OUP user menu

Genetic control of sodium channel function

Hanno L Tan, Connie R Bezzina, Jeroen P.P Smits, Arie O Verkerk, Arthur A.M Wilde
DOI: http://dx.doi.org/10.1016/S0008-6363(02)00714-9 961-973 First published online: 15 March 2003

Abstract

Sodium ion (Na) influx through cardiac Na channels triggers the action potential in cells of the working myocardium and the specialized conduction system. Na channels thus act as key molecular determinants of cardiac excitability and impulse propagation. Na channel dysfunction may cause life-threatening arrhythmias. Here, we review the ways in which Na channel function can be aberrant due to genetic changes. We discuss how biophysical studies of mutant Na channels combined with precise clinical phenotyping may improve our understanding of Na channel function in health and disease and may be useful as a model from which to derive improved treatment strategies for common disease.

Keywords
  • Arrhythmia mechanisms
  • Long QT syndrome
  • Na-channel
  • Sudden death
  • Ventricular arrhythmias

Time for primary review 34 days.

Na channel dysfunction, present in various clinical conditions (ischemia, drug use, electrolyte imbalance), may cause life-threatening arrhythmias by various electrophysiological mechanisms. Reduced Na current during ischemia (secondary to depolarization of the resting membrane potential) may cause conduction slowing or block and evoke reentrant tachyarrhythmias [1]. The excess mortality in patients with active ischemia who received Na channel blocking antiarrhythmic drugs in the CAST study [2,3] suggests that the combined effects of Na current reduction during ischemia and pharmacological Na channel blockade may facilitate these arrhythmias. Conversely, enhanced Na current may trigger afterdepolarizations and torsade de pointes ventricular tachycardia (VT) in the Long QT (LQT) syndrome (see below).

The cardiac Na channel is a transmembrane protein composed of the main pore-forming α-subunit and two subsidiary β-subunits (β1 and β2). Several studies report modulating effects of β-subunits on the function of heterologously expressed (mutant) α-subunits [4], but a physiological role of β-subunits is controversial [5]. The identification and analysis of genetic variants of SCN5A, the gene encoding the α-subunit, underlying various inherited idiopathic (in the absence of structural heart disease) arrhythmia syndromes, has proven particularly significant. The biophysical characterization of the resulting aberrant Na channels has improved our understanding of the ways in which various aspects of Na channel function (gating) and the associated structural domains may alter cardiac excitability. These studies are particularly powerful where correlations between the biophysical findings and the clinical observations were carefully explored. The clinical data thus serve as in vivo models of genetic control of Na channel function. These studies have enhanced our understanding of the functions of various channel domains, rendering them candidate targets to develop drugs not only designed for these disorders, but also for common disease. This review will deal with the inherited arrhythmia syndromes that arise from aberrant Na channels encoded by mutant SCN5A genes. We will discuss the SCN5A related LQT syndrome (LQT3), Brugada syndrome, and isolated conduction disease, and we will review the ways in which changes of distinct gating functions evoke the clinical phenotypes. Since the first biophysical studies on these groups of rare inherited syndromes (LQT3 in 1995 [6,7], Brugada syndrome in 1998 [8], isolated conduction disease in 2001 [9]), numerous reports have revealed different functional changes of individual SCN5A mutants. From this wealth of data, the recognition is emerging that some themes are consistently recurring and may be applicable to the pathophysiology and therapy of common disease. Furthermore, some new themes have recently emerged. First, single SCN5A mutations may cause ‘overlap syndromes’, i.e., phenotypes that combine features of LQT3, Brugada syndrome, and conduction disease [4]. Second, trafficking disorders may underlie reduced Na current of some SCN5A mutants.

1 Structure and function of the cardiac Na channel

A brief general outline of the structure and function of the cardiac Na channel α-subunit is now provided. In-depth discussions of its domains and their gating functions are available elsewhere [4,10–13]. The α-subunit contains four homologous domains, each composed of six membrane-spanning segments, linked by linker segments (Fig. 1). These domains are bracketed by an intracellular N terminus and C terminus. The Na channel may assume different conformations. Upon depolarization, it changes from a resting (closed) to an ion-conducting activated (open) state. This state is short-lived, as channel opening is coupled to and rapidly followed by inactivation [14]. The channels have to recover from this state and return to the resting state before they can reopen. Several inactivation processes, linked to conformational changes in distinct channel regions, are distinguishable and clinically relevant. Fast (time constant: few ms) and intermediate (50–100 ms) inactivation are distinguished by their speed of development and recovery. The magnitude of their time constants indicate that both processes are relevant for the cardiac action potential. Closed-state inactivation (inactivation from closed states without prior activation) may also be clinically relevant [15]. The ion conducting pore is lined by the linker segments (P loop) between transmembrane segments 5 and 6 in each domain. Some P loop residues confer ion selectivity [16]. Segment 4 in each domain contains positively charged residues, which impart sensitivity to the membrane voltage [17]. Outward S4 movement upon membrane depolarization opens the ion-conducting pore. Subsequent fast inactivation, coupled to this process [14], involves a hinged lid process. The intracellular linker between domains III and IV acts as a lid, which docks at the inner vestibule of the pore to occlude it [14,17,18]. Intermediate inactivation involves residues within the outer pore and the C-terminus [10,19–21]. The regions involved in closed-state inactivation await identification. This spatial organization of gating functions suggests that the effects of missense mutations or deletions are predictable from their position and alteration of charge or other biophysical properties. These predictions have been verified in some, but not all, mutations. Similarly, mutations with equal gating changes may be clustered within the same segments, or be located apart. For instance, enhanced intermediate inactivation is observed in mutants in the C-terminus [22–24], but also remote from it [25–27].

Fig. 1

Diagrammatic representation of the human cardiac sodium channel displaying locations of mutations associated with Long QT syndrome type 3, Brugada syndrome, isolated cardiac conduction disease, and overlap syndromes. DI, domain I; DII, domain II; DIII, domain III; DIV, domain IV.

2 Biophysical and clinical characterization of SCN5A mutations

Biophysical characterization of mutant Na channels, i.e., analysis of gating changes and expression levels, is usually performed using patch-clamp studies in heterologous expression systems: Xenopus laevis oocytes and transiently transfected mammalian cells (HEK293, tsA201). These studies generally yield consistent results among various experimental systems, but some inconsistencies exist, particularly between oocytes and mammalian cells (e.g., Refs. [28], and [6] versus [29]). Studies on transgenic mice have recently appeared [30,31]. Sarcolemmal Na channel expression and trafficking disorders are studied by immunocytochemical techniques [32,33].

Clinical characterization of SCN5A mutants includes: classification of the arrhythmia types using electrocardiogram (ECG) analysis, particularly at different heart rates; provocation of ECG abnormalities utilizing antiarrhythmic drugs; studying the triggers for arrhythmias (rest/exercise, slow/fast heart rates, sudden heart rate changes, particular circumstances including swimming, sudden noises [34,35]). Equally important is the clinical follow-up, e.g., event-free survival with different therapies: antiarrhythmic drugs, implantable cardioverters/defibrillators (ICDs), pacemakers. The rhythm storage capabilities of ICDs and pacemakers may be invaluable to unravel the electrophysiological properties of the individual SCN5A mutants. Additional clinical aspects are discussed below.

3 Long QT syndrome (LQT3)

The LQT syndrome is characterized by ECG markers of abnormal repolarization: QT interval prolongation, abnormal T waves, prominent U waves. It is associated with torsade de pointes that may be self-terminating or sustained and/or degenerate to ventricular fibrillation (VF), thereby causing cardiac arrest and sudden death. These ventricular tachyarrhythmias may be triggered by early afterdepolarizations arising from the abnormally prolonged action potentials [36]. The LQT syndromes have initially been subdivided into a congenital form (inherited in an autosomal dominant or autosomal recessive fashion), now known to be based on mutant potassium (K) or Na channels, and an acquired form provoked by environmental factors, notably antiarrhythmic and non-antiarrhythmic drugs that block various K channels. There is a strong suspicion that the acquired forms are partly based on mutant ion channels or polymorphisms thereof, that render the affected individual more susceptible to QT prolongation (reduced ‘repolarization reserve’) [37]. We therefore pool all known SCN5A mutations identified in patients with congenital or acquired (drug-induced) LQT syndrome. The congenital LQT syndromes were the first group of arrhythmias to be linked to a genetic basis and are considered the paradigm for inherited arrhythmia syndromes [37]. Seven types are known (LQT1–LQT7). SCN5A mutations cause LQT3.

Fig. 2 summarizes the biophysical changes identified by patch-clamp studies of SCN5A mutants in LQT3 [6,24,29,30,38–62]. Figs. 3 and 4 list these changes for SCN5A mutants in Brugada syndrome and conduction disease. Table 1 summarizes the effects of these changes on Na current magnitude. SCN5A mutants whose biophysical properties have not been reported are listed in Table 2 [8,63–68].

Fig. 4

(A) Representative electrocardiogram of isolated conduction disease (40 ms/div). Note marked QRS widening and PQ interval prolongation. (B) SCN5A mutations associated with isolated conduction disease: summary of changes in their biophysical properties. Same abbreviations as in Fig. 2.

Fig. 3

(A) Representative electrocardiogram of Brugada syndrome (40 ms/div). Note ST segment elevation (coved type) with negative T waves, typically seen in right precordial leads V1–V3 (here V1). There is also marked PQ interval prolongation. (B) SCN5A mutations associated with Brugada syndrome: summary of changes in their biophysical properties. Same abbreviations as in Fig. 2B.

Fig. 2

(A) Representative electrocardiogram of Long QT syndrome type 3 (40 ms/div). Note marked QT interval prolongation with late peaked T waves. There is also sinus bradycardia. (B) SCN5A mutations associated with Long QT syndrome type 3: summary of changes in their biophysical properties. τfast current decay, time constant of fast component of sodium current decay (fast inactivation); inact, inactivation; V1/2 of inactivation, voltage at which 50% of sodium channels are inactivated; V1/2 of activation, voltage at which 50% of sodium channels are activated; − shift, shift to negative voltage; + shift, shift to positive voltage; ↓, reduction; ↑, increase; =, unchanged; –, not reported.

View this table:
Table 2

SCN5A mutations without studies of biophysical properties

MutationLocalizationRefs.
LQT3:  
D1114NDIIDIII[63]
T1304MDIIIS4[64]
K1500delDIIIDIV[65]
L1501VDIIIDIV[63]
F1617delDIVS3S4[63]
R1623LDIVS4[63]
T1645MDIVS4[64]
S1787NC terminus[63]
D1840GC terminus[66]
Brugada syndrome:  
R27HN terminus[67]
K126EN terminus[68]
A226VDIS4[67]
I230VDIS4[67]
IVS7DS+4DIS5–S6[67]
R282HDIS5–S6[67]
V294MDIS5–S6[67]
G319SDIS5–S6[65]
F393delDIS6[67]
H681PDIDII[67]
A735EDIIS1[65]
IVS14–1G>CDIIS2[65]
F851LDIIS5[67]
S871fs+9XDIIS5S6[67]
F892IDIIS5S6[67]
C896SDIIS5S6[67]
S910LDIIS5S6[67]
R965CDIIDIII[65]
E1053KDIIDIII[65]
D1114NDIIDIII[65]
Q1118XDIIDIII[67]
K1236NDIIIS1S2[67]
E1240QDIIIS2[67]
F1293SDIIIS3S4[67]
V1398XDIIIS5–S6[8]
G1466fs+12XDIIIS6[67]
K1500delDIIIDIV[67]
G1740RDIVS5–S6[67]
E1784KC terminus[67]
V1951LC terminus[67]
Conduction disease:  
IVS22DS+2DIIIS4S5[120]
delG5280 [120]
View this table:
Table 1

Effects of changes in biophysical properties on sodium current magnitude

Less NaMore Na
currentcurrent
τfast of current
decay  
Intermediate
inactivation  
Persistent current
Current density
V1/2 of inactivation− shift+ shift
V1/2 of activation+ shift− shift
  • Na, sodium; τfast of current decay, time constant of fast component of sodium current decay (fast inactivation); V1/2 of inactivation, voltage at which 50% of sodium channels are inactivated; V1/2 of activation, voltage at which 50% of sodium channels are activated; ↓, reduction; ↑, increase; + shift, shift to positive voltage; − shift, shift to negative voltage.

QT prolongation may result from increased depolarizing or reduced repolarizing current. Accordingly, LQT3 associated SCN5A mutants produce increased Na current. Their almost universal pathophysiological mechanism is enhanced persistent Na current [6,24,29,30,39–41,43–45,47–59,61,62]. While normal Na channels have virtually complete fast inactivation shortly following opening, these mutant channels exhibit a resistance to inactivation. An enhanced propensity for reopening from the inactivated state causes a persistent current during the action potential plateau [43,54]. The first biophysically characterized LQT3-related SCN5A mutant, ΔKPQ [6], provides a plausible explanation for this gating change: the three residue deletion in the linker between domains III and IV, the putative fast inactivation lid, may destabilize its binding to its receptor at the inner vestibule of the channel pore. Similarly, the next reported LQT3 associated mutants are located in the S4–S5 linkers and in close proximity to the docking site for the fast inactivation lid [69]. Regions of the C terminus may also be involved in fast inactivation, as engineered mutations here disrupt fast inactivation [70] and enhance persistent Na current [71], and some LQT3 associated mutants are clustered here (Fig. 1). While the persistent current is relatively small (not exceeding 5% of peak Na current), it may cause significant action potential prolongation, afterdepolarizations, and torsade de pointes, because it occurs during a phase of the cardiac cycle when the membrane potential is governed by a delicate balance of depolarizing and repolarizing forces, and the membrane resistance is high. Thus, small disruptions of this balance may profoundly affect membrane potential [72].

Clinical studies indicate that QT prolongation is most severe and arrhythmias most prevalent at slow heart rates, and that there is abnormally strong QT shortening upon heart rate increases [73]. Accordingly, the persistent Na current is prominent at slow heart rates [6,48,72] and reduced at fast rates, because, in some SCN5A mutants, its recovery from inactivation is slow and therefore incomplete at fast rates [22,24,48]. Conversely, many LQT3 patients (30–40%) experience cardiac arrest during exercise or emotional stress [35,74]. Enhanced adrenergic tone during these circumstances may facilitate the early afterdepolarizations that trigger their tachyarrhythmias. Alternatively, these patients may have QT prolongation at fast heart rates, because their SCN5A mutants do not cause QT prolongation by enhancing persistent Na current with a slow recovery from inactivation, but by other mechanisms. Challenging this view, however, is a study in transgenic mice (knock-in of ΔKPQ, a mutant with pronounced persistent Na current), where sudden pacing-induced heart rate increases caused paradoxical QT prolongation [30]. While this observation is of obvious clinical relevance, its underlying mechanism is unclear. Functional AV block was also reported. Here, repolarization delay is so excessive that the action potential duration exceeds the beat-to-beat interval of the underlying rhythm. Thus, cardiac excitability is not restored when the wavefront of the following beat arrives. The site of AV block is the ventricular working myocardium or the specialized conduction system (Purkinje fibers, bundle branches), but not the AV node [57]. Some mutants do not exhibit enhanced Na current in patch-clamp studies, but seem to exert electrophysiological effects only in conjunction with environmental factors (acquired LQT syndrome) [38].

4 Brugada syndrome

The Brugada syndrome, inherited in an autosomal dominant fashion, has a high incidence of sudden death (usually at night or rest), due to cardiac tachyarrhythmias, notably VF and polymorphic or monomorphic VT [67,75]. It has characteristic ECG features that are dynamic and may be temporarily absent [76]: ST segment elevations in the right precordial leads (V1–V3), often with signs of conduction slowing: prolongation of PQ and His-ventricle intervals and QRS width, leftward deviation of the frontal QRS axis, (pseudo) right bundle branch block [77–79]. So far, the only gene with a proven involvement is SCN5A. However, SCN5A mutations are found only in 30% of these patients [80]. This suggests involvement of other genes, but they still await discovery [81]. The clinical and experimental data provide strong evidence that impediment of activation is pivotal in the pathophysiology of Brugada syndrome. Given their role in cardiac excitability [9,82], the disease-causing SCN5A mutants should exhibit reduced Na current. Accordingly, reduced Na current was found by patch-clamp studies in virtually all SCN5A mutants, but the underlying mechanisms may be classified into different categories (Fig. 3) [8,22–24,26,27,32,33,59,61,83–93]. First, gating changes may reduce Na current [22–24,27,32,61,83–85,88,91,93]. Second, mutations within the ion-conducting pore may impede Na permeation [59,68,83,86]. Third, introduction of a premature stop codon produces a truncated, nonconducting, Na channel [8]. Fourth, mutant proteins may be retained in the endoplasmic reticulum (ER) through the actions of the ER quality control systems (see below), thus reducing sarcolemmal expression of SCN5A mutants and Na current density [32,33]. The gating changes may be mapped to particular Na channel regions. Notably, enhanced intermediate inactivation is commonly observed in mutations within the C terminus, as domains herein are effectors of intermediate inactivation. For instance, the calmodulin binding ‘IQ’ domain modulates calcium (Ca)-dependent intermediate inactivation [93,94]. For some mutants (R1512W, A1924T, T1620M), biophysical studies have not yet demonstrated reduced Na current. This may be (partly) due to experimental conditions, e.g., T1620M exhibits reduced Na current only at 32 °C, but not at 22 °C [88].

Clinical and experimental studies also suggest reduced Na current. Clinically, in addition to the frequent association with conduction slowing [78], the strongest evidence is exacerbation/unmasking of the phenotype by Na channel blockers [34,95–101]. This is reproducible in experimental models of Brugada syndrome [102,103]. In line with expectations, patients who have a proven SCN5A mutation carry stronger signs of conduction slowing, e.g., longer PQ and His-ventricle intervals, and more QRS widening upon challenge by Na channel blockers [80], than those who have not. While reduced Na current is strongly supported by clinical and experimental data, the mechanism whereby it causes the phenotype remains controversial. The hypothesis that currently receives the most support places the spatially heterogeneous magnitude of the transient outward K current, Ito, at its core [102]. This current is large in epicardial cells, particularly in the right ventricle [104], but absent from endocardial cells [105,106], while its β-subunit KChiP2 is also differentially expressed between these cell layers [107]. According to this hypothesis, this strong repolarizing current renders epicardial cells more sensitive than endocardial cells to reductions of depolarizing Na current. In epicardial cells, Na current reduction causes the loss of the action potential plateau, because the L-type Ca channel (ICa-L), responsible for it, is not activated at the less positive membrane voltages that result from reduced Na current. Concomitantly, the plateau is retained in endocardial cells, because reduced net depolarizing force is not exacerbated by the repolarizing force of Ito. The resulting potential gradient between these cell layers during the action potential plateau phase produces the characteristic ST elevations and facilitates arrhythmias based on (phase 2) reentry. Indirect evidence supports this hypothesis. Clinically, the ST elevations are suppressed by isoproterenol, presumably through its ICa-L enhancing action [100,108]. Also, right ventricular epicardial activation recovery intervals, measures of action potential duration, are abbreviated concomitantly with the emergence of ST elevations upon infusion of the Na channel blocker pilsicainide [109]. Experimental models of Brugada syndrome employ (ATP sensitive) K channel openers to induce ST elevations and VT/VF, while the Ito blocker 4-aminopyridine counteracts these effects [102]. Similarly, quinidine may confer some protection from ST elevations and VT/VF, presumably because this drug blocks Ito (in addition to the Na current) [110–112]. Nevertheless, other findings challenge this hypothesis. It is not clear how the reported strong expression of Ito in midmyocardium [113] ties up with this explanation. A loss of the plateau phase in this large cell layer would constitute such a large electrical sink to the remaining small layer where the plateau would be retained (endocardium), that the action potential plateau should also be abolished in endocardial cells and excitation–contraction coupling effectively eliminated. Furthermore, significant Na current reduction (assessed by QRS width) by Na channel blockers in patients without SCN5A mutations does not evoke the ST elevations of Brugada syndrome.

Despite this controversy, studies of Brugada syndrome associated mutant Na channels have clearly enhanced our recognition that Na current reduction may be a general mechanism of arrhythmias in common disorders [20]. Of particular interest is intermediate inactivation, which determines Na channel availability [114]. Na current reduction secondary to enhanced intermediate inactivation may underlie reentrant tachyarrhythmias in ischemia [115]. These arrhythmias may evolve from areas of slow conduction near the ischemic border zone [116]. Interestingly, Na channel blockers elicit proarrhythmic effects (rate-dependent conduction slowing and facilitation of reentry) in epicardial fibers isolated from infarcted hearts [117,118]. Furthermore, cells from the epicardial border zone of a canine subacute infarction model exhibit a delayed recovery from inactivation upon sustained depolarization, which suggests enhanced intermediate inactivation [119]. Finally, these cells exhibit an increased sensitivity to drug block by Na channel blockers [119]. This is indicative of enhanced intermediate inactivation, since facilitated block by these drugs is linked to structural rearrangements involved in intermediate inactivation [21].

5 Isolated conduction disease

The latest group of idiopathic inherited arrhythmia syndromes linked to SCN5A mutations is isolated (in the absence of Brugada or LQT3 syndrome) conduction disease (Fig. 4 [9,25,27,86]). We found that a mutation in the linker between domains I and II (G514C) exhibited opposing gating changes [9]: resistance to voltage-dependent activation would reduce Na current, while destabilization of closed-state inactivation would increase it. The net effect is attenuated Na current reduction. The seemingly paradoxical increase of Na channel availability resulting from destabilization of closed-state inactivation is required to recapitulate the phenotype (conduction slowing), as modeling studies indicated that the magnitude of the resistance to voltage-dependent activation, if left unopposed, would cause Brugada syndrome. One conclusion may be that conduction disease and Brugada syndrome are manifestations of different degrees of severity in the spectrum of Na current reduction, with Brugada syndrome constituting the most severe form. The observation that Brugada syndrome often encompasses some degree of conduction slowing supports this view. Conversely, some evidence undermines this concept. In particular, the first reported SCN5A mutants linked to isolated conduction disease were predicted to lack large segments based on the introduction of premature stop codons near the N terminus [120]. These channels are expected to yield no Na current in heterologous expression systems. Similarly, a missense mutation in a pore-lining segment resulted in the absence of macroscopic Na current [86]. From these observations it seems hard to predict the putatively milder phenotype isolated conduction disease. Without compensatory increases in the expression of the wild-type allele (experimental validation thereof is lacking), these changes result in haploinsufficiency, which is not present in most Brugada syndrome related SCN5A mutants. It is not obvious how any gating change, however harsh, would cause more severe Na current reduction than the complete absence of an ion-conducting Na channel. A dominant negative effect, as in K channels, might provide an explanation, but its mechanism must be different from K channels. In K channels, one mutant domain may disrupt the function of the assembly of the four separate domains that constitute the K channel tetramer [121]. The Na channel, however, is not an assembly, as its four domains form a functional ion channel by themselves.

6 Overlap syndromes

Given the pivotal role of Na channels in cardiac excitability, it is not surprising that Na channel dysfunction may cause complex effects on cardiac rate and rhythm, as shown in SCN5A knock-out mice [31]. Mice heterozygous for the SCN5A deletion exhibited atrial and atrioventricular conduction slowing, intraventricular conduction defects (rightward QRS axis shift), prolonged ventricular refractoriness, and reentrant VT. Furthermore, recent studies report single SCN5A mutants that cause a phenotype which combines features of LQT syndrome, Brugada syndrome, and conduction disease.

The first such mutant was 1795insD. Patch-clamp [22] and modeling studies [122] have provided plausible explanations for its mixed phenotype. Not only has this explanation invoked differential effects on distinct gating processes, but the electrophysiological alterations predicted from these changes were also confirmed by clinical observations. Thus, QT prolongation was explained by enhanced persistent Na current, while Na current reduction was predicted from enhanced intermediate inactivation. Based on its slow recovery, intermediate inactivation is predicted to accumulate at short diastolic intervals, reducing Na current at fast heart rates. Accordingly, clinically, the ST elevations were most prominent at fast heart rates [22]. Subsequently, other mixed phenotypes were recognized upon closer examination of previously reported mutants and in newly discovered ones. For instance, the LQT3 associated mutants ΔKPQ [123] and D1790G [124] displayed conduction slowing and sinus arrest, while G1406R segregated with conduction disease or Brugada syndrome in different family branches of one kindred [86]. Interestingly, Brugada syndrome was found exclusively in males, while six of eight patients afflicted by conduction disease were females. This gender inequality mirrors the observation that in some regions, particularly Southeast Asia, Brugada syndrome is far more prevalent in males than in females [78]. These studies provide strong impetus to the suspicion that gender or yet unidentified modifier genes contribute decisively to the phenotype [125].

The emerging notion that one single mutation may cause gating changes with opposing effects highlights the necessity of precise clinical phenotyping. This may direct patient management. For instance, while the most eye-catching features of 1795insD are signs of Brugada and LQT3 syndrome, its clinical relevance may stem from its heart rate slowing effect [126], that results both from sinus exit block and slowing of the pacemaker firing rate. While intuition comfortably links exit block to reduced Na current (conduction block to the atrial tissues that surround the sinus node), a slower firing rate may also result from impaired Na channels, given their role in pacemaker activity in the peripheral sinus node [127,128]. Long follow-up periods of the large kindred in whom 1795insD was described revealed that excess mortality, while high in untreated patients, was abolished after pacemaker implantation, indicating that death is related to bradycardia, rather than tachycardia [126]. Holter ECG analysis revealed severe bradycardia episodes (sinus arrest and sinus exit block), but no bradycardia induced tachycardias such as torsade de pointes. Also, ST elevations were attenuated at slow heart rates [22]. Taken together, these observations render it likely that death in 1795insD is due to impediment of excitation and bradycardia, but not (bradycardia induced) VT/VF, unlike other Brugada or LQT3 syndrome associated SCN5A mutants [67,129]. Treatment with a pacemaker rather than ICD, the generally accepted best course for Brugada syndrome patients [130], thus appears strongly validated for this mutant.

Another instance where careful clinical phenotyping has proven its worth is the unexpected finding that flecainide treatment, intended as a therapy in LQT3 patients [124], caused right precordial ST elevations, i.e., the Brugada syndrome phenotype, in these patients [131]. This is at first glance surprising, since flecainide [52] and other Na channel blockers (lidocaine [43,44,132], mexiletine [45]), preferentially attenuate the persistent Na current that causes QT prolongation. Clinically, this causes QT interval shortening, and these drugs have therefore been proposed as specific treatments for LQT3 [71,124]. Far from being merely an untoward effect and an additional challenge to safe and effective drug treatment of LQT3, this unexpected finding has offered the opportunity to deepen our understanding of Na channel gating and the molecular pharmacology of Na channel blockers [20]. ST elevation following flecainide in some LQT3 patients may result from the fact that their mutant Na channels exhibit enhanced closed-state inactivation. This gating change sensitizes 1795insD and ΔKPQ Na channels to tonic block by flecainide, i.e., Na current reduction at slow heart rates [23] (similarly, it sensitizes the LQT3 associated mutant R1623Q to tonic block by lidocaine [53]). Concomitantly, Na channels with the 1795insD mutation exhibit enhanced intermediate inactivation, which sensitizes them to use-dependent flecainide block, i.e., block at fast heart rates [23].

7 Trafficking disorders

It is becoming clear that failure of proper trafficking of Na channel proteins to the sarcolemma and their retention in the ER may contribute to the pathophysiology of Brugada syndrome [32,33] and conduction disease [120]. The ER is the first compartment for the synthesis and processing of membrane proteins. When errors in polypeptide folding occur, quality control mechanisms within the ER ensure that the misfolded polypeptides are recognized and degraded. Often the mutation is relatively minor and those molecules that escape the ER quality control systems and reach their proper cellular localization retain some function [56,133,134]. Consequently, the identification of strategies to correct diseases of protein folding is at the forefront of clinical research [135,136]. Trafficking-deficient mutants have been rescued by lowering the incubation temperature or using ‘chemical chaperones’, e.g., glycerol and dimethylsulfoxide [137,138]. These strategies may stabilize conformations that escape degradation by the ER. The former is only applicable in vitro while the latter is not clinically applicable since concentrations that would be needed to attain the desired effect are too high to be achieved in vivo and are unsafe. More recently, it was demonstrated that substrates and blockers that bind to the target protein can achieve similar stabilizing effects to rescue ER-trapped mutant proteins [139–142]. The selectivity that can be achieved using such ‘pharmacological chaperones’ (ligand-mediated) makes it more likely that such a strategy could ultimately become clinically relevant.

The feasibility of this approach was demonstrated for mutant HERG K channels that were rescued by the HERG blockers E-4031, astemizole, and cisapride. Interestingly, rescue may be accomplished without inducing channel block [143]. This may also apply to trafficking-deficient Na channels that were rescued by the Na channel blocker mexiletine [56]. While it is clear that rescuing SCN5A mutants and enhancing Na current is an highly attractive novel concept with potentially far-reaching therapeutic implications, several theoretical and practical issues need to be resolved prior to its clinical application [144].

8 Conclusion and future directions

The identification of mutant Na channels in inherited arrhythmia syndromes and the studies of their functional properties have significantly enhanced our understanding of Na channel function and its modulation by disease and antiarrhythmic drugs. These studies have been particularly powerful where biophysical studies are combined with precise clinical phenotyping. Eight years into the study of these mutants, intensive research efforts have identified a host of biophysical mechanisms of Na channel dysfunction. Some themes are consistently recurring. It is hoped that fathoming these themes in these rare inherited disorders will enhance our understanding of arrhythmogenesis in common disorders and that this will ultimately result in better therapy strategies, including drug design, for common disease. At the same time, it is clear that this requires, as the next great challenge, the integration of the studies on isolated Na channel α-subunits into the ensemble of collaborating and modulating molecules that govern the cardiac rhythm (e.g., other ion channels and exchangers, β-subunits, trafficking machinery and targeting molecules, anchoring proteins), in order to grasp the complex multifactorial pathophysiology of common disease [5].

Acknowledgements

Dr. Tan is supported by a fellowship of the Royal Netherlands Academy of Arts and Sciences (KNAW) and by The Netherlands Heart Foundation (NHS 2002B191); Dr. Bezzina by The Netherlands Heart Foundation (NHS 2000.059); Dr. Smits by The Netherlands Organization for Health Research (ZonMW 902-16.193) and the Interuniversity Cardiology Institute Netherlands (ICIN project 27); Drs. Verkerk and Wilde by The Netherlands Organization for Scientific Research (NWO 805-06.155 [AOV] and NWO 902-16.193 [AAMW]). We thank for Dr. M.W. Veldkamp for valuable discussions.

References

  1. [1]
  2. [2]
  3. [3]
  4. [4]
  5. [5]
  6. [6]
  7. [7]
  8. [8]
  9. [9]
  10. [10]
  11. [11]
  12. [12]
  13. [13]
  14. [14]
  15. [15]
  16. [16]
  17. [17]
  18. [18]
  19. [19]
  20. [20]
  21. [21]
  22. [22]
  23. [23]
  24. [24]
  25. [25]
  26. [26]
  27. [27]
  28. [28]
  29. [29]
  30. [30]
  31. [31]
  32. [32]
  33. [33]
  34. [34]
  35. [35]
  36. [36]
  37. [37]
  38. [38]
  39. [39]
  40. [40]
  41. [41]
  42. [42]
  43. [43]
  44. [44]
  45. [45]
  46. [46]
  47. [47]
  48. [48]
  49. [49]
  50. [50]
  51. [51]
  52. [52]
  53. [53]
  54. [54]
  55. [55]
  56. [56]
  57. [57]
  58. [58]
  59. [59]
  60. [60]
  61. [61]
  62. [62]
  63. [63]
  64. [64]
  65. [65]
  66. [66]
  67. [67]
  68. [68]
  69. [69]
  70. [70]
  71. [71]
  72. [72]
  73. [73]
  74. [74]
  75. [75]
  76. [76]
  77. [77]
  78. [78]
  79. [79]
  80. [80]
  81. [81]
  82. [82]
  83. [83]
  84. [84]
  85. [85]
  86. [86]
  87. [87]
  88. [88]
  89. [89]
  90. [90]
  91. [91]
  92. [92]
  93. [93]
  94. [94]
  95. [95]
  96. [96]
  97. [97]
  98. [98]
  99. [99]
  100. [100]
  101. [101]
  102. [102]
  103. [103]
  104. [104]
  105. [105]
  106. [106]
  107. [107]
  108. [108]
  109. [109]
  110. [110]
  111. [111]
  112. [112]
  113. [113]
  114. [114]
  115. [115]
  116. [116]
  117. [117]
  118. [118]
  119. [119]
  120. [120]
  121. [121]
  122. [122]
  123. [123]
  124. [124]
  125. [125]
  126. [126]
  127. [127]
  128. [128]
  129. [129]
  130. [130]
  131. [131]
  132. [132]
  133. [133]
  134. [134]
  135. [135]
  136. [136]
  137. [137]
  138. [138]
  139. [139]
  140. [140]
  141. [141]
  142. [142]
  143. [143]
  144. [144]
View Abstract