Cardiac sodium channel mutations: Why so many phenotypes?

Abstract

Mutations of the cardiac Na⁺ channel (Na_v1.5) induce channel gain and loss of function. Gain-of-function mutations cause Long QT syndrome type 3 and possibly atrial fibrillation, while a wider variety of phenotypes are associated with loss-of-function channel mutations, such as Brugada syndrome, cardiac conduction disease, sick sinus syndrome, and possibly dilated cardiomyopathy. The phenotypes of Na_v1.5 mutations vary with the mutation-mediated effects on channel biophysics, with age, gender, body temperature, and time, and within regions of the heart. This phenotypic variability makes genotype-phenotype correlations difficult. In this Perspective, we review the clinical syndromes seen with monogenetic SCN5A mutations and propose that phenotypic variability not ascribed to mutation-dependent changes in channel function may be the result of variations in additional modifiers of channel behavior. Consideration of these modifiers could help improve genotype-phenotype correlations and lead to new therapeutic strategies.

Introduction

A functional heart pumps blood throughout the body by regular, rhythmic contractions, which are controlled by electrical impulses usually initiated at specialized pacemakers and conducted through the heart, in part, by a myocardial conducting system. Under usual circumstances, the sinoatrial (SA) node, also known as the pacemaker, generates the initial depolarization that stimulates the atrial muscle to contract. From there, the signal travels to the atrioventricular (AV) node, conducts through the bundle of His and bundle branches to the Purkinje fibers. The Purkinje fibers articulate with the subendocardial muscle, ensuring ventricular contraction from apex to base of the heart. This sequence of electrical activity can be monitored in the electrocardiogram (ECG).

The importance of cardiac Na⁺ channel (Na_v1.5) for cardiac electrical stability is highlighted by lethal arrhythmias as the result of inherited Na_v1.5 channel gene defects. Alternations in cardiac Na_v1.5 have also been implicated in arrhythmic risk observed with acquired heart disease.^{1, 2} The cardiac Na_v1.5 is responsible for the fast inward Na⁺ current (I_Na) that initiates the depolarizing phase 0 of the cardiac action potential (AP), except in the SA and AV node. In addition to intercellular communication via gap junctions, Na_v1.5 is a major determinant of impulse conduction velocity of cardiac tissues.³

Na⁺ channel biophysics is complex. The α subunit of Na_v1.5 encoded by SCN5A forms the ion conducting pore and changes its structural conformation on a time scale of milliseconds in response to the changes in membrane potential. This process is termed voltage-dependent gating.⁴ The α subunit is composed of intracellular N and C termini, and four homologous domains (I-IV) that form a pore to conduct Na⁺ ions across the cell membrane.⁵ The four domains are attached to one another by cytoplasmic linker sequences with important residues for channel gating and post-translational modifications such as phosphorylation. Each domain contains six transmembrane segments (S1–S6). The four positively charged S4 segments constitute the voltage sensor, responsible for increased channel permeability during membrane depolarization.⁶ The S5 and S6 segments of all four domains form the ion-conducting pore, and the interconnecting P-loops make up the channel’s selectivity filter.^{7, 8} The α subunit interacts with accessory proteins, including β subunits and other interacting proteins, to form a macromolecular complex that modulates channel expression and function.⁹

Na_v1.5 Biophysical Properties

Mutations in SCN5A produce a variety of clinical phenotypes. Some portion of phenotypic variability is the result of mutations’ effects on the Na_v1.5 biophysical properties. These effects are often divided into gain and loss of channel function (i.e. increased or decreased I_Na). In gain-of-function mutations, it is generally thought that an increase in the persistent late I_Na (I_NaL) during the AP plateau, rather than in peak I_Na, is responsible for the phenotype. As illustrated in Figure 1, gain-of-function SCN5A mutations can result in increased late I_Na, leading to type 3 long-QT syndrome (LQT3). Loss-of-function SCN5A mutations, on the other hand, can lead to decreased peak I_Na, resulting in phenotypes including Brugada syndrome (BrS), sick sinus syndrome (SSS), cardiac conduction diseases (CCD) and possibly dilated cardiomyopathy (DCM). Some SCN5A mutations have also been shown to cause mixed phenotypes (overlap syndromes) or linked to familial lone atrial fibrillation (AF). Occasionally, mutations can cause both an increase in I_NaL and a simultaneous reduction in peak I_Na, thereby having properties of both gain and loss of channel function. The effect of mutations and their contribution to phenotypic variability have been extensively studied and reviewed.^10–14

Figure 1.

Mutations in SCN5A can produce various clinical phenotypes. Gain-of-function SCN5A mutations can result in increased late I_Na, leading to type 3 long-QT syndrome (LQT3). Loss-of-function SCN5A mutations can lead to decreased peak I_Na, the phenotypes associated with which include Brugada syndrome (BrS), sick sinus syndrome (SSS), cardiac conduction diseases (CCD) and possibly dilated cardiomyopathy (DCM). Moreover, SCN5A mutations with both gain late I_Na and loss of peak I_Na can be associated with a mixed phenotype (for example, BrS + LQT3, indicated as overlap syndromes). Similarly, both gain and loss of function mutations have been associated with familial lone atrial fibrillation (AF).

Other Sources of Phenotypic Variability

As discussed above, different SCN5A mutations can generate different clinical syndromes. These syndromes can manifest in a particular chamber or region of the heart. In addition to this variability, the phenotype of individual SCN5A mutations can show a variable extent of penetrance between individuals and can vary within an individual as a result of such factors as age, time of day, and body temperature. The origin of this phenotypic variability is unclear, and this degree of variability makes genotype-phenotype correlations difficult. The lack of correlation complicates medical decision-making in patients with known mutations.

In Figure 2, we present examples of factors that can contribute to the phenotypic variability of a SCN5A mutation. The clinical phenotypic penetrance and manifestation of a SCN5A mutation can vary with gender (upper left panel: LQT3), the presence of genetic modifiers (upper right panel: H558R polymorphism), circadian rhythm (lower left panel: BrS), and ageing (lower right panel: CCD).

Figure 2.

Examples of phenotypic variability with time, ageing, genetic background and gender in patients carrying a SCN5A mutation. Upper left panel: Gender can impact the phenotype of LQT3 syndrome. Among LQT3 patients, men tend to have longer QTc intervals than women.¹⁵ Upper right panel: Phenotype of a SCN5A mutation can be altered by genetic variants within or outside of SCN5A gene. For example, SCN5A H558R polymorphism has been shown to rescue the phenotype of P2006A and R282H mutations in SCN5A.^{27, 28} Lower left panel: BrS is known to have circadian changes in ECG patterns and the susceptibility to lethal ventricular fibrillation. Typical Brugada ECG with ST segment elevation in precordial leads is more frequently observed in the midnight-to-early morning period during sleep.¹¹¹ Lower right panel: SCN5A mutations causing progressive CCD have been shown to demonstrate deterioration in ventricular conduction and widened QRS complex with ageing.¹¹² Illustrations adapted from references¹¹¹, and¹¹² with permission.

Gender is known to influence the effect of SCN5A mutations. In LQT3 patients, men tend to have longer QTc intervals than women (Figure 2, upper left panel).¹⁵ In addition, BrS affects far more men than women.^16–18 Gender can influence the clinical phenotype of an individual mutation. For example, within a single family, four male members carrying a loss-of-function SCN5A mutation, G1406R, had BrS and seven other family members carrying the same mutation, six of which were women, had CCD.¹⁷ The mechanism of gender’s effects is still unclear. Sex hormone may cause differences in QT intervals and gene expression of cardiac ion channels.¹⁹ For female patients, hormonal changes due to menses and pregnancy could induce QT prolongation and increase the susceptibility to arrhythmias.²⁰

The phenotype of SCN5A mutations can be age-dependent. For example, despite carrying the mutation since birth, BrS patients develop arrhythmic phenotype between the age of 34 and 53 years.²¹ Interestingly, the male predominance of arrhythmic phenotype is not observed in children with BrS.²² Age-dependent hormonal changes has been proposed to explain why the risk of spontaneous arrhythmias in male increases after puberty. Within a patient with a BrS-associated SCN5A mutation, there is phenotypic variability in the ECG manifestations of the disease as a function of time. As shown in Figure 2 lower left panel, a BrS patient presented circadian changes in ECG patterns and the susceptibility to lethal ventricular fibrillation, and ECG manifestations can worsen with fever.²³ In addition, CCD can be progressive and worsen with age (Figure 2 lower right panel).

It is difficult to explain all of the phenotypic variability observed clinically by only considering mutation-mediated changes in channel biophysics. We propose that additional phenotypic variation is the result of variability in channel modifiers. Established modifiers include additional genetic variants in the SCN5A gene (Figure 2, upper right panel). Other potential modifiers include alterations in the expression or function of the mutated channel, hereinafter called intrinsic modifiers, or changes in the expression of other channels or genes that can modify the overall electrical effect of an individual SCN5A mutation, hereinafter called extrinsic modifiers.

Modifiers within the SCN5A Gene

The variability of the phenotypes associated with an SCN5A mutation can result from another genetic variant that co-exists within SCN5A gene. For example, two BrS-associated SCN5A variants T1620M and R1232W are known to cause minor changes in Na_v1.5 channel properties when exist independently.²⁴ When they co-exist, however, the T1620M/R1232W double mutant produces no sodium current.²⁵ Subjects carrying heterozygous loss-of-function mutations W156X or R225W do not have a clinical phenotype.²⁶ Compound heterozygosity for W156X and R225W mutations, however, leads to severe conduction disease, dilated cardiomyopathy and premature death at young age.²⁶ These examples demonstrate that the co-existence of a genetic variant within the SCN5A gene can affect the severity of the phenotype associated with another SCN5A mutation.

In addition, the SCN5A H558R polymorphism has been found to be a disease-modifying genetic variation that can rescue gain-of-function²⁷ or loss-of-function SCN5A mutations (Figure 2, upper right panel),²⁸ providing a potential mechanism to account for the variable penetrance that is often observed with a SCN5A mutation. Taken together, the presence of genetic modifiers within the SCN5A gene may contribute to the variability observed in SCN5A mutation phenotypes, even in familial cases sharing the same mutation.

Other Intrinsic Modifiers

Any stage in the cardiac Na_v1.5 life cycle represents a potential modification point to amplify, localize, or minimize a mutation-mediated phenotype. The life cycle of cardiac Na_v1.5 includes gene transcription, post-transcription RNA processing, translation (protein synthesis and assembly), post-translational trafficking and modifications, and degradation, as demonstrated in Figure 3.^{29, 30} Modifiers such as transcription factors, microRNAs, and kinases regulate these processes and result in functional changes of cardiac Na_v1.5. Although not having established roles in modifying inherited Na_v1.5 channelopathies, these modifiers can vary with age, gender, clinical states of the patient, and regions within the heart. Therefore, conceivable variability in these modifiers (summarized in Figure 4) may contribute to the overall phenotypic variability observed. Moreover, these modifiers are candidates to explain regional localization of Na⁺ channel-mutation effects.

Figure 3.

The life cycle of cardiac Na_v1.5 and potential phenotypic modifiers. The life cycle starts from SCN5A channel gene transcription in nucleus, which can be regulated by transcription factors such as TBX5, NF-κB, and Foxo1. The next steps are channel protein translation and assembling in the endoplasmic reticulum (ER) and Golgi. Properly folded Na_v1.5 protein is trafficked to the cell membrane, while misfolded channel protein as a result of mutations and splicing variants can activate the PERK pathway of the unfolded protein response to downregulate the SCN5A mRNA level. miRNAs also regulate the SCN5A mRNA level. Channel trafficking can be modulated by PKA, PKC and caveolae. Once expressed on the cell membrane, Na⁺ channel interacting proteins (NaChIPs) join the α and β subunits to form a macromolecular complex, which can be regulated by PKA, PKC, oxidative stress (ROS), and metabolic states (NADH/NAD⁺). Eventually, the channel will go through ubiquitin-mediated degradation that can be regulated by Nedd4.

Figure 4.

The “second hit” theory of phenotypic variability of SCN5A mutations. The effects of a SCN5A mutation are subjected to the modulation by secondary modifiers at multiple levels, including transcriptional regulation (such transcription factors and alternative splicing), post-transcriptional regulation (such as miRNAs), translational regulation (such as protein translation and unfolded protein response), post-translational regulation, circadian rhythm, and the regulation by accessory subunits, components of Na_v1.5 macromolecular protein complex, or other ion channel proteins. The alteration of these secondary modifiers can alter the phenotype variability of a mutation, potentially explaining variability among patients with the same mutation, effects of a mutation confined to a particular region of the heart, and variability as a function of time within an individual.

Transcriptional Modifiers

Differential transcriptional regulation

Transcriptional dysregulation of SCN5A (lower part of Figure 3) can lead to either up- or downregulation of Na_v1.5 membrane expression and macroscopic I_Na. This change in current can enhance or mask the phenotype of a mutation. Moreover, regional differences in transcriptional regulation might explain how various areas of the heart and conduction system are most affected.

Some transcriptional factors play essential roles in signaling normal SCN5A expression, dysregulation of which may disrupt SCN5A expression and cause phenotypic variability. TBX5, a T-box transcription factor, is essential for Na_v1.5 expression in the ventricular conduction system.³¹ Removal of TBX5 causes a dramatic reduction of Na_v1.5 and connexin 40 membrane expressions, leading to conduction velocity slowing, cardiac arrhythmias, and increased animal mortality.³¹ Alterations in TBX5 expression could modulate I_Na in the conduction system and contribute to conduction-specific phenotypes. Bmal1, encoding a core molecular clock transcription factor, also plays an important role in SCN5A expression and cardiac arrhythmia susceptibility. Cardiomyocyte-specific deletion of Bmal1 causes slowed heart rate, prolonged electrocardiographic RR and QRS intervals, and increased episodes of arrhythmia in mice as a result of loss of SCN5A expression.³² Alterations in Bmal1 may help explain diurnal variations in arrhythmic risk in the presence of a given mutation.

The SCN5A promoter region contains consensus binding sites for several transcription factors, such as NF-κB and Forkhead box O 1 (Foxo1). When activated by angiotensin II (AngII) or oxidative stress, NF-κB and Foxo1 are upregulated and induce a significant decrease in Na_v1.5 expression and I_Na.^{33, 34} Interestingly, transforming growth factor β1 can counteract Foxo1’s effect and increase the SCN5A mRNA level and I_Na.³⁵ This suggests that the phenotypic penetrance of a mutation could vary as a result of changes in these transcription factors. In addition, single nucleotide polymorphisms (SNPs) in the promoter region of SCN5A have been shown to modulate the transcriptional activity of SCN5A³⁶ and to affect the phenotypic severity in heterozygous carriers of an SCN5A loss-of-function mutation.³⁷ These data suggest genetic variants in the SCN5A promoter region can contribute to the phenotypic variability of SCN5A mutations through regulating SCN5A transcription.

Differential mRNA processing and regulation

Another potential modifier of Na_v1.5 mutations is at the stage of mRNA processing. Recently, our group has shown that AngII and hypoxia can regulate SCN5A mRNA splicing and generate truncated, non-functional Na⁺ channels in human HF.² These abnormally spliced, truncated channels also show a dominant negative effect on the full-length SCN5A mRNA transcript by activating the unfolded protein response sensor protein kinase-R like ER kinase (PERK) (Figure 3).³⁸ AngII can be differentially regulated in local regions of the heart. For example, AngII levels are elevated in ventricular cells but not in atrial cells of type 1 diabetic rats.³⁹ Combining local regulation with induction of abnormal mRNA processing could contribute to a phenotype that varies with clinical states.

MicroRNAs (miRNAs) are endogenous non-coding small ribonucleic acids emerging as important regulators for gene expression by either restricting translation or inducing degradation of the target mRNA. Altered expression of miRNAs, mainly upregulation, has been widely reported to reduce the expression of cardiac ion channels in the pathological studies of cardiovascular diseases.^40–42 In HF, suppression of Na_v1.5 protein expression and I_Na reduction are observed with upregulated miR-125a/b.⁴² A stress inducible miRNA, miR-195, which is upregulated in cardiac hypertrophy and HF,⁴³ can downregulate SCN5A, leading to I_Na reduction.⁴¹ In myocardial infarction, upregulation of let-7f and miR-378 also cause I_Na reduction.⁴² Therefore, changes in miRNAs with time may contribute to phenotypic variability.

Other miRNAs may have indirect effects on the Na_v1.5. For example, miR-199a is downregulated after induction of ischemia, leading to an upregulation of its target hypoxia-inducible factor 1 α. This factor can activate RBM25⁴⁴ and enhance abnormal SCN5A mRNA splicing to generate truncated, non-functional Na⁺ channels.^{2, 38} In human AF and ventricular arrhythmia, miR-1 is markedly decreased, which can induce upregulation of Kir2.1 channel.⁴⁵ This channel conducts an inward K⁺ current I_K1 that controls the resting membrane potential. An increase of I_K1 can lead to membrane hyperpolarization, which can either amplify I_Na reduction induced by loss-of-function SCN5A mutations or minimize I_NaL increase induced by a gain-of-function mutation. In a canine tachy-paced AF model, miR-328 level is increased, which leads to a reduction of its target L-type Ca²⁺ channel current and a shortened atrial APD.⁴⁶ These changes could summate with those of Na_v1.5 loss-of-function mutations to worsen arrhythmic risk.

Translational Modifiers

At the translational level, many modifiers regulate Na_v1.5 protein synthesis, folding, trafficking, and assembling with accessory subunits and interacting proteins. For example, activated mitogen-activated protein kinases (MAPKs) such as p38-MAPK and JNK,^{47, 48} can affect translation of Na_v1.5 channel and connexin 43 and contribute to arrhythmia.^49–52

Differential trafficking to cell membrane

After SCN5A is translated to protein and assembled in the ER, the channel is transited to the Golgi complex, where Na_v1.5 channel trafficking begins. The Golgi vesicle containing the Na_v1.5 channel is then transported to the plasma membrane area. Modifiers such as protein kinase A and C (PKA and PKC) regulate channel trafficking to the plasma membrane. For instance, PKC activation can decrease Na_v1.5 channel trafficking to the plasma membrane, while PKA can play an opposite role.^{53, 54} This suggests that a patient’s clinical state can potentially affect the penetrance of a mutation.

Caveolae are potential Na_v1.5 modifiers, since colocalization of the Na_v1.5 protein and caveolin-3 is observed in ventricular tissues.⁵⁵ Caveolae constitute a reservoir of I_Na that can be modulated by caveolar neck opening and closing.⁵⁶ Caveolae are also enriched in signaling molecules, such as PKC, PKA, Src family tyrosine kinases, and G proteins.^57–59 A complex formed with PKA, caveolin-3 and Gsα protein in caveolae modulates Na_v1.5 function.^{59, 60} PKA and PKC regulate Na_v1.5 gating properties by phosphorylation.⁶¹ A caveolin-3 mutation results in a ~3-fold increase of I_NaL, which could enhance the penetrance of LQT3 induced by Na_v1.5 gain-of-function mutations.⁶² Any changes of these molecules will affect Na_v1.5 membrane availability and alter the phenotype of Na_v1.5 mutations.

Differential channel complex regulation

The Na_v1.5 α subunit interacts with accessory proteins (β subunits and interacting proteins, reviewed in ref⁶³) to form a macromolecular complex (Figure 5). Mutations of these accessory proteins have been reported to cause BrS. For example, BrS patients with mutations in the Na⁺ channel β1 and β3 subunit show impaired Na_v1.5 channel function.^{12, 64, 65} The β4 subunit expression level is reported to affect the severity of CCD in mice by modulating Na_v1.5 channel activation.⁶⁶ Therefore, differential expression of accessory channel subunits may modulate the phenotype of any given Na_v1.5 mutation.

Figure 5.

The Na_v1.5 channel is part of a macromolecular complex. The channel α subunit interacts with multiple cellular proteins, including accessory proteins such as β subunits, caveolin 3 (CAV3), Ankyrin (Ank), MOG1, syntrophin, and cytoskeleton at the cell membrane, forming a macromolecular complex. Na_v1.5 channel activity can be modified by the altered expression or function of the components of the macromolecular protein complex.

Na_v1.5 interacting proteins such as ankyrin (Ank), MOG1, and syntrophin interact with the α and β subunits of Na⁺ channel. Mutations and misregulation of these proteins can disturb Na_v1.5 membrane expression and function. For example, a mutation in Ank-B causes LQT syndrome type 4, resulting from disruption of Ank-B interactions with Na_v1.5, the Na⁺-Ca²⁺ exchanger, and inositol-1,4,5-trisphosphate receptors.⁶⁷ MOG1 mutations are found in patients with BrS⁶⁸ (E83D) and AF⁶⁹ (E61X). The A257G syntrophin mutation induces LQT3 because of a gain-of-function modulation of Na_v1.5 gating properties.⁷⁰ Therefore, differential regulation of the macromolecular complex is another potential modifier of the Na_v1.5 mutation-induced phenotypes.

Post-translational Modifiers

PKC and PKA

The most well-known post-translational modifiers of cardiac Na_v1.5 are PKA and PKC that can modify the Na_v1.5 channel membrane expression and gating properties. Summated effects of these modifiers with the baseline genetic defect may explain phenotypes that vary with time, patient state, and region of the heart.

Protein kinases regulate Na_v1.5 in fast and complex ways. For example, PKA and PKC regulate the Na_v1.5 by affecting channel conductance, trafficking and membrane expression.^{1, 53, 54, 71} PKCs are a family of serine/threonine protein kinases with at least 15 isoforms in humans, with the PKCα, δ, and ε isoforms being most intensively investigated in cardiomyocytes.⁷² PKC decreases the cardiac I_Na by channel redistribution away from plasma membrane.⁵⁴ In addition, PKCδ reduces I_Na by inducing mitochondrial reactive oxygen species (ROS) overproduction and decreasing the single channel conductance.^{1, 61, 73, 74} This effect occurs in minutes without altering SCN5A mRNA abundance and channel membrane expression, suggesting these effects are in addition to any effect on channel trafficking. Therefore, PKC activation generally reduces I_Na and could exacerbate the phenotype of loss-of-function SCN5A mutations or mitigate the phenotype of gain-of-function SCN5A mutations. PKCε has not been found to regulate Na_v1.5 directly, but activation of PKCε has been reported to exhibit beneficial antiarrhythmic effect in an PKCε agonist transgenic mouse model with ischemia/reperfusion.⁷⁵ This indicates that the state of PKCε activation may modulate the phenotypic variance of Na_v1.5 mutations.

PKA is a tetrameric serine/threonine kinase with broad specificity that is known to upregulate Na_v1.5 through changing the channel trafficking⁵³ and conductance.⁶¹ Important PKA phosphorylation sites on human cardiac Na_v1.5 at S525/528 and RRR533-535 have been identified.^{53, 76, 77} RRR533-535 is important for the PKA-dependent increases in I_Na.⁶¹ PKA can also upregulate Na_v1.5 through minimizing mitochondrial ROS production.^{1, 73} Furthermore, PKA activation of connexin 43 stimulates gap junctional intercellular communication and alters the electrical distance, over which Na_v1.5 mutations have their effects.^78–80

ROS

Major sources of cardiac ROS include mitochondrial electron transport chain, uncoupled nitric oxide synthase, NAD(P)H oxidase, cyclooxygenase, and xanthine oxidase. Enzymes such as superoxide dismutase, glutathione peroxidase, and catalase form the defensive system. ROS elevation is a common feature shared by various cardiac diseases.^{81, 82} Increasing evidence suggests that elevated ROS alter the cardiac ion channels (such as Na_v1.5, K_v1.4, K_v4.2, K_v4.3, and Ca_v1.2), Ca²⁺ handling proteins, and gap junction.⁸³ ROS can downregulate Na_v1.5 at the transcriptional level by reducing the SCN5A gene expression^{33, 38} and at the post-translational level by affecting channel phosphorylation and the channel conductance.⁶¹ For instance, mitochondrial ROS can modify Na_v1.5 directly through altering channel post-translational modifications known to decrease I_Na.^{1, 73} Mitochondrial ROS can also activate PKC, which can further downregulate channel activity by phosphorylation⁶¹ or impairing channel trafficking.⁸⁴ Therefore, fluctuations in ROS are another source of phenotypic variability of Na_v1.5 mutations.

Alterations in metabolism

HF and cardiac injury from many causes are associated with altered metabolism and downregulated Na_v1.5.^85–88 NADH is known to oscillate with myocardial ischemia, and mitochondrial injury is associated with increased NADH and ROS levels.^{89, 90} Elevated NADH results in PKC activation, mitochondrial ROS overproduction, and decreased I_Na. Many such alterations could be local, following ischemic segments, for example. NAD⁺ antagonizes the I_Na reduction seen with a rise of internal NADH by activating PKA, reducing mitochondrial ROS production, and enhancing I_Na. In an ex vivo study, NADH perfusion leads to programmed electrical stimulation (PES)-induced ventricular tachycardia (VT) in wide-type mouse hearts, while NAD⁺ perfusion diminishes PES-induced VT in SCN5A^+/− mouse hearts.¹ Therefore, any alteration in cardiac metabolism could alter the phenotype of Na_v1.5 mutations.

Degradation Pathways of Na_v1.5

The last step of the Na_v1.5 channel life cycle is degradation. A balance between synthesis and degradation of the channel determines the available channel amount to generate currents. Ubiquitination of membrane proteins by ubiquitin-protein ligases is one common signal for protein internalization and degradation. Neural precursor cell-expressed, developmentally downregulated isoform 4-2 (Nedd4-2) is an ubiquitin-protein ligase to regulate the degradation of cardiac Na_v1.5, K_vLQT1, and Ca_v1.2 channels. In Xenopus oocytes, Nedd4-2 significantly reduces these channels’ membrane expression levels, and attenuates ion currents to 40–75%.^{49, 91–94} An inactive Nedd4 mutant can increase the Na_v1.5 density on the membrane and elevate I_Na.⁹¹ Nedd4 also regulates the degradation of neuronal Na⁺ channels (Na_v1.2, Na_v1.3, Na_v1.7, and Na_v1.8), which are expressed in cardiomyocytes and affect the cardiac function (as discussed below).^93–95 Nedd4 can be regulated by PKA, MAPKs, serum and glucocorticoid-induced kinase, and 14-3-3.^{49–51, 96, 97} Therefore, any changes of Nedd4 and its modifiers could affect the membrane expression of cardiac ion channels, which will give rise to various phenotypes when combined with Na_v1.5 mutations.

Extrinsic Modifiers

Other Na⁺ Channel Isoforms

Besides cardiac Na_v1.5, the neuronal and skeletal isoforms of voltage-gated Na⁺ channels (Na_v1.1 – Na_v1.8) are expressed in the heart as well. In adult mouse heart, the Na_v1.5 is responsible for 92% of I_Na, while the other isoforms account for 8%.⁹⁸ Na_v1.1 and Na_v1.6 participate in pacemaking of the SA node⁹⁹ and in EC coupling.¹⁰⁰ Na_v1.6 is also shown as a depolarizing reserve to assure excitation.¹⁰⁰ Recently, mutations in Na_v1.8 encoded by SCN10A have been associated with prolonged cardiac conduction and a higher risk of heart block.¹⁰¹ Na_v1.8 is mainly expressed in intracardiac neurons of the heart and plays an important role in regulating the AP firing frequency in mice.¹⁰² This channel also contributes to I_NaL and its inhibition may be antiarrhythmic.¹⁰³ Association of common variants at SCN10A, SCN5A and a transcriptional regulator gene HEY2 (encoding hairy/enhancer-of-split related with YRPW motif protein 2) is reported in a genome-wide association study (GWAS) of BrS patients.¹⁰⁴ The accumulated effect of the three loci significantly decreased cardiac conduction velocity and increased susceptibility to BrS. Differential regulation of these neuronal and skeletal isoforms of Na⁺ channels could modulate the phenotype of any Na_v1.5 mutation.

Other Cardiac Ion Channels

The fact that other channel gene mutations are associated with BrS and LQT3 suggests that many different channels can be involved in generating one clinical phenotype. This idea implies that changes in these other ion channels can affect the phenotype of any Na_v1.5 mutation. For example, SCN5A mutation D1275N has been shown to cause atrial standstill if this mutation is combined with polymorphisms in the atrial-specific gap junction protein connexin 40.¹⁰⁵ The K⁺ current I_to conducted by K_v4.2 and K_v4.3 and I_K1 conducted by Kir2.1 have shown modifications on I_Na and affect the APD and conduction velocity. I_to counteracts the depolarizing currents during the early phase of the AP plateau, resulting in an AP notch (phase 1) and a “spike-and-dome” morphology. In a canine heart model, Hoogendijk et. al.¹⁰⁶ have found that increased I_to at the early phase of AP can potentially cause conduction slowing. When combined with I_Na reduction, I_to that is larger in subepicardial than in subendocardial myocytes can lead to loss of the AP dome and early completion of repolarization in the right ventricular subepicardium but not the subendocardium. This can result in ST-segment elevation of the BrS ECG.¹⁰⁶ Therefore, alterations or mutations of K_v4.3 and its accessory proteins could amplify EADs or mitigate early repolarization in BrS.

I_K1 controls the resting membrane potential that determines the voltage-gated Na_v1.5 availability. A recent study shows that I_K1-I_Na interactions involve a reciprocal modulation of their respective channel expressions within a macromolecular complex.¹⁰⁷ An increase of I_K1 can lead to membrane hyperpolarization, which can either amplify I_Na reduction induced by loss-of-function SCN5A mutations or minimize an I_NaL increase induced by a gain-of-function mutation.

Genetic and Genomic Background

Genetic background may contribute to the variable phenotypes seen with SCN5A mutations. One recent study in families with BrS carrying SCN5A mutations suggest that the presence of Brugada phenotype is not only determined by SCN5A mutation, but also by the genetic background of the subject.¹⁰⁸ The SCN5A^1798insD/+ mutation has been found to generate a more severe conduction slowing phenotype in mouse with 129P2 background than those with FVB/N background.⁶⁶ A recent GWAS in Brugada patients has identified that polymorphisms in SCN10A and HEY2 can have a strong impact on the phenotype of BrS.¹⁰⁴ The fact that multiple genetic mutations in unrelated genes act to lower I_Na and cause BrS implies the importance of considering genetic and genomic modifiers when undertaking genotype-phenotype correlations.^{64, 65, 109, 110}

Summary

Mutations in the cardiac Na_v1.5 result in a number of clinical syndromes. Some of this phenotypic variability can be explained by the effect of the mutation on channel properties. On the other hand, it is difficult to explain manifestation in a region or chamber of the heart, circadian variation and age-dependent arrhythmic episodes, incomplete penetrance in families, and variations with clinical state within an affected individual. We propose that this variability may be the result of secondary modifiers of the Na_v1.5 channel. We have divided these modifiers into those within the SCN5A gene, those intrinsic to the life cycle of Na_v1.5, and those extrinsic to Na_v1.5 that can modify overall electrical function. It is important, however, to acknowledge that much of the evidence for Na_v1.5 expression/functional modifiers has been derived from the heart failure literature. More experimental evidence will be required to support and generalize this “second hit” hypothesis of the phenotypic variability of SCN5A mutation-related arrhythmia syndromes.

It is conceivable that Na_v1.5 channel modifiers can alleviate or aggravate the phenotype of a SCN5A mutation, depending on their impact on sodium current or the overall balance of currents. Localized modifier changes may explain regional expression of mutation effects. Changes in modifiers could explain why some conduction disease phenotypes progress and others do not. Variations in modifiers could explain why ECG phenotypes and arrhythmic risk varies with time of day and with age. The above considerations may explain why there is less phenotypic variability in gain-of-function mutations since there are less modifiers that affect inactivation alone than the number that can affect overall channel current. Nevertheless, it is possible to see how the net effect of a gain-of-function mutation is modulated by the total available channels, allowing for variable penetrance of the phenotype with time or between regions of the heart.

In summary, considering these modifiers may improve genotype-phenotype correlation and risk prediction algorithms. Moreover, targeting modifiers may prove an effective strategy to prevent Na_v1.5 mutation-associated diseases.

Key points:

• Multiple clinical syndromes are observed among subjects carrying cardiac Na_v1.5 mutations,

• Variability in clinical presentation is only partially explained by mutation-associated changes in channel biophysical properties.

• Phenotypic variability not easily explained includes manifestations localized to a particular cardiac chamber or region, variable penetrance within families, and phenotypic changes within an individual as a result of such factors as age, gender, time of day, and body temperature.

• Additional modifiers of the cardiac Na_v1.5 may help explain the phenotypic variability.

• These additional modifiers may include other mutations in the SCN5A gene, intrinsic modifiers of the Na_v1.5, and extrinsic modifiers affecting the overall balance of channel currents.

• Considering both the mutation-induced Na_v1.5 changes and secondary modifiers may help improve the genotype-phenotype correlations and direct therapeutic strategies.

Review Criteria.

PubMed was used to search for original papers studying various phenotypes of cardiac sodium channel mutations and channel modifiers. The search terms were “SCN5A mutations,” “arrhythmia,” and “sodium channel regulation & modification.” Papers used were written in English. In addition, we searched the references of the identified papers to discover related articles not included in the initial search.

Author biographies

Dr. Man Liu is currently a research assistant professor of Cardiovascular Institute of Lifespan and Brown University. Her research focuses on the mechanisms of metabolic regulation of cardiac Na⁺ channels and arrhythmias. Dr. Liu obtained her PhD degree in Biophysics from Johann Wolfgang Goethe-Universität (Germany). She finished her postdoctoral training in Stockholm University (Sweden) and State University of New York at Stony Brook. Dr. Liu became a research assistant professor in University of Illinois at Chicago in 2010.

Dr. Kai-Chien Yang is currently a post-doctoral research associate at the Cardiovascular Institute in Lifespan and Brown University. He obtained his PhD degree in Molecular Genetics and Genomics from Washington University in St Louis and MD degree from National Taiwan University. His research project is funded by the American Heart Association and focuses on the roles of oxidative stress in gap junction remodeling and cardiac arrhythmias.

Dr. Samuel Dudley is the Ruth and Paul Levinger Professor and Chief of Cardiology at the Warren Alpert Medical School of Brown University and Director of the Cardiovascular Institute at Lifespan. He holds doctoral degrees in Medicine and in Physiology from the Medical College of Virginia. He is a member of the Association of University Cardiologists and the American Society for Clinical Investigation. His research program is funded by multiple sources including the National Institutes of Health and the Department of Veterans Affairs and focuses on mechanisms and treatments of arrhythmia and heart failure with preserved left ventricular function.