PHYSIOLOGY IN MEDICINE In collaboration with The American Physiological Society, Thomas E. Andreoli, MD, Editor
Molecular Biology of Sodium Channels and Their Role in Cardiac Arrhythmias Augustus O. Grant, MB, ChB, PhD The sodium channel is an integral membrane protein that plays a central role in conduction of the cardiac impulse in working cardiac myocytes and cells of the His-Purkinje system. The channel has two fundamental properties, ion conduction and gating. Specific domains of the channel protein control each of these functions. Ion conduction describes the mechanisms of the selective movement of sodium ion across the pore in the cell membrane. The selectivity of the channel for sodium ions is at least 10 times greater than that for other monovalent cations; the channel does not normally conduct divalent cations. Gating describes the opening and closing of the sodium channel pore. Sodium channels open transiently during membrane depolarization and close by a process termed inactivation. The cardiac sodium channel protein is a multimeric complex
consisting of an ␣ and an auxiliary -subunit. The genes encoding the sodium channel have been cloned and sequenced. The ␣ subunit gene, SCN5A is sufficient to express a functional channel. However,  subunit co-expression increases the level of channel expression and alters the voltage dependence of inactivation. Mutations of the sodium channel may result in incomplete inactivation during maintained depolarization, a decrease in the level of channel expression or acceleration of inactivation. The resulting clinical phenotypes include long QT syndrome, type III (LQT III), Brugada syndrome, and heart block. LQT III and Brugada syndromes have a high case fatality rate and are best treated with an implantable defibrillator. Am J Med. 2001; 110:296 –305. 䉷2001 by Excerpta Medica, Inc.
C
closely related ions with a selectivity ratio of 10 to 100:1. They proposed that gating involved transitions between three groups of states, resting, open, and inactivated (Figure 1). The state of the channel is governed by the position of activation or m gates and an inactivation or h gate. At the normal resting potential of the cell, sodium channels are in the closed resting state. In this state, the activation gates are in a closed position and the inactivation gate is an open position. Membrane depolarization induces a transition from a closed resting state to the open state. Both the activation and inactivation gates occupy an open conformation in this channel state. However, even in the face of maintained depolarization, the inactivation gate assumes a closed conformation, blocking movement of Na⫹ across the cell membrane. During inactivation, the sodium channel is converted to a refractory state. The membrane must repolarize before a second conductance increase can occur. The channel then assumes its resting state with closed activation and open inactivation gates. Subsequent studies have confirmed these essential features of sodium channel gating in most excitable cells, including the heart. Over the past decade, the techniques of molecular genetics and biophysics have provided remarkable advances in our understanding of sodium channel structure and the relationship of function to structure. Is there a compelling reason to acquaint the clinician with these new developments? The answer is a resounding “yes.”
onduction in working myocardial cells of the atrium and ventricle, and in cells of the His-Purkinje, is sustained by transient increases in membrane permeability to sodium ions. This function resides in a family of integral membrane proteins, the voltagegated sodium channels. The functional properties of the neuronal channel were worked out almost half a century ago in a classic series of experiments by Hodgkin and Huxley (1). They described two sets of properties of the sodium channel: conductance or the mechanisms by which ions cross the membrane and gating, the mechanism(s) that open and close the channel. They established that conductance was time invariant; the action potential was not generated by changing cation selectivity of a single class of channels, but by the gating of channels with fixed conductance and selective permeability to a single ion, for example, Na, K, Ca, or Cl⫺. The selective permeability of ion channels is especially efficient, because they may exhibit a throughput rate of greater than 106 ions/second, and yet distinguish between
From the Cardiology Division, Duke Medical Center, Durham, North Carolina. Requests for reprints should be addressed to Augustus O. Grant, Duke University Medical Center, Box 3504, Durham, North Carolina 27710; e-mail:
[email protected] 296
䉷2001 by Excerpta Medica, Inc. All rights reserved.
0002-9343/01/$–see front matter PII S0002-9343(00)00714-2
Sodium Channels and Cardiac Arrhythmias/Grant
Figure 1. The Hodgkin-Huxley model of sodium channel gating. The upper trace shows a voltage step and the middle trace, the resulting membrane current. A physical model to account for the transient current is shown in the lower trace. The sodium channel is represented as a pore spanning the membrane. In the resting state, the activation (m) gate is in the closed position and the inactivation (h) gate is in the open position. After depolarization, the m gate assumes the open position, and with both gates open, sodium ions move into the cell. The h gate then moves into the closed position blocking ion movement. When the membrane is returned to the resting level, the m gate moves into the closed position, a process termed deactivation. After a variable interval, the h gate moves into the open position, a process termed recovery from inactivation (not shown).
The new developments have provided a fundamental understanding of the inherited cardiac arrhythmias. They may also provide clues to the basis of more common arrhythmias and apparently idiosyncratic drug reactions. In this contribution to the series on the molecular biology of ion channels, I shall review the structure–function relationships of the sodium channels, as revealed by studies over the past decade. This will be followed by a critical review of mutations of the sodium channel as a basis for certain inherited arrhythmias.
BIOCHEMICAL PROPERTIES OF THE SODIUM CHANNEL The sodium channels of brain and the eel electroplax, a form of skeletal muscle, were partially purified by the classical techniques of protein biochemistry (2). The brain sodium channel consists of an ␣ subunit of approx-
imately 250 KD and smaller subunits 1 and 2 of approximately 30 to 40 KD. The stoichiometry of the subunits is 1:1:1. There are fewer data on the subunit composition of the cardiac sodium channel. The ␣ and 1 subunits of the cardiac sodium channel have a size similar to that of their neuronal counterpart. The presence of a similar 2 subunit in the heart is uncertain. Noda et al used the partial sequence of the eel electroplax to clone the eel Na channel ␣ subunit (3). They then used probes from the eel Na channel primary sequence to clone three neuronal sodium channel isoforms, rat brain I to III (4). Rogart et al (5) cloned the full-length sequence of the rat heart Na channel from a newborn heart cDNA library using probes derived from the rat brain II sequence. They showed that the same RNA species (termed rSKM2) was expressed in denerved skeletal muscle, but not in innervated skeletal muscle or brain. Gellens et al (6) used probes derived from rSKM2 to clone a cardiac Na channel from an adult human heart cDNA library. The human channel, hH1 had 90% homology to the rat counterpart. When hH1 was expressed in a heterologous system, it reproduced the characteristics of cardiac Na channels in native cells, including TTX resistance (IC50 of 50 M), resistance to the marine toxin contotoxin (skeletal muscle ␣-subunit–specific toxin) and sensitivity to transitional metal cations, such as cadmium.
TOPOLOGY OF THE CARDIAC NA CHANNEL Because of its size and hydrophobic nature, the secondary and tertiary structure of the Na channel has not been resolved. Elements of the tertiary structure have been inferred from hydrophobicity plots, location of probable sites of glycosylation and phosphorylation, location of epitopes to antibodies, and more recently, analogy with another membrane channel of known structure (2,7,8). A tentative structure for the Na channel ␣ subunit is illustrated in Figure 2. The amino- and carboxy termini are intracellular. The remainder of the polypeptide is organized into four homologous domains, DI to DIV. Each domain consists of six transmembrane segments, S1 to S6. These segments are generally hydrophobic and have an ␣-helical conformation of sufficient length to cross the membrane. The fourth transmembrane segment, S4, is highly charged with conserved arginine or lysine residues at every third position. This motif is conserved among voltage-gated Na, Ca, and K channels and is tentatively identified as the voltage sensor that responds to the changes in membrane potential. The domains are connected by intracytoplasmic interdomain loops IDI/II to IDIII/IV. The first loop, IDI/II has multiple consensus sequences for protein kinase A– dependent phosphorylation. IDI/II and IDII/III are March 2001
THE AMERICAN JOURNAL OF MEDICINE威
Volume 110 297
Sodium Channels and Cardiac Arrhythmias/Grant
Figure 2. Structure of the cardiac sodium channel ␣ and  subunits. The ␣ subunit consists of four homologous domains, DI to DIV. Each domain consists of six transmembrane segments, S1 to S6. The amino and carboxy termini are intracellular. Some of the mutations associated with LQT3 are identified. The three amino acid deletion mutation ␦KPQ in the linker between DIII and DIV produces the most severe defect. The 1 subunit consists of a single membrane-spanning segment and the intracellular carboxy terminus.
among the most variable regions of the ␣ subunit. IDIII/IV is short and highly conserved among Na channel isoforms. The loops between the individual transmembrane segments alternate between intracytoplasmic and extracellular sites. The loop between S5 and S6 is long and is believed to curve back into the membrane to form the pore or P loop.
CHANNEL CONDUCTANCE
LOCALIZATION OF FUNCTIONAL DOMAINS OF THE NA CHANNEL The availability of small rigid neurotoxins of known structure, antibodies to specific epitopes, enzyme susceptibility, and site-directed mutagenesis have provided a wealth of data relating channel structure to function (2). Site-directed mutagenesis performed in the systematic approaches, such as alanine scanning, and cysteine accessibility has been particularly revealing. Alanine scanning involves the sequential mutation of each amino acid in a region of interest to alanine. Alanine substitution is usually a conservative change. The cysteine accessibility approach involves the mutation of single amino acids to cysteine followed by the application of thiol-reactive reagents, applied from either side of the cell membrane as transmembrane potential is varied. The approach provides insight into conformational changes of the protein during gating. The comparative properties of the Na 298
March 2001
THE AMERICAN JOURNAL OF MEDICINE威
channel isoform have also provided important structure– function clues about the channel. A complete structure– function analysis of the channel should account for its basic functions, such as TTX resistance and Cd2⫹ sensitivity, and the elementary properties of Na⫹-selective conduction, and transient opening and closing.
Volume 110
The neurotoxins tetrodotoxin and saxitoxin are of clinical and experimental interest. They are highly potent neurotoxins that are produced by micro-organisms, and concentrated in fish and shellfish at higher levels of the food chain. Tetrodotoxin is responsible for sporadic cases of fish poisoning as a result of improperly prepared puffer fish (or fugu, an Asian delicacy). Saxitoxin is responsible for outbreaks of paralytic shellfish poisoning (associated with “red tide”). These toxins are small rigid molecules of known crystal structure and have proved crucial in defining the ion conduction pathways of sodium channels. The toxins have no effect on channel gating but block ion permeation from an extracellular site. The cardiac sodium channel is relatively resistant, with an IC50 in M range, whereas the neuronal channel has an IC50 in the nM range. In comparing the putative pore loop residues between neuronal (TTX sensitive) and cardiac (TTX resistant) sodium channel isoforms, only two residues dif-
Sodium Channels and Cardiac Arrhythmias/Grant
fer in the four P loops. A cysteine replaces phenylalanine, and arginine replaces asparagine in the D1 S5–S6 loop of the cardiac isoform. Cysteine is a logical residue for Cd2⫹ binding. Satin et al replaced cysteine in the cardiac isoform with phenylalemine or tyrosine and recapitulated the high tetrodotoxin sensitivity characteristic of the neuronal sodium channel (2,9). Backx et al (10) replaced tyrosine in the tetrodotoxin-sensitive skeletal muscle isoform with cysteine and markedly reduced toxin sensitivity. Similar mutagenesis studies of the other amino acid characteristics of the cardiac isoform show that it has minimal effects on toxin sensitivity. The cysteine mutations and Cd2⫹ susceptibility identified a channel region that was crucial to toxin binding, and by inference to ion permeation. Strichartz modeled the secondary structure of this region of domain I of the channel as a  hairpin (11). They docked the tetrodotoxin molecular onto this structure. The domain II  hairpin was then added. The remaining energetically important binding could be accounted for by hydrogen bonds between two hydroxyl groups of the toxin with carboxyl groups of domain II. The P loops of domains III and IV were positioned symmetrically to allow nonbinding interactions. Alternative secondary structures of the P loops have been proposed based on cysteine-scanning mutagenesis. However, details of toxin binding are not provided with these models (12,13). When the toxin was removed in the Lipkind-Fozzard model, the binding site consisted of a funnel-like structure 12 angstroms deep and 12 angstroms wide at its widest end, with a narrow mouth of 3 ⫻ 5 angstroms. The narrow mouth was surrounded by aspartate, glutamate, lysine, and alanine residues from domains I to IV and forms the selectivity filter region. Lipkind and Fozzard explored the interaction of Na⫹ with the putative pore structure. The binding energy of Na⫹ with the aspartate and glutamate carboxyl residues of domains I and II was approximately 100 K-cal/mol, sufficient energy to dehydrate the Na⫹ as it crossed the pore. Four carboxyl side chains of glutamic acid occupy the equivalent position in the calcium channel (14). Mutation of the lysine and alanine residues of the selectivity filter region to glutamine converted the sodium channel to a calcium selective channel (14). This is an important experiment, because it indicated that the Na⫹:Ca2⫹ selectivity of the sodium channel is imprinted in the channel structure and is not likely to be induced by such hormones as atrial naturetic peptide or -adrenergic agonists (15,16).
CHANNEL ACTIVATION The opening of the sodium channel involves two processes. Changes in membrane potential must first be sensed by the channel protein. This signal must eventually be transmitted to the ion permeation pathway to ini-
tiate transmembrane permeation. In their original model of sodium channel gating, Hodgkin and Huxley (1) proposed that the outward movement of regions of the channel protein was responsible for the process of activation or opening of the channel. This postulate was confirmed many years later by the recording of small outward gating currents that preceded channel opening (17). As soon as the primary sequence of the sodium channel ␣ subunit was solved, the S4 segment of each domain was proposed as candidate structures for voltage sensing. The S4 segments are conserved among the four domains and among voltage-gated ion channels and consist of positively charged lysine or arginine residues occurring at every third position in S4. The proposed secondary structure of the S4 segment is an ␣ helix. Such a structure would depose the lysine and arginine residue as a spiral strip on the helix. It would rotate outward in response to membrane depolarization. Site-directed mutagenesis is the most direct approach to examining the role of the S4 segment in channel activation. Stuehmer et al (18) showed that the voltage dependence of activation was reduced in proportion to the decrease in the charge of DI–S4 up to 3. Mutation of the charged residues in DII–S4 contributed less to the voltage dependence of activation. Replacement of DIV–S4 residues with cysteine and evaluation of the reactivity of the substituted residues to methanethiosulfonate derivatives support the hypothesis of outward movement of S4 with channel activation (19). Much less is known about the mechanism(s) by which outward movement of the S4 segments increase Na⫹ permeation. A logical proposal is that the outward movement of the S4 segment is transmitted to the S5–S6 loop that forms the selectivity filter. In principle, movements of fractions of an angstrom could lead to substantial change in permeability of the ion conduction pathway. Mutagenesis of the selectivity filter residues changes conductance but not activation gating. Activation is probably the result of the transmission of conformation changes over a range of structures within the protein.
CHANNEL INACTIVATION Early sodium current measurements provide clues to the location of channel structures that are crucial to inactivation. Perfusion of the cytoplasmic surface of the membrane with endopeptidases disrupted inactivation, but left activation intact (20). Experiments by Bezanilla and Armstrong (21) showed that inactivation was weakly voltage dependent. This observation suggested that the structures that control inactivation did not extend deeply into the membrane field. Antibodies directed at epitopes in the linker between DIII and DIV, IDIII/IV, disrupted inactivation (22). A mutation in the IDIII/IV linker that separated DI–DIII from DIV disrupted inactivation (18). Patton et al (23) systematically deleted 10 amino segments in March 2001
THE AMERICAN JOURNAL OF MEDICINE威
Volume 110 299
Sodium Channels and Cardiac Arrhythmias/Grant
the IDIII/IV linker and was able to localize the segment that was crucial for inactivation. Mutation of the hydrophobic triplet isoleucine, phenylalenine, methionine (IFM) to glutamine, glutamine, glutamine (QQQ) completely disrupted inactivation (24). When the phenylalenine residue was replaced by cysteine, reactivity of the cysteine changed with the onset of inactivation (25). The experiments suggest that a portion of the IDIII/IV linker acts as a lid that folds back in the membrane and occludes the channel pore during inactivation. Segments at the cytoplasmic end of DIII, S5, 6, and DIV S6 act as a receptor or docking site for the IDIII/IV linker (26,27). More recent studies have shown that mutations over a fairly wide area of the sodium channel ␣ subunit can destabilize inactivation. This suggests that the structural requirement for normal inactivation is fairly precise.
COUPLING OF ACTIVATION AND INACTIVATION
ANTIARRHYTHMIC DRUG BINDING SITE
Hodgkin and Huxley proposed that sodium channel activation and inactivation were independent processes. Early studies supported independence of the two processes. When inactivation was “removed” by perfusion of the cytoplasmic surface of the membrane with endopeptidase, activation remained intact (20). More recent single channel measurements support the coupling of activation and inactivation. Using single channel recordings, Aldrich and Stevens (28) showed that the inactivation of closed channels was slow and strongly voltage dependent. However, inactivation of open channels was rapid and weakly voltage dependent. Some mutations in IDIII/IV, for example, the naturally occurring three amino-acid deletion mutation ⌬KPQ destabilizes inactivation, converting it to a rapidly reversible process, and also speeds up sodium channel activation (29). Kuo and Bean (30) proposed a model of sodium channel gating in which the rate of channel inactivation increases as the channel makes successive transitions along the pathway to opening.
AUXILIARY BETA SUBUNIT The expression of the sodium channel ␣ subunit alone in frog oocytes produces current with slowed inactivation. Coexpression with low molecular weight RNAs recapitulates the rapid inactivation of the wild-type channel (31,32). Further studies have shown that the modulating factor is the 1 subunit (33). This subunit is expressed in neuronal tissues, heart, and skeletal muscle. The single gene that encodes the 1 subunit in all three tissues has been cloned and sequenced (34). The 1 subunit has a small carboxy terminus, a single membrane-spanning segment, and a long extracellular amino terminus with immunoglobulin-like folds and concensus sites for Nlinked glycosylation (35). Initial studies suggest that the 300
March 2001
THE AMERICAN JOURNAL OF MEDICINE威
1 subunit interacts with domains I and IV of the ␣ subunit (36). The sodium channel complex may be coupled to the subsarcolemmal cytoskeleton through the 1 subunit and the protein ankyrin (37). Coexpression of the 1 subunit with the neuronal or skeletal muscle ␣ subunit in frog oocytes increases the level of channel expression and the rate of channel activation and inactivation. The voltage dependence of steady-state inactivation is shifted to more negative membrane potentials. The functional role of 1-subunit coexpression with the cardiac sodium channel ␣ subunit is uncertain; some studies show no effect on gating, whereas others show an acceleration of channel inactivation (38,39). When the ␣ and 1 subunits are coexpressed in mammalian cells, for example, HEK 293 cells, voltage dependence of inactivation is shifted to more depolarized potentials (40).
Volume 110
Sodium channel blockers, such as lidocaine and quinidine, remain a widely prescribed group of drugs. They terminate arrhythmias by slowing cardiac conduction and prolonging refractoriness. Models of drug action postulate that they exert their effects by binding to a specific site on the channel protein (41,42). By analogy with the calcium channel blockers, DIV/S6 was examined as a site of action of sodium channel blockers using alanine scanning (43). Sites for access to and for binding were first identified for local anesthetics on the brain sodium channel. Mutations in the corresponding regions of the cardiac sodium channel confirm the importance of the DIV/S6 sites for access and binding of the local anestheticclass anti-arrhythmic drugs (44). The changes in binding produced by mutations at the putative receptor site were modest. Further, these mutations also modified channel gating, and the changes in gating may contribute to the lower drug affinity. Recently, Sunami et al have shown that the cysteine residue in DI/S5– 6 that is important in tetrodotoxin resistance may provide a cardiac-selective binding site for sodium channel blockers (45). Identification of the other residues that contribute to binding need to be explored. When coupled with data on tertiary structure, synthesis of more specific drugs would be feasible. Several significant hurdles need to be overcome before the potential of such a direct approach to drug design can be realized.
ROLE OF THE CARDIAC SODIUM CHANNEL IN GENESIS OF INHERITED CARDIAC ARRHYTHMIAS The cardiac action potential is orchestrated by the sequential changes of the permeability of specific ion chan-
Sodium Channels and Cardiac Arrhythmias/Grant
perspective, I shall describe the abnormal phenotypes first, followed by a description of their molecular basis.
Loss of Function Mutations: Brugada Syndrome
Figure 3. Membrane current that generates the action potential in working myocardial cells (atrial and ventricular) and pacemaker cells (sino-atrial node). Facsimile of typical action potentials of atrial and ventricular and sino-atrial node cells are presented in the middle of the figure. INa ⫽ sodium current; ICa-L ⫽ the L-type calcium current; ICaT ⫽ the T-type calcium current; INS ⫽ nonspecific cation current; INa/Ca ⫽ the sodium– calcium exchanger current; If ⫽ the hyperpolarization activated current; INa-B ⫽ the background sodium current; IK, IK1, ITO, and I(ACh) ⫽ potassium currents; ICl ⫽ the chloride current.
nels. The ion channels that contribute to the generation of the action potential are summarized in Figure 3. The sodium channel makes a major contribution to two phases of the action potential, the period of rapid depolarization or phase zero, and the plateau or phase two of the action potential. Mutations in the sodium channel contribute to the genesis of cardiac arrhythmias by one of two mechanisms: 1) loss of function mutations that lead to the synthesis of nonfunctional channels or channels that inactivate rapidly; those mutations lead to a reduction in available sodium current during phase zero of the action potential; 2) gain of function mutations that result in slowing or an increase in the reversibility of inactivation and an increase in the late component of sodium current. The increase in late current prolongs the action potential duration and the QT interval on the ECG. Historically, the abnormal phenotypes were described first. Only years later were the genetic and functional bases of the phenotype identified. In keeping with this
In 1992, Brugada and Brugada (46) described a syndrome of recurrent episodes of aborted sudden death, right bundle branch block, persistent ST segment elevation in leads V1 to V3, and normal QT interval in 8 patients with structurally normal hearts (Figure 4). The index arrhythmia was polymorphic ventricular tachycardia. A similar arrhythmia was inducible at electrophysiology in 4 patients by programmed stimulation. The HV interval was prolonged in these patients. Follow-up of larger groups of patients identified additional features of the syndrome (47,48). Transient normalization of the ECG may be observed. In patients showing normalization, the characteristic ECG pattern could be unmasked by administration of the sodium channel blockers ajmaline, procainamide, and flecainide (49). The case fatality rate was high. Over a mean follow-up of 3 years, mortality was 31% in patients receiving no therapy, 36% in patients treated with amiodarone and/or -adrenergic blockers, and 0% in patients receiving an implantable defibrillator. In the 1980s, the Centers for Disease Control and Prevention reported a high incidence of sudden death among young males of Southeast Asian descent. Nademanee et al (50) showed that a subgroup of these individuals had the ECG features characteristic of Brugada syndrome. Chen et al (51) used the candidate gene approach to examine six families with Brugada syndrome. In one family two C to T base substitutions were identified in the sodium channel gene SCN5A. One substitution, an arginine by a tryptophan at codon 1232 (R1232W) was located in the S1–S2 extracellular loop of domain III. This is believed to be a rare polymorphism. The second substitution was of a highly conserved threonine by methionine at codon 1620 (T1620 M), also located in an extracellular loop, S3–S4, of domain 4. When expressed in frog oocytes, the R1232W/T1620 M gene gave sodium currents with half-potential for inactivation shifted to more positive potential. The shift is increased when the double mutant is coexpressed with the  subunit. When expressed in mammalian cells (eg, HEK 293 cells) the T1620 M mutation produced sodium currents with accelerated decay kinetics. Raising the temperature from 22 to 32⬚ C markedly increased the rate of inactivation (52). In another family a two-nucleotide insertion interrupted the splice donor sequence of an intron. In another family a single nucleotide deletion produced a stop codon, resulting in the elimination of DIII/S6 and DIV S1–S6. In approximately 50% of patients with Brugada syndrome, no genetic defect has been identified. Yan and Antzelevitch (53) have proposed a mechanism for the ST segment shift and the occurrence of ventricular fibrillation in Brugada syndrome. The ST segment elevaMarch 2001
THE AMERICAN JOURNAL OF MEDICINE威
Volume 110 301
Sodium Channels and Cardiac Arrhythmias/Grant
Figure 4. ECG in Brugada syndrome. The 12-lead ECG was obtained on a 67-year-old patient with a 10-year history of recurrent syncope. The ECG shows J point element in leads V1 to V3, consistent with “apparent RBBB and T-wave inversion in leads V1 to 4
tion is the result of an epicardium to endocardium voltage gradient during the early repolarization phases of the action potential. The transient outward current, ITO, is the dominant outward current during this phase of the action potential. ITO magnitude is greater in epicardial myocytes than in endocardial myocytes, and greater in the right ventricle than in the left ventricle. A reduction in the size of the balancing inward sodium current would produce the most marked reduction in the action potential duration in the epicardium myocytes of the right ventricle, and would produce the epicardial to endocardial voltage gradient. Brugada syndrome–associated mutations reduce the number of functional sodium channels or accelerate their inactivation. The proposed mechanism is consistent with observations that the characteristic ECG pattern may be precipitated in patients with the latent syndrome by administration of sodium channel blocking drugs. An increase in the magnitude of the transient inward potassium ITO would have a similar effect, contributing to the dispersion of repolarization between epicardial and endocardial myocytes. However, this mechanism has not been identified in Brugada syndrome patients to date.
Loss of Function Mutations: Progressive Conduction System Disease Lenegre (54) and Lev et al (55) described a progressive conduction system disease with right or left bundle branch block leading to complete heart block. Middleaged or elderly patients usually present with syncope or complete heart block. Histologically, the heart demonstrates fibrosis of the conduction system. However, similar fibrosis may be observed in patients without conduction system disease. Recently, Schott et al described two families with syncope and right or left bundle branch block (56). The de302
March 2001
THE AMERICAN JOURNAL OF MEDICINE威
Volume 110
fect mapped to the locus of the sodium channel gene SCN5A. There was a donor splicing in intron 22 that produced a sodium channel lacking DIII S4. Although no functional data were presented, one would predict that the mutant channel would be nonfunctional. The reason that this defect would present only later in life is not clear. The conduction defect may require the presence of the mutant channel and the structural changes that occur with aging for its expression. In principle, a mutation that results in the decrease of the number of functional channels may also cause Brugada syndrome. However, the ECG changes characteristic of that syndrome were not identified in either family.
Gain of Function Mutations: LQT3 The syndrome of familial QT interval prolongation, polymorphic ventricular tachycardia (VT), and sudden death (LQTS) has been linked to inherited defects of membrane ion channels or their regulatory subunits. The ion channel defect maps to a gene that encodes the sodium channel, SCN5A in 10% to 25% of patients with LQTS. The subgroup of patients with sodium channel defects (LQT3) presents later in life than those with potassium channel defects. However, the initial event is more likely to be fatal (57). Patients with LQT3 tend to have a resting bradycardia. However, shortening of the QT interval in response to exercise is exaggerated. Attacks of ventricular tachycardia or cardiac arrest are more likely to occur at rest or during sleep in LQT3. Sodium channel blockers, for example, mexiletine and flecainide shorten the QT interval in LQT3 (58,59). However, they have not been shown to prevent arrhythmia recurrence in LQT3. Expression of both the wild-type and LQT3 mutant sodium channels in the frog oocyte and mammalian cells have permitted exploration of the molecular basis of LQT3. At most membrane potentials, sodium channels
Sodium Channels and Cardiac Arrhythmias/Grant
potentials such that repetitive depolarizations occur at the level of the action potential plateau (Figure 5B). These depolarizations are termed early afterdepolarizations. When the gating defects of LQT3 are incorporated into a computer model of the cardiac action potential, prolongation and early afterdepolarizations could be reproduced (63). Early afterdepolarizations are believed to initiate polymorphic VT and ventricular fibrillation in LQTS. These arrhythmias may be sustained by reentry. The bradycardia and the frequency dependence of the QT interval in LQT3 have not been adequately explained to date. The sodium-channel– blocking drugs, lidocaine, mexiletine, and flecainide, selectively block the late component of sodium current in wild-type and LQT3 mutant channels (40,64). The selective blockade of the late current could account for the QT interval shortening without QRS prolongation observed with mexiletine and flecainide. Figure 5. Single sodium channel currents recorded from cells expressing the wild-type (A) and LQT3 mutant sodium channel ␦KPQ (B) and the presumed cellular mechanism of the associated arrhythmia. In panels A and B, single sodium channel currents were recorded with the patch-clamp technique. Membrane currents were elicited with 200-ms depolarizing pulses from ⫺100 to ⫺40 mV. Each panel shows the current observed with 10 consecutive depolarizing trials. Sodium channel openings are downward deflections from the baseline. Panel C shows the sequence of events associated with the development of abnormal activity in LQT3. Action potential prolongation precedes the development of single or multiple secondary depolarizations at the plateau level of membrane potential. Those secondary depolarizations occurring at the level of the plateau are examples of abnormal automaticity, termed early afterdepolarizations.
open once in response to depolarization and close by inactivation. The inactivation process is usually not reversible unless repolarization occurs. In fewer than 5% of depolarizations, sodium channels may re-open at a later time during depolarization only to close again by inactivation (Figure 5A). As a rare event (approximately .1% of depolarizations), the inactivation process may fail altogether and the sodium channel may open repetitively for hundreds of milliseconds. These late components of sodium current contribute inward depolarizing current during the plateau of the action potential, prolonging its duration. Several sodium channel mutations in the cytoplasmic loop between domains III and IV (IDIII/IV), consisting of single base substitutions, deletions, or insertions decrease the stability of inactivation and produce an increase in the size of the late components of sodium current (60). Other mutations slow the rate of inactivation by prolongation of the duration of individual channel openings (61,62). Marked prolongation of the action potential alters the voltage-time trajectory of the action
The Overlap Syndrome As genetic studies are applied to larger groups of patients, it is becoming clear that there may be considerable overlap in syndromes associated with sodium channel mutations. Bezzina et al (65) described an eight-generation kindred with a high incidence of nocturnal sudden death, QT interval prolongation, and ECG changes characteristic of Brugada syndrome. The molecular defect was the insertion of an aspartate residue in the carboxy terminal domain of the sodium channel ␣ subunit. The functional consequences included generation of a persistent component of sodium current during maintained depolarization and an increase in inactivation (66,67). The former effect would prolong the action potential duration and the QT interval; the latter effect would reduce the available sodium current during phase one of the action potential. As we have discussed above, flecainide blocks the late component of sodium current and has been proposed as a treatment for LQTS. Priori et al (68) administered flecainide intravenously to 13 patients with LQT3. As expected, the QT interval was shortened. However in 6 patients, concomitant ST-segment elevation in leads V1 to V3 characteristic of Brugada syndrome was observed. These observations suggest that the relationship between the underlying gene defect and the phenotype is dynamic. The phenotype will depend on the ionic current that makes contributions to the action potential under specific circumstances. A view of these overlapping relationships are summarized in Figure 6.
SUMMARY The cardiac sodium channel is a complex multimeric protein consisting of an ␣ subunit and a regulatory 1 March 2001
THE AMERICAN JOURNAL OF MEDICINE威
Volume 110 303
Sodium Channels and Cardiac Arrhythmias/Grant
Figure 6. Potential relationship of the inherited defects of the cardiac sodium channel. A single mutation may produce lone ventricular tachycardia/fibrillation of the type LQTS or Brugada syndrome, or heart block.
subunit. The basic properties of ion conduction and gating reside in the ␣ subunit. The use of specific toxins and site-directed mutagenesis has localized the channel domains that are responsible for ion conduction and inactivation. Mutations in the sodium channel gene result in the long QT and Brugada syndromes. These syndromes are associated with a high case fatality rate. The implantable defibrillator is the most effective form of treatment for either syndrome.
REFERENCES 1. Hodgkin AL, Huxley AF. A quantitative description of membrane current and its application to conduction and excitation in nerve. J Physiol. 1952;117:500 –544. 2. Fozzard H, Hanck DA. Structure and function of voltage-dependent sodium channels: comparison of brain II and cardiac isoforms. Physiolog Rev. 1996;76:887–926. 3. Noda M, Shimizu S, Tanabe T, et al. Primary structure of electrophorus electricus sodium channel deduced from cDNA sequence. Nature. 1984;312:121–127. 4. Noda M, Ikeda T, Kayano T, et al. Existence of distinct sodium channel messenger RNAs in rat brain. Nature. 1986;320:188 –192. 5. Rogart RB, Cribbs LL, Muglia LK, Kephart DD, Kaiser MW. Molecular cloning of a putative tetrodotoxin-resistant rat heart Na⫹ channel isoform. Proc Natl Acad Sci. 1989;86:8170 – 8174. 6. Gellens ME, George AL Jr, Chen L, et al. Primary structure and functional expression of the human cardiac tetrodotoxin-insensitive voltage-dependent sodium channel. Proc Natl Acad Sci. 1992; 89:554 –558. 304
March 2001
THE AMERICAN JOURNAL OF MEDICINE威
Volume 110
7. Cohen SA, Barchi RL. Voltage-dependent sodium channels. Internat Rev Cytol. 1993;137C:55–103. 8. Doyle DA, Cabral JM, Pfuetzner RA, et al. The structure of the potassium channel: molecular basis of K⫹ conduction and selectivity. Science. 1998;280:69 –77. 9. Satin J, Kyle JW, Chen M, et al. A mutant of TTX-resistant cardiac sodium channels with TTX-sensitive properties. Science. 1992;256: 1202–1205. 10. Backx PH, Yue DT, Lawrence JH, Marban E, Tomaselli GF. Molecular localization of an ion-binding site within the pore of mammalian sodium channels. Science. 1992;257:248 –251. 11. Lipkind GM, Fozzard HA. A structural model of tetrodotoxin and saxitoxin binding site of the Na⫹ channel. Biophys J. 1994;66:1–13. 12. Nuss HB, Balser JR, Orias DW, Lawrence JH, Tomaselli GF, Marban E. Coupling between fast and slow inactivation revealed by analysis of a point mutation (F1304Q) i 1 rat skeletal muscle sodium channels. J Physiol. 1996;494:2: 411– 429. 13. Perez-Garcia MT, Chiamvimonvat N, Marban E, Tomaselli GF. Structure of the sodium channel pore revealed by serial cysteine mutagenesis. Proc Natal Acad Sci USA. 1996;93:300 –304. 14. Heinemann SH, Terlau H, Stuehmer W, Imoto K, Numa S. Calcium channel characteristics conferred on the sodium channel by single mutations. Nature. 1992;356:441– 443. 15. Santana LF, Gomez AM, Lederer WJ. Ca2⫹ flux through promiscuous cardiac Na⫹ channels: slip-mode conductance. Science. 1998; 279:1027–1033. 16. Sorbera LA, Morad M. Atrionaturitic peptide transforms cardiac sodium channels into calcium-conducting channels. Science. 1990; 247:969 –973. 17. Armstrong CM. Sodium channels and gating currents. Physiol Rev. 1981;61:644 – 683. 18. Stuehmer W, Conti F, Suzuki H, et al. Structural parts involved in activation and inactivation of the sodium channel. Nature. 1993; 339:597– 603. 19. Yang N, Horn R. Evidence for voltage-dependent S4 movement in sodium channels. J Gen Physiol. 1995;15:213–218. 20. Armstrong CM, Bezanilla F, Rogas E. Destruction of sodium conductance inactivation in squid axons perfused with pronase. J Gen Physiol. 1973;62:375–391. 21. Armstrong CM, Bezanilla F. Inactivation of the sodium channel. II. Gating current experiments. J Gen Physiol. 1977;70:567–590. 22. Vassilev PM, Scheuer T, Catterall WA. Identification of an intracellular peptide segment involved in sodium channel inactivation. Science. 1988;241:1658 –1661. 23. Patton DE, West JW, Catterall WA, Goldin AL. Amino acid residues required for fast Na⫹-channel inactivation: charge neutralizations and deletions in the III-IV linker. Proc Natl Acad Sci. 1992;89: 10905–10909. 24. West JW, Patton DE, Scheuer T, Wang Y, Goldin AL, Catterall WA. A cluster of hydrophobic aminoacid residues required for fast Na⫹channel inactivation. Proc Natl Acad Sci. 1992;89:10910 –10914. 25. Kellenberger S, Scheuer T, Catterall WA. Movement of the Na⫹ channel inactivation gate during inactivation. J Biol Chem. 1996; 271:30971–30979. 26. McPhee JC, Ragsdale DS, Scheuer T, Catterall WA. A critical role for transmembrane segment IVS6 of the sodium channel ␣ subunit in fast inactivation. J Biological Chem. 1995;270:12025–12034. 27. Smith MR, Goldin AL. Interaction between the sodium channel inactivation linker and domain III S4-S5. Biophys J. 1997;73:1885– 1895. 28. Aldrich RW, Stevens CF. Voltage-dependent gating of single sodium channels from mammalian neuroblastoma cells. J Neurosci. 1987;7:418 – 431. 29. Chandra R, Starmer CF, Grant AO. Multiple effects of KPQ dele-
Sodium Channels and Cardiac Arrhythmias/Grant
30. 31. 32.
33.
34.
35.
36.
37.
38.
39.
40.
41. 42.
43.
44.
45.
46.
47.
48.
tion mutation on gating of human cardiac Na⫹ channels expressed in mammalian cells. Am J Physiol. 1998;274:H1643–H1654. Kuo C-C, Bean BD. Na⫹ channels must deactivate to recover from inactivation. Neuron. 1994;12:819 – 829. Goldin AL. Accessory subunits and sodium channel inactivation. Curr Opin Neurobiol. 1993;3:272–277. Krafte DS, Snutch TP, Leonard JP, Davidson N, Lester HA. Evidence for the involvement of more than one mRNA species in controlling the inactivation process of rat and rabbit brain Na channels expressed in Xenopus oocytes. J Neurosci. 1988;8:2859 –2868. Isom LL, Scheuer T, Brownstein AB, Ragsdale DS, Murphy BJ, Catterall WA. Function co-expression of the 1 and type IIA ␣ subunits of sodium channels in a mammalian cell line. J Biol Chem. 1995;270:3306 –3312. Makita N, Bennett PB Jr, George AL Jr. Voltage-gated Na⫹ channel 1 subunit mRNA expressed in adult human skeletal muscle, heart and brain is encoded by a single gene. J Biol Chem. 1994;269:7571– 7578. Isom LL, De Jongh KS, Patton DE, et al. Primary structure and functional expression of the 1 subunit of the rat brain sodium channel. Science. 1992;256:839 – 842. Makita N, Bennett PB, George AL Jr. Molecular determinants of 1 subunit-induced gating modulation in voltage-dependent Na⫹ channels. J Neurosci. 1996;16:7117–7127. Malhotra JD, Kazen-Gillespie K, Hortsch M, Isom LL. Sodium channel  subunits mediate homophilic cell adhesion and recruit Ankyrin to points of cell– cell contact. J Biol Chem. 2000;275: 11383–11388. Makita N, Bennett PB Jr, George AL Jr. Voltage-gated Na⫹ channel 1 subunit mRNA expressed in adult human skeletal muscle, heart and brain is encoded by a single gene. J Biol Chem. 1994;269:7571– 7578. Nuss HB, Chiamvimonvat N, Perez-Garcia MT, Tomaselli GF, Marban E. Functional association of the 1 subunit with human cardiac (hH1) and rat skeletal muscle (l) sodium channel ␣ subunits expressed in Xenopus oocytes. J Gen Physiol. 1995;106:1171– 1191. An RH, Wang XL, Kerem B, et al. Novel LQT-3 mutations affects Na⫹ channel activity through interactions between ␣- and 1-subunits. Circ Res. 1998;83:141–146. Hille B. Local anesthetics: hydrophilic and hydrophobic pathways for the drug-receptor reaction. J Gen Physiol. 1977;69:497–515. Hondeghem LM, Katzung BG. Time- and voltage-dependent interactions of antiarrhythmic drugs with cardiac sodium channels. Biochim et Biophys Acta. 1977;472:373–398. Ragsdale DS, McPhee JC, Scheuer T, Catterall WA. Molecular determinants of state-dependent block of Na⫹ channels by local anesthetics. Science. 1994;265:1724 –1728. Ragsdale DS, MePhee JC, Scheuer T, Catterall WA. Common molecular determinants of local anesthetic, antiarrhythmic and anticonvulsant block of voltage-gated Na⫹ channels. Proc Natl Acad Sci USA. 1996;93:9270 –9275. Sunami A, Dudley SC Jr, Fozzard HA. Sodium channel selectivity filter regulates antiarrhythmic drug binding. Proc Natl Acad Sci USA. 1997;94:14126 –14131. Brugada P, Brugada J. Right bundle branch block, persistent ST segment elevation and sudden cardiac death: a distinct clinical and electrocardiographic syndrome. J Am Coll Cardiol. 1992;20:1391– 1396. Brugada J, Brugada P. Further characterization of the syndrome of right bundle branch block, ST segment elevation, and sudden cardiac death. J Cardiovasc Electrophysiol. 1997;8:325–331. Brugada J, Brugada R, Brugada P. Right bundle-branch block and ST-segment elevation in leads V1 through V3. Circulation. 1998;97: 457– 460.
49. Brugada R, Brugada J, Antzelevitch C, et al. Sodium channel blockers identify risk for sudden death in patients with ST-segment elevation and right bundle branch block but structurally normal hearts. Circulation. 2000;101:510 –515. 50. Nademanee K, Veerakul G, Nimmannit S, et al. Arrhythmogeneic marker for the sudden unexplained death syndrome in Thai men. Circulation. 1997;96:2595–2600. 51. Chen Q, Kirsch GE, Zhang D, et al. Genetic basis and molecular mechanism for idiopathic ventricular fibrillation. Nature. 1998; 392:293–296. 52. Dumaine R, Towbin JA, Brugada P, et al. Ionic mechanisms responsible for the electrocardiographic phenotype of the Brugada syndrome are temperature dependent. Circ Res. 1999;85:803– 809. 53. Yan G-X, Antzelevitch C. Cellular basis for the Brugada syndrome and other mechanisms of arrhythmogenesis associated with STsegment elevation. Circulation. 1999;100:1660 –1666. 54. Lenegre J. Etiology and pathology of bilateral bundle branch block in relation to complete heart block. Prog Cardiovasc Dis. 1964;6: 409 – 444. 55. Lev M, Kinare SC, Pick A. The pathogenesis of atrioventricular block in coronary disease. Circulation. 1970;42:409 – 425. 56. Schott JJ, Alshinawi C, Kyudt F, Probst V, Hoorntje TM, Hulsbeek M, Wilde AA, Esconde D, Manners MM, Le Marec H. Cardiac conduction defect associated with SCN5A. Nature Genetics. 1999; 23:20 –21. 57. Zareba W, Moss AJ, Schwartz PJ, et al. Influence of the genotype on the clinical course of the long-QT syndrome. N Engl J Med. 1998; 339:960 –965. 58. Benhorin J, Taub R, Goldmit M, et al. Effects of flecainide in patients with new SCN5A mutation. Circulation. 2000;101:1698 – 1706. 59. Schwartz PJ, Priora SG, Locati EH, et al. Long QT syndrome patients with mutations of the SCN5A, and HERG genes have differential responses to Na⫹ channel blockade, and to increases in heart rate. Circulation. 1995;92:3381–3386. 60. Bennett PB, Yazawa K, Makita N, George AL Jr. Molecular mechanism for an inherited cardiac arrhythmia. Nature. 1995;376:683– 685. 61. Kambouris NG, Nuss HB, Johns DC, Tomaselli GF, Marban E, Balser JR. Phenotypic characterization of a novel long-QT syndrome mutation (R1623Q) in the cardiac sodium channel. Circulation. 1997;97:640 – 644. 62. Makita N, Shirai N, Nagashima M, et al. A de novo missense mutation of human cardiac Na⫹ channel exhibiting novel molecular mechanisms of long QT syndrome. FEBS Letters. 1998;423:5–9. 63. Clancy CE, Rudy Y. Linking a genetic defect to its cellular phenotype in a cardiac arrhythmia. Nature. 1999;400:566 –569. 64. Nagatomo T, January CT, Makielski JC. Preferential block of late sodium current in the LQT3 ␦KPQ mutant by the class IC antiarrhythmic flecainide. Molec Pharmacol. 2000;57:101–107. 65. Bezzina C, Veldkamp MW, van den Berg MP, et al. A single Na⫹ channel mutation causing both long-QT and Brugada syndromes. Circ Res. 1999;85:1206 –1213. 66. Veldkamp MW, Viswanathan PC, Bezzina C, Baartscheer A, Wilde AAM, Balser JR. Two distinct congenital arrhythmias evoked by a multidysfunctional Na⫹ channel. Circ Res. 2000;86:E91–E97. 67. Wang DW, Markita N, Kitabatake A, Balser JR, George AL Jr. Enhanced Na⫹ channel intermediate inactivation in Brugada syndrome. Circ Res. 2000;87:E37–E43. 68. Priori SG, Napolitano C, Schwartz PJ, Bloise R, Crotti L, Ronchetti E. The elusive link between LQT3 and Brugada syndrome. Circulation. 2000;102:945–947.
March 2001
THE AMERICAN JOURNAL OF MEDICINE威
Volume 110 305
