasmic N‐ and C‐terminal domains (Fig. 5). ³¹ , ³² The subunit consists of six transmembrane domains with helical structure. The pore region is the intervening loop inserted between segments 5 and 6, which is the responsible for the ionic selectivity and the conduction. ³³ The amino acid sequence of this region (TxGYG) is highly conservative among Kv channels. The S domain contains numerous positively charged amino acids and plays a role as a voltage sensor for voltage dependent activation kinetics. ³³ Two splice variants of the KvLQT1 subunit have been identified, the isoform‐1 and isoform‐2. ³⁴ Isoform‐2 is an N‐terminally truncated version of the isoform‐1, having suppressive effects on the current generated by the full length KvLQT1 protein. Five different tetramers can be constructed from the two isoforms: two homotetramers and three heterotetramers. The maximum conductance as well the time constants for activation were shown to be different for these channels suggesting that the level of isoforms‐2 expression can be important for the I_Ks density. ³⁵ For instance, the transmural heterogeneity of in repolarization was attributed to heterogeneous I_Ks channel density. KvLQT1 subunit alone forms voltage gated potassium channel with rapid activation kinetics, but when the MinK subunit binds to the KvLQT1 unit, the activation slows dramatically and the voltage dependence of activation shifs toward positive potentials. ³⁶

Figure 5.

Schematic representation of the principal subunits to form I_Ks channel. Panel A. Relation of the two major subunits (KvLQT1 and MinK) in the cell membrane. Panel B. Formation of the aqueous pore by four subunits. In both panels, numbers indicate the six alpha helical domains traversing the cell membrane. EC and IC mean extracellular and intracellular region, respectively.

The MinK protein (KCNE1 gene) named also as I_sK consists of 126–130 amino acids with a single transmembrane segment. The N‐terminal is N‐glycosylated and located on the extracellular side of the membrane. ¹⁹ Subunit MinK cannot form a pore alone, it only modifies the kinetic properties of the KvLQT1 subunit, when its transmembrane domain interacts with the pore‐forming segment of KvLQT1. ²⁷ There are several other known members of the KCNE family (KCNE2, KCNE3 KCNE4, and KCNE5), which also interact with KCNQ1, influencing the behavior of the I_Ks current, and with other Kv family currents (Kv4.3 or HERG). ³³ , ³⁷ , ³⁸ Mutations in MinK protein may also lead to development of the Long‐QT (LQT4 and LQT5) syndrome. ³⁹ , ⁴⁰ , ⁴¹

Table 1 summarizes the gene and protein structure forming the I_Ks channel.

Table 1.

The I_Ks Channel Pore Forming α and Auxiliary β‐Subunits

Subunit	Protein	Gene	Locus in Human	Current
Alpha (pore)	KvLQT1	KCNQ1	11p15	IKs
Beta (auxiliary)	MinK, IsK	KCNE1	21q22	IKs

PHARMACOLOGY OF THE I_Ks CURRENT

Several activators and blockers of I_Ks have been developed in the last decade. The failure of the I_Kr blockers and the Sanguinetti hypothesis, ¹¹ which proposed positive rate dependent prolongation after inhibition of I_Ks, turned the attention of the pharmaceutical industry to develop I_Ks blockers as new antiarrhythmics devoid of reverse rate dependent action potential duration (APD) lengthening effects. Until now, three compounds and their analogs have been described as selective blockers of the I_Ks channel.

The Hoechst/Aventis compound chromanol 293B has been reported as the first selective blocker of the I_Ks. ⁴² Several studies showed its efficacy on the I_Ks channel parallel with its purported Class III antiarrhythmic efficacy. ⁴³ , ⁴⁴ However, it was shown that chromanol 293B, at the concentration required for the full I_Ks block (about 30‐50 μM), affects I_to, another important repolarizing current, ⁴³ which made the interpretation that selective I_Ks block lengthens repolarization substantially, and thereby elicits antiarrhythmic effect, became rather uncertain. However, in these studies, the investigators used chromanol 293B concentrations up to 100–150 μM, were the agent looses its selectivity even regarding I_Kr. ⁴⁵ , ⁴⁶ In vivo studies with chromanol 293B were also contradictory. ⁴⁷ , ⁴⁸ Our group reported no discernible changes in QTc interval following administration of chromanol 293B in anaesthetised dogs ⁴⁹ and Langendorff perfused rabbit hearts, ¹² whereas others observed T wave abnormalities and QT prolongation. ⁴⁸ In the late nineties, by variation of the aromatic substituent, a more potent analog of chromanol, the HMR‐1556, was developed at Aventis. ⁵⁰ This compound ⁵¹ has a higher potency and better selectivity compared to chromanol. Its EC₅₀ value is in the submicromolar range (about 30–50 nM), ⁵¹ which made HMR‐1556 one of the most suitable tools for testing the properties of I_Ks current in vitro ⁵¹ , ⁵² , ⁵³ and in vivo ⁵⁴ , ⁵⁵ as well.

The second chemical classes of I_Ks channel blockers are benzodiazepine derivates synthesized by Merck‐Sharpe&Dohme Pharma (MSD). Several from this family of compounds, such as L‐735,821 ⁵⁶ and L‐768,673, ⁵⁷ proved to be very effective (EC₅₀ about 1–20 nM) and selective I_Ks blockers. ⁵⁸ In our own studies, L‐735,821 failed to lengthen significantly the rabbit, ¹² dog, ⁴⁹ and human ⁵⁹ APD, while in other studies, L‐768,673 lengthened the guinea pig ⁵⁸ and rabbit ⁶⁰ APD. L‐768,673 was shown to prevent certain types of arrhythmias in both anesthetized and conscious dogs. When L‐768.673 was combined with the beta‐adrenoceptor blocker timolol, additional antiarrhythmic effect was reported in dogs with myocardial infarction. ⁶¹

The third class of I_Ks blockers was recently developed by Bristol Myers Squibb. It is represented by the benzamide derivative BMS‐208782. The compound has a high potency to inhibit I_Ks, (EC₅₀ is about 9 nM), but no other meaningful pharmacological data have so far been reported regarding this compound. ⁶²

Several other drugs, in addition to blocking different transmembrane ion channels, inhibit the I_Ks current as well. For example, such drugs are amiodarone, ⁶³ azimilide, ⁶⁴ bepridil, ⁶⁵ cibenzoline, ⁶⁶ clofilium, ⁶⁷ imipramine, ⁶⁸ indapamide, ⁶⁹ mibefradil, ⁶⁵ propafenone, ⁷⁰ terfenadine, ⁷¹ tedisamil. ⁷²

Activators of I_Ks were also described. DIDS ³¹ and mefenamic acid, ³¹ , ⁷³ at relatively high concentrations (100–200 μM), are able to increase the amplitude of I_Ks, but these drugs at such a high concentrations affect also other transmembrane ion currents. A benzodiazepine derivate L‐364,373 (R‐L3) ⁷³ from MSD was first reported to enhance I_Ks in the guinea pig when applied in the submicromolar concentration range, but the effect seems dependent either on the species/tissue used for the investigation, or on other experimental conditions, since in later studies, the same compound showed less potency in the rabbit, ⁷⁴ or even lack of effectiveness in the dog ventricle. ⁷⁵

The therapeutic usefulness of I_Ks activators is uncertain, but one might speculate that enhancing I_Ks would strengthen the repolarization reserve, thereby decreasing the risk of torsade de pointes arrhythmias in situations where potassium channels are inhibited by different drugs or when the channels are downregulated (for example by heart failure, diabetes) or when these channels have loss of function due to genetic causes, like in congenital long QT syndromes. ⁷⁵ , ⁷⁶

THE ROLE OF THE I_Ks CURRENT IN THE HUMAN CARDIAC REPOLARIZATION

Previously, it was generally accepted that inhibition of I_Ks would result in a significant lengthening of repolarization, thereby providing Class III antiarrhythmic effect to avoid reverse‐rate dependent action potential prolongation. This was suggested by Jurkiewicz and Sanguinetti, ¹¹ who based this on their earlier observation in guinea pig ventricular myocytes, where the deactivation of I_Ks is slower than the normal length of the diastole. Consequently, significant accumulation of I_Ks could be expected during fast rate, i.e., short diastole. This accumulation of I_Ks was supposed to contribute to the repolarization‐shortening usually seen during short cycle length. Therefore, pharmacological block of I_Ks was believed to elicit more APD lengthening at fast than at normal or slow rates, and there has been an effort to develop selective I_Ks blockers as potential antiarrhythmic blockers agent without the risk of generating torsade de pointes arrhythmias. ¹¹

The absence of selective I_Ks blockers until the last decade made it impossible to directly evaluate the physiological role of I_Ks in determining cardiac action potential configuration. Early studies with selective I_Ks blockers provided controversial results as to how inhibition of I_Ks would affect repolarization in the ventricle. Some studies provided evidence that selective inhibition of I_Ks resulted in significant APD or QT prolongation, ⁴⁵ , ⁴⁶ , ⁴⁸ but other studies ⁴⁹ , ⁵⁴ did not confirm it. Repeated later investigations in the dog, ⁴⁹ rabbit, ¹² and human ⁵⁹ ventricle in the absence of sympathetic stimulation showed that inhibition of I_Ks by chromanol 293B, L‐735,821 and HMR‐1556 did not cause significant lengthening of repolarization (Fig. 6) at a wide range of stimulation frequency. The explanation for these unexpected results came from the careful experimental analysis of the properties of I_Ks itself. Since I_Ks activates slowly (τ≈ 900‐1000 ms) at more positive voltages, there is only very little current that can be expected to flow during the time and voltage courses of the plateau of the normal action potential, which is usually shorter than 200 ms, and seldom gets to more positive voltages than 20–30 mV. ¹² , ¹⁶ , ⁴⁹ , ⁵⁹ However, in situations where the APD is prolonged beyond normal, there is a possibility for more I_Ks activation to allow an important means of limiting excessive APD lengthening by a negative feedback mechanism (Fig. 7) providing more safety of the repolarization process. ¹² , ⁴⁹ , ⁵⁹ In the presence of enhanced sympathetic stimulation, however, the magnitude of I_Ks is increased, its activation speeds up, and shifted into more negative voltages. ⁵⁴ In addition, due to the simultaneous sympathetic enhancement of the L‐type calcium current (I_Ca), the plateau voltage of the action potential would shift upward in more positive direction, further increasing the activation of I_Ks. ⁷⁷ Therefore, I_Ks block in the presence of elevated sympathetic activity may result in a significant repolarization lengthening (Fig. 8). This latter effect may be further enhanced by the lack of the outward I_Ks, which does not oppose the augmented inward I_Ca current. ⁷⁷ In general, it can be concluded that the pharmacologic block of I_Ks does not provide antiarrhythmic benefit, but, on the contrary, it can decrease the safety of repolarization and consequently increase the proarrhythmic risk, especially if sympathetic tone is enhanced. ⁵⁴ , ⁵⁹

Figure 6.

Frequency‐dependent effect of selective I_Kr and I_Ks block on action potential duration (APD) in dog (Panel A), rabbit (Panel B), and human (panel C) ventricular papillary muscles. Abscissa = pacing cycle length in milliseconds; ordinate = percentile changes in APD₉₀[from Ref. 12 (Panel A), Ref. 49 (Panel B), and Ref. 59 (Panel C), used with permission].

Figure 7.

Panel A. L‐735,821 sensitive (I_Ks) currents in human ventricular myocytes after application of short (150 ms, upper) and long (500 ms, bottom) depolarizing test pulses. Insets show applied voltage protocols. Panel B. Average I_Ks currents at the end of a short (150 ms) a long (500 ms) depolarizing test pulses to +30 mV, and peak tail currents at −40 mV. Panel C. L‐735,821 sensitive (I_Ks) difference current recorded during an “action‐potential‐like” test pulse in human ventricular myocytes in the absence of any sympathetic agonist. The dotted line shows superposed the E‐4031 sensitive (I_Kr) current measured in similar condition [modified from Ref. 59, used with permission].

Figure 8.

Effect of I_Ks block in the presence of elevated sympathetic activity. I_Ks blocker HMR‐1556 frequency dependently lengthens APD in dog (Panel A) and human (Panel B) ventricular myocytes, when the repolarization reserve is attenuated either by sympathetic stimulation (by isoprenaline, ISO) or combined I_Kr block with stimulation (by dofetilide + adrenaline, DOF + ADR) [from Refs. 12 (Panel A), Ref. 54 (Panel A) and Ref. 59 (Panel B), used with permission].

BLOCK OF THE I_Ks CURRENT CONTRIBUTES TO THE DEVELOPMENT OF EXPERIMENTAL TORSADE DE POINTES MODELS

The recent observations made with pharmacological block of I_Ks have contributed to the development of experimental LQT1 and/or torsade de pointes models. Volders et al. ⁷⁸ reported that 3–4 weeks after AV ablation in dog, significant bradycardia and cardiac hypertrophy developed. In this model, several potassium currents, most notably I_Ks, were downregulated, and APD and QTc were prolonged. ⁷⁸ In these dogs, drug treatments very often caused torsade de pointes arrhythmias, making this experimental model very useful method to test the proarrhythmic potency of drugs, which interfere with repolarization in general. ⁷⁸ Similarly pharmacological block of I_Ks by chromanol 293B or HMR‐1556 in rabbit ⁷⁹ and dogs ⁴⁵ , ⁴⁶ , ⁸⁰ did not prolong QTc markedly, but made the heart susceptible toward eliciting torsade de pointes arrhythmias by I_Kr block with dofetilide. In these experiments, dofetilide alone did not or only occasionally induced torsade de pointes, but applying it after full I_Ks block, the drug increased the reverse‐rate dependent APD prolongation excessively, and the incidence of torsade de pointes arrhythmias were more than 70%, providing a useful model for safety pharmacology studies (Fig. 9). ⁷⁹ , ⁸⁰ Recently, it was also suggested the increased short‐term beat to beat APD/QT variability as a more sensitive measure of the proarrhythmic risk is related to decreased function of I_Ks. ⁸¹ Tachypaced rabbits showed marked downregulation of I_Ks, and developed torsade de pointes arrhythmias after dofetilide administration. On the contrary, bradypaced rabbits developed spontaneous torsade de pointes arrhythmia with the concomitant downregulation of both I_Ks and I_Kr channels. ⁸²

Figure 9.

Representative examples of Poincaré plots of the QT interval from an anesthetized rabbit treated with the I_Kr blocker dofetilide (25 μg/kg i.v., DOF) (Panel A), and from another one treated with the I_Ks blocker HMR 1556 (0.1 mg/kg i.v.) followed by dofetilide (25 μg/kg i.v.) (Panel B). Short‐term QT interval variability was increased after dofetilide administration. Panel C. Combined I_Kr and I_Ks block markedly increased QT variability and provoked torsade de pointes arrhythmia (TdP) (Panel D). ^#P < 0.05 versus control in the same group, *P < 0.05 versus dofetilide treatment, n = 7 animals/group. (Unpublished results from the present authors and Dr. I. Baczkó; partially presented as abstract at the Heart&Rhythm Meeting‐2006; Ref. 80.)

THE REPOLARIZATION RESERVE AND THE I_Ks CURRENT

The concept of “repolarization reserve” was suggested by Dan Roden. ⁸³ , ⁸⁴ , ⁸⁵ According to this terminology, the normal repolarization is accomplished by multiple different potassium channels providing a strong safety reserve for normal repolarization. Thus, under normal conditions, the pharmacological block or impairment of one single type of potassium channels does not necessarily lead to marked QT interval prolongation. However, in the presence of a subclinical impairment in the repolarization process, i.e., an otherwise mild potassium channel block (e.g., due to inhibition of the I_Ks current resulting in a decrease in repolarization reserve) ⁴⁹ , ⁵⁹ may precipitate marked QT prolongation, which can result in life‐threatening torsades de pointes arrhythmia.

CONGENITAL LONG QT SYNDROME AND I_KS CURRENT

Congenital LQT syndrome is rather infrequent in the general population (1/5000), and is identified by QT prolongation on the surface ECG during clinical evaluation of unexplained syncope. ⁸⁶ The cause of the syncope that occurs in this disorder is due to transient rapid polymorphic ventricular tachycardia known as torsades de pointes linked to delayed repolarization of the cardiac ventricular muscle. So far, seven forms of LQT (LQT1‐7) have been well characterized. They are caused by several hundred different mutations in the I_Ks, I_Kr, I_Na, I_K1 alpha and beta subunits. ⁸⁷ , ⁸⁸ , ⁸⁹ , ⁹⁰

“Among the three most common LQTS genotyped (LQT1‐3), LQT1 probably hosts the largest percentage of genotype‐positive individuals displaying a normal/borderline resting QTc” as described in a recent review of Ackerman. ⁹¹ Similarly, simple pharmacological modulation of I_Kr, I_Na and I_K1 channels also prolongs ventricular repolarization resulting in drug‐induced LQT. ⁹² In contrast, as mentioned before, even full pharmacological block of I_Ks does not lengthen repolarization in the ventricle, unless sympathetic tone is enhanced ⁵⁹ , ⁹³ or repolarization had been delayed previously by other means. ⁴⁹ Based on these observations, it was concluded that I_Ks (KvLQT1 + MinK) channel is not a major contributing factor to “normal” repolarization but it is an important source of the repolarization reserve that opposes excessive lengthening of APD and consequently protects against torsades de pointes arrhythmia during possible impairment or change in normal function of other transmembrane ion channels. A recent study from Boulet et al. ⁹⁴ describes a loss of function mutation in the KCNQ1 gene underlying the pore‐forming unit of I_Ks channel in a 40‐year‐old women. She had experienced torsade de pointes arrhythmia with QTc interval of 430 ms (which is well within the normal range in women), and this may also indicate that a genetic loss of I_Ks function does not necessarily prolong repolarization, as shown by the normal QTc of the patient. However, under some unfavourable conditions (e.g., hypokalemia, drug effects, downregulation of potassium channels), the impairment of the repolarization reserve could not provide the necessary protection, making this patient more vulnerable toward arrhythmia than those who lack defective I_Ks channels. In accordance with this speculation, Kääb et al. ⁹⁵ reported that patients who experienced torsades de pointes arrhythmia with QT‐prolonging drugs developed more QTc lengthening after i.v. sotalol, an I_Kr blocking drug, than those of the control group consisting of patients without history of torsades de pointes. The interesting observation in this study was that in both groups the baseline QTc was normal and did not differ from each other (Fig. 10). ⁹⁵ Although genetic testing has not been carried out in these patients, it can be assumed that the individuals who responded to sotalol might have “subclinical‐concealed‐silent” LQTS due to defective I_Ks channels. Based on these results, the authors suggested that the administration of provocative drug test under controlled situation might help identify selected patients at risk for developing torsades de pointes arrhythmia.

Figure 10.

Individual QTc intervals in human control and study groups before and after sotalol. The dotted line indicates a cutoff value of 480 ms that distinguished best between the study and the control group (from Ref. 95, used with permission).

DOWN‐REGULATION OF I_KS AS A POSSIBLE LINK TO DECREASED REPOLARIZATION RESERVE

I_Ks can be decreased not exclusively by ion channel mutation but it can be down‐regulated due to diseases such as heart failure, diabetes, and cardiac hypertrophy. Also, it is not known whether possible gene polymorphisms or simply the variation of the expression level of this channel in normal individuals play a similar role. This can be an important point, since even large variations in the I_Ks density cannot be expected to influence normal QTc duration significantly on the surface ECG, but they may substantially determine repolarization reserve and the stability of repolarization. It is of interest to note in this context, based on a recent study, that the coexpression of KvLQT1 cDNA with HERG cDNA increased the current‐carrying properties and trafficking of I_Kr channels. ⁹⁶ In other words, it is possible that there is an alpha subunit interaction between I_Ks and I_Kr and such changes at the expression level of I_Ks can secondarily alter I_Kr. It would be interesting to know how mutated KvLQT1 channels could behave in this setting.

CONCLUSION

The slow delayed rectifier potassium current (I_Ks) operates in mammalian ventricular muscle including human. In normal situation, this current contributes to the repolarization process only to a minimal extent, but has a vital role in the repolarization reserve. Accordingly, when repolarization is prolonged beyond normal and/or sympathetic tone is elevated, I_Ks current provides an important safety mechanism preventing excessive and dangerous repolarization lengthening. Therefore, down‐regulation, genetical loss of function or pharmacological inhibition of I_Ks is not manifested in a marked repolarization elongation, but makes the repolarization less stable and the heart vulnerable toward repolarization abnormalities and consequently torsades de pointes arrhythmias.