Clinical, Genetic, and Biophysical Characterization of SCN5A Mutations Associated With Atrioventricular Conduction Block
Background— Three distinct cardiac arrhythmia disorders, the long-QT syndrome, Brugada syndrome, and conduction system disease, have been associated with heterozygous mutations in the cardiac voltage-gated sodium channel α-subunit gene (SCN5A). We present clinical, genetic, and biophysical features of 2 new SCN5A mutations that result in atrioventricular (AV) conduction block.
Methods and Results— SCN5A was used as a candidate gene in 2 children with AV block. Molecular genetic studies revealed G to A transition mutations that resulted in the substitution of serine for glycine (G298S) in the domain I S5-S6 loop and asparagine for aspartic acid (D1595N) within the S3 segment of domain IV. The functional consequences of G298S and D1595N were assessed by whole-cell patch clamp recording of recombinant mutant channels coexpressed with the β1 subunit in a cultured cell line (tsA201). Both mutations impair fast inactivation but do not exhibit sustained non-inactivating currents. The mutations also reduce sodium current density and enhance slower inactivation components. Action potential simulations predict that this combination of biophysical abnormalities will significantly slow myocardial conduction velocity.
Conclusions— A distinct pattern of biophysical abnormalities not previously observed for any other SCN5A mutant have been recognized in association with AV block. These data provide insight into the distinct clinical phenotypes resulting from mutation of a single ion channel.
Received July 24, 2001; revision received November 7, 2001; accepted November 8, 2001.
Three distinct inherited cardiac arrhythmia disorders, long-QT syndrome, Brugada syndrome, and conduction system disease, have been associated with heterozygous mutations in the cardiac voltage-gated sodium channel α-subunit gene (SCN5A).1–3⇓⇓ Mutations resulting in long-QT syndrome cause persistent sodium current (INa) and delay repolarization, thereby creating a predisposition toward the distinctive polymorphic ventricular tachycardia, torsades de pointes.4–7⇓⇓⇓ Brugada syndrome exhibits a characteristic electrocardiographic pattern consisting of ST elevation in the right precordial leads and apparent right bundle-branch block; associated SCN5A mutations reduce INa, alter transmural myocardial voltage gradients, and increase the risk for ventricular fibrillation.2,7,8⇓⇓ Most recently, cardiac conduction system disease, manifesting as slowed intramyocardial conduction and in some cases progressive atrioventricular (AV) block, has been linked to SCN5A mutation.3
Familial AV conduction block, characterized by progressive “degree of block” in association with variable apparent “site of block,” may be transmitted as an autosomal dominant trait. Two genetically distinct forms of AV conduction block have been identified.9 Brink et al10 established a genetic link between AV block and a genetic locus at chromosome 19q13. Schott et al3 mapped AV block to chromosome 3p21, where the cardiac sodium channel, SCN5A, is encoded and identified 2 SCN5A mutations. Recently, Tan et al11 characterized an SCN5A mutation (G514C) that resulted in an isolated cardiac conduction defect. In contrast to SCN5A mutants resulting in long-QT or Brugada syndrome, G514C exhibited opposing gating effects, including a depolarizing shift in the voltage-dependence of activation and enhanced fast inactivation, that are predicted to result in isolated conduction slowing.
We report clinical, genetic, and biophysical characterizations of 2 new SCN5A mutations associated with AV block presenting during childhood. The mutants reveal a distinct pattern of biophysical abnormalities not previously observed for any other SCN5A mutants.
Informed consent was obtained from study participants in accordance with the Medical University of South Carolina Institutional Review Board for Human Research. During electrocardiographic assessment of irregular heartbeat, second-degree AV block with a normal corrected QT interval was diagnosed in the probands CHB2 and CHB17 at ages 6 and 9 years, respectively. Other family members were evaluated (Figures 1 and 2⇓).
Molecular Genetic Methods
Genomic DNA was isolated from the blood of study participants and the polymerase chain reaction was used to amplify coding region and flanking intronic sequence of SCN5A as previously described.12 Sequencing reactions were performed in the presence of fluorescent-labeled dideoxynucleotides and additional primer for exon specific sequencing in both sense and antisense direction on isolated polymerase chain reaction product.
Sodium currents were recorded using the whole-cell patch clamp technique.8 The holding potential was −120 mV for all experiments, and details of each pulse protocol are given schematically in the figures and explained in Results. Data were analyzed using pCLAMP 8.0 (Axon Instruments, Inc), and figures were prepared with SigmaPlot 2000 (SPSS Inc).
Action Potential Simulations
Simulations were conducted using the theoretical dynamic model of a mammalian ventricular action potential (Luo-Rudy [LRd] model).13 The transient potassium current, ITO, was introduced into the LRd model as previously described.14 A 1-dimensional model fiber composed of 130 serially arranged ventricular cells each of LRd formulation was used.15 Action potential simulations were performed by stimulating the fiber 10 times (to achieve steady state) at a constant cycle length (300 ms or 800 ms).
The sodium channel in the LRd model incorporates 2 inactivation gates, h and j, with fast and slow kinetic behavior, respectively.13 To examine the overall effect of increased slow inactivation on conduction velocity in D1595N, the contribution (amplitude) of the j gate in the LRd model was increased to reflect the experimental result at 32°C (Figure 2C), as follows:
where Vm is the membrane potential.
The other j∞ parameters were not altered from the LRd (guinea pig myocyte) formalism, given the relative uncertainty of these parameters in the human heart at 37°C. Slowing of fast inactivation was introduced into the model by increasing the time constant of the h gate 1.5-fold (Figure 3B). The maximum conductance of D1595N channels was also reduced to reflect their lower expression (Figure 3E). The D1595N simulations reflect the heterozygous condition, assuming equal penetration from a 50/50 mixture of mutant and wild-type channels. The model time constants for fast and slow inactivation for both mutant and wild-type were obtained from experiments at 22°C, where clamp control was optimal, and then adjusted using Q10 factors, 1.4 for wild-type and 2.3 for D1595N. The Q10 factors were determined from measured rates of fast inactivation (Figure 3A) at both 22°C and 32°C using the following relation:
where A represents any parameter measured at temperature T (t1=22°C and t2=32°C).
Second-degree heart block was diagnosed in 2 children during electrocardiographic evaluation of irregular heart rate. CHB17 belongs to a 3-generation white kindred with inherited cardiac conduction disease (Figure 1E). The asymptomatic proband (III-1) was diagnosed at age 9 years. Before elective pacemaker implantation at age 12 years, the PR interval was prolonged (≈280 ms) during conducted beats, and a prolonged HV interval of 210 ms and normal AH interval of 70 ms were recorded; conduction block occurred in the HV interval (not shown). Complete AV block was present at age 20 years (Figure 1C). The father (II-2) and sister (III-2) had ECG evidence of right bundle-branch block, left axis deviation, and a normal PR interval (Figure 1A and 1D). The corrected QT interval was normal (<420 ms) in individuals II-2, III-1, and III-3. Family history was negative for sudden death. Individual I-2, who lived to 85 years of age, was described as having had a mild “heart attack” in his 50s; medical records providing details of this event were not available.
To determine heart rate dependence of electrocardiographic abnormalities, we evaluated 2 simultaneously recorded ECG leads during 24-hour ambulatory ECG monitoring of individual II-2. Examination was performed during a variety of conditions, including sleep and jogging and resulting in cycle lengths ranging from 1000 to 340 ms. There was no evidence of second- or third-degree AV block or ST-T wave elevation, and QRS duration showed no rate dependent trends (130 to 140 ms, data not shown).
The family of the second proband, CHB2, emigrated from Mexico. Second-degree AV block, diagnosed at age 6 years, later progressed to third-degree AV block (Figure 4A). His mother does not have electrocardiographic evidence of AV conduction disturbance and his father declined participation. Family history is negative for sudden death.
Molecular Genetic Findings
During systematic survey of all SCN5A coding exons, 2 heterozygous nucleotide changes (G892A and G4783A) were identified in exons 7 and 27, respectively. The nucleotide changes altered the coding sense from aspartic acid to asparagine (D1595N, Figure 1F) in CHB17 and glycine to serine (G298S, Figure 4C) in CHB2. The sequence changes create an AluI restriction enzyme site (G892A) or abolish a Taq I restriction enzyme site (G4783A) that allowed independent confirmation of the G to A transitions (data not shown). All 3 members of CHB17’s family with an ECG phenotype exhibited the D1595N mutation. The mother of CHB2 had a normal ECG and did not carry G298S. Neither sequence change was found in unaffected family members or 100 normal unrelated chromosomes.
Figure 1G depicts amino acid alignment for a portion of SCN5A domain IV S3 segment including aspartic acid 1595, which is conserved in all sodium channel genes. Glycine is conserved at position 298 in the rat cardiac sodium channel but not in other mammalian isoforms. Figure 4D depicts an SCN5A cartoon showing location of G298S, D1595N, and previously identified SCN5A mutations resulting in conduction system disease.
Biophysical Characterization of G298S and D1595N
We engineered G298S and D1595N into the recombinant human heart sodium channel (hH1) for transient expression in a cultured mammalian cell line (tsA201) and functionally characterized each using whole-cell patch clamp recording. All electrophysiological experiments were performed in cells co-transfected with the human sodium channel β1 subunit (hβ1).
G298S and D1595N Impair Fast Inactivation
In Figure 3A, direct comparisons of sodium currents recorded from cells expressing either WT-hH1, G298S, or D1595N illustrate that the mutants exhibit a less rapid decay in whole-cell current elicited by a 20 ms test potential. The time to 50% current decay at 22°C was 0.54±0.03 ms for WT-hH1, 0.71±0.02 ms for G298S, and 1.20±0.06 for D1595N (n=17 for each, P<0.001). A similar difference in the time course of fast inactivation was observed at 32°C (time to 50% current decay was 0.23±0.01 for WT-hH1, 0.29±0.01 ms for G298S, and 0.42±0.01 for D1595N; n=5 for each, P<0.005). We further quantified the time course of fast inactivation by fitting the data with double exponential functions, and the voltage-dependence of time constants (τ1, τ2) for fast inactivation are illustrated in Figure 3B. These data indicate that the mutants exhibit a significant defect in fast inactivation that manifest over a wide range of membrane potentials. Neither G298S nor D1595N exhibited a sustained, non-inactivating current component during longer (200 ms) test depolarizations (data not shown).
The mutations also affect sodium channel availability and recovery from fast inactivation. Channel availability for both G298S and D1595N is significantly shifted toward more depolarized potentials (Figure 3C; WT-hH1, V1/2= −100.7±1.4 mV, n=19; G298S, V1/2=−93.3±0.5 mV, n=7; D1595N, V1/2=−96.5±1.5 mV; P<0.05 as compared with WT). Mutant channels also have delayed recovery from fast inactivation (Figure 3D). Over a wide range of test potentials, the mutant channels expressed a lower level of current density in transfected cells than wild-type channels (Figure 3E). In contrast with an earlier report regarding another SCN5A mutation linked to inherited cardiac conduction system disease,11 G298S and D1595N have no impact on the voltage-dependence of activation (Figure 3F).
G298S and D1595N Affect Slower Forms of Inactivation
We tested cells expressing either WT-hH1 or mutant channels for differences in channel availability when preconditioned with longer (≥1000 ms) membrane depolarizations. In these experiments, effects of fast inactivation were eliminated by a brief recovery pulse to −120 mV before channel availability was assessed with a −10 mV test potential. Figure 2A illustrates that G298S and D1595N channels more readily inactivate and exhibit slowed recovery from inactivation induced by a sustained depolarization (Figure 2B). In WT and mutant channels, a 1000 ms depolarization was sufficient to induced an intermediate component of recovery from inactivation that was slower in both mutants (reflected in the τ2 time constants; see Figure 2B legend). Additionally, a greater proportion of mutant channels partition into this slower inactivated state during the 1000 ms prepulse (reflected in the larger τ2 amplitudes; see Figure 2B legend).
We characterized the temporal onset of the slower inactivation components by the experiment illustrated in Figure 2C. With progressively longer prepulse durations, sodium current diminished more in cells expressing G298S or D1595N than those transfected with WT-hH1. In both mutant channels, time-dependent loss of availability was multi-exponential with a very rapid component (<2 ms), as well as intermediate (30 to 60 ms) and much slower components (>10 sec). The more rapid components were not easily distinguishable in the WT channel and the dominant kinetic behavior of the slowest component was smaller in amplitude relative to the mutants. Overall, the data indicate that both mutations cause the recruitment of slower inactivation components including both intermediate and slow kinetic components.
Physiological Consequences of SCN5A Dysfunction
We used computer simulations of the mammalian cardiac action potential utilizing the LRd model13 to determine if the functional gating defects associated with these 2 SCN5A mutations could explain the observed conduction abnormality. We stimulated the model fiber at a constant cycle length of either 800 ms or 300 ms and computed the maximum upstroke velocity (dV/dtmax in V/sec) for action potentials simulated using wild-type sodium channel kinetics or channel properties modified to match those observed for D1595N (see Methods). For cycle lengths of 800 ms, the model predicts a substantial decrease (wild-type, 264 V/sec versus D1595N, 174 V/sec; 34% decrease) in the upstroke velocity for the heterozygous condition consistent with conduction slowing in mutation carriers. At faster cycle lengths (300 ms), dV/dTmax was reduced further and exhibited a bigeminal pattern of alternating values between 161 and 176 V/sec. This observation is consistent with the lack of effect of heart rate on the degree of block observed in the CHB17 family. These simulations provide additional evidence that the functional abnormalities observed for D1595N are consistent with a defect in cardiac conduction and based on the similarity of their functional gating defects for G298S by inference.
Mutations in SCN5A have been associated with 3 distinct arrhythmia syndromes. In this report, we present clinical, genetic, and biophysical characterizations of 2 new alleles associated with AV block. The clinical, molecular genetic, and biophysical characterizations of these diseases are providing refined genotype-phenotype characterizations that lead to improved understanding of the pathophysiology of specific SCN5A mutations. Such distinctions may ultimately provide the basis for improved diagnosis and treatment.
Functional characterization of G298S and D1595N revealed a distinct pattern of abnormalities not previously observed for any other single SCN5A allele. Both of the mutations we report here are associated with reduced levels of sodium current density in transfected cells, and we speculate that a similar reduction may occur in vivo. In addition, the mutations cause disturbances in multiple forms of sodium channel inactivation that differ based on their time course and molecular mechanisms. Fast sodium channel inactivation occurs with a time course of a few milliseconds and is the type of channel gating originally described by Hodgkin and Huxley16 involving occlusion of the inner channel mouth by a cytoplasmic gate. Sodium channels also exhibit slower forms of inactivation that proceed over several seconds to minutes and are physiologically important for setting the level of sodium current density in excitable membranes.
Our studies demonstrate a significant effect of G298S and D1595N on the kinetics of fast inactivation distinct from SCN5A mutations associated with long-QT syndrome4–6⇓⇓ and the SCN5A mutant G514C that is associated with cardiac conduction system disease.11 Another important biophysical disturbance exhibited by G298S and D1595N is enhancement of slower forms of inactivation, which is expected to cause an activity-dependent loss of sodium channel availability, resulting in reduced sodium current density.17 Reduced myocardial sodium current density will slow the rise time of the cardiac action potential and slow conduction velocity. Therefore, the combination of reduced sodium current density and enhanced slow inactivation provides a plausible mechanism for the observed conduction disease. This mechanism is well supported by data from action potential simulations that predict slowing of myocardial conduction in the presence of the functional anomalies observed for D1595N sodium channels.
This work was supported in part by a grant from the American Heart Association Southeast Affiliate to Dr Wang and by grants from the National Institutes of Health (HD39946 to Dr Benson, NS32387 to Dr George, HL46681 to Drs George and Balser, and GM56307 to Dr Balser). Drs George and Balser are established investigators of the American Heart Association. The authors thank the family members for their participation. The studies would not have been possible without the technical assistance of Ping Lu and Nila Gillani, and the assistance of Maggie Fischer, Peter Karpawich, MD, and Brian Olshansky, MD, in collecting patient material.
- ↵Dumaine R, Wang Q, Keating MT, et al. Multiple mechanisms of Na+ channel-linked long-QT syndrome. Circ Res. 1996; 78: 916–924.
- ↵Wang DW, Yazawa K, George AL Jr, et al. Characterization of human cardiac Na+ channel mutations in the congenital long QT syndrome. Proc Nat Acad Sci. 1996; 93: 13200–13205.
- ↵Wang DW, Makita N, Kitabatake A, et al. Enhanced Na+ channel intermediate inactivation in Brugada syndrome. Circ Res. 2000; 87: E37–E43.
- ↵McKusick-Nathans Institute for Genetic Medicine, Johns Hopkins University (Baltimore, MD), and National Center for Biotechnology Information, National Library of Medicine (Bethesda, MD). Online Mendelian Inheritance in Man, OMIM (TM). Available at http://www.ncbi.nlm.nih.gov/omim/. Accessed December 5, 2001.
- ↵Brink PA, Ferreira A, Moolman JC, et al. Gene for progressive familial heart block type I maps to chromosome 19q13. Circulation. 1995; 39: 1633–1640.
- ↵Luo CH, Rudy Y. A dynamic model of the cardiac ventricular action potential. I. Simulations of ionic currents and concentration changes. Circ Res. 1994; 74: 1071–1096.
- ↵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.
- ↵Viswanathan PC, Shaw RM, Rudy Y. Effects of IKr and IKs heterogeneity on action potential duration and its rate dependence: a simulation study. Circulation. 1999; 99: 2466–2474.
- ↵Hodgkin AL, Huxley AF. The dual effect of membrane potential on sodium conductance in the giant axon of Loligo. J Physiol (Lond). 1952; 116: 497–506.
- ↵Veldkamp MW, Viswanathan PC, Bezzina C, et al. Two distinct congenital arrhythmias evoked by a multidysfunctional Na+ channel. Circ Res. 2000; 86: E91–E97.