Abstract
Background: Brugada syndrome (BrS) is a genetically determined cardiac electrical disorder, characterized by typical electrocardiography (ECG) alterations, and it is an arrhythmogenic syndrome that may lead to sudden cardiac death. The most common genotype found among BrS patients is caused by mutations in the SCN5A gene, which lead to a loss of function of the cardiac sodium (Na+) channel (Nav1.5) by different mechanisms. Methods: The assay of confocal laser microscopy and western blot were used to identify the expression and location of L812Q at the cell surface. Characterization of Nav1.5 L812Q mutant Na+ channels was text by patch-clamp recordings, and the PHYRE2 server was used to build a model for human Nav1.5 channel. Results: Here, we report that a novel missense SCN5A mutation, L812Q, localized in the DII-S4 transmembrane region of the Nav1.5 channel protein, was identified in an index patient who showed a typical BrS type-1 ECG phenotype. The mutation was absent in the patient's parents and brother. Heterologous expression of the wild-type (WT) and L812Q mutant Nav1.5 channels in human embryonic kidney cells (HEK293 cells) reveals that the mutation results in a reduction of Na+ current density as well as ∼20 mV hyperpolarizing shift of the voltage dependence of inactivation. The voltage dependence of activation and the time course for recovery from inactivation are not affected by the mutation. The hyperpolarizing shift of the voltage dependence of inactivation caused a reduction of the Na+ window current as well. In addition, western blot and confocal laser microscopy imaging experiments showed that the mutation causes fewer channel to be expressed at the membrane than WT channel. A large proportion of the mutant channels are retained in the cytoplasm, probably in the endoplasmic reticulum. Conclusion: The decrease of channel expression, hyperpolarizing shift of voltage dependence of inactivation, and a decline of Na+ window current caused by L812Q mutation lead to a reduction of Na+ current during the upstroke and the repolarization phases of cardiac action potential, which contribute to the development of BrS.
Introduction
Brugada syndrome (BrS) is an autosomal-dominantly inherited cardiac arrhythmogenic syndrome. BrS is characterized by an electrocardiographic (ECG) phenotype consisting of ST-segment elevation in the right precordial leads V1-V3, often referred to as a type-1 BrS pattern, with atypical right bundle branch block; it predisposes to a highly increased risk for syncope and sudden cardiac death (SCD) as a result of polymorphic ventricular tachycardia (VT) or ventricular fibrillation (VF) in the absence of structural heart disease [1,2,3]. BrS is thought to account for 4% of all SCDs and for up to 20% of unexplained sudden death in patients without cardiac structural disease [4]. However, some patients display a more benign course. The diagnosis of BrS requires the presence of type-1 Brugada ECG phenotype in the right precordial leads (i.e.V1-V3), either spontaneously manifesting or unmasked by sodium channel blockers [5]. Right bundle branch block may be associated with BrS, but it is not a requirement for the diagnosis [1,6,7].
Voltage-gated sodium (Na+) channel Nav1.5 plays a very important role in cardiac impulse propagation. The activation of Na+ channels is responsible for the generation of the rapid upstroke of the cardiac action potential [8]. Each Na+ channel is composed of a pore-forming α-subunit and one or more modulating β-subunits [9,10]. The SCN5A gene encodes the α-subunit of Nav1.5 channel, a 2016-amino acid protein. The Nav1.5 α subunit consists of four homologous domains (DI-DIV), each with six transmembrane α helices (S1-S6) linked by intracellular loops [9,11]. The repeated S1-S4 domains form the channel's voltage sensing domains, primarily mediated by positively charged arginines and lysines positioned at every third residue within each S4 segment [12]. S5, S6 and S5-S6 loops (P-loops) form the functional pore and selectivity filter of Na+ channel [13]. Mutations in the SCN5A gene comprise the most common genotypes found among BrS patients (≈20% of Brugada syndrome patients). Over 300 mutations in the SCN5A gene have been identified [3,14,15,16]. BrS-associated SCN5A mutations cause variable reductions of inward Na+ current by alterations of channel gating such as delayed activation, enhanced inactivation, slowed recovery from inactivation [17], and/or by impaired trafficking of the channel, the latter reducing expression at cell membrane [18,19,20]. All of these mechanisms lead to a loss of function of the cardiac Na+ channel. Decreased inward Na+ current can affect depolarization and/or repolarization of the cardiac action potential. However, the pathophysiologic mechanism underlying BrS phenotype is still under debate [1].
In the present study, we have screened the SCN5A gene from a patient who displayed a characteristic type-1 BrS pattern in ECG. A novel mutation, L812Q, located in the S4 segment of domain II (DII) of the Nav1.5 α-subunit protein has been identified. The functional consequences of L812Q mutation have been examined by measuring the current of the Nav1.5 L812Q mutant channel in a mammalian-cell expression system. We demonstrate that the mutation enhances the Na+ channel inactivation process and impairs the channel`s expression at the membrane, the probable cause of BrS in the patient.
Materials and Methods
Patient
This study was performed in accordance with the protocol approved by the local ethics committees. Informed consent was obtained from the patient, who was initially presented to the emergency room and had an ECG examination. A blood sample was obtained from the patient for genotype analysis. After the patient's SCN5A mutation was identified, other family members of the patient were invited to participate for the investigation of detailed disease history, ECG examination and blood sample testing.
Genetic analysis
Genomic DNA was purified from peripheral blood lymphocytes using the TiANampBlood DNA kit (TianGen). All exons of SCN5A were amplified by polymerase chain reaction (PCR) and sequenced by Bei Jing AuGCT Biotechnology (Sanger sequencing method).
Plasmid constructions
The pEGFP-C1-hH1 plasmid was a gift of Dr. Thomas Zimmer. The cloning of the relevant fragment of SCN5A mutation (L812Q) was generated using overlap extension PCR and inserted as an EcoR1/ACC1 fragment into the EcoR1/ACC1 site of SCN5A cDNA of pEGFP-C1-hH1. All constructs were purified using Qiagen columns (Qiagen Inc). The cDNA was sequenced to confirm the presence of the correct fragment, containing the mutation.
Cell culture and transient expression in HEK293 cells
HEK293 cells were cultured in Dulbecco modified Eagle's medium (DMEM, PAA Laboratories GmbH, Pasching, Austria) supplemented with 10% fetal bovine serum (FBS, PAA Laboratories GmbH, Pasching, Austria) and 1% penicillin-streptomycin. Cells were maintained at 37°C in a humidified chamber with 95% air and 5% CO2. To express WT or L812Q mutant Nav1.5 channels, the cells were transiently transfected with plasmids of pEGFP-C1-hH1(WT, 4µg) or pEGFP-C1-hH1-L812Q(4µg) using Lipofectamine 2000 (8µl) transfection reagents (Invitrogen Co., Carlsbad, CA, USA) in 6 well plates. To co-express WT and L812Q mutant channel, equal amounts of pEGFP-C1-hH1 (2µg) and pEGFP-C1-hH1-L812Q (2µg) plasmids (0.5:0.5 ratio) were used to transfect HEK293 cells. The transfected cells were incubated at 37°C for 48 to 72 hours before performing patch-clamp, imaging and biochemical experiments.
Confocal laser microscopy assay
To determine subcellular localization of the WT, WT/ L812Q, and L812Q Nav1.5 channels in HEK293 cells, transfected HEK293 cells were placed on poly-D-lysine-coated cover slips. The cells were fixed with 4% formaldehyde for 20 minutes, and then incubated with 0.5% Triton X-100. After washed three times with PBS, the cells were incubated with a goat anti-mouse antibody (Millipore), and measured by confocal laser microscopy (Leica).
Immunofluorescence microscopy
To determine the effect of L812Q on location of Nav1.5, we also performed immunofluorescence staining using the Na/K-ATPase (Abcam), SCN5A (Abcam) and Calnexin (Abcam 1:100) anti-body. After 48h, the transfected L812Q cell lines were fixed with 4% formaldehyde for 20 minutes, then incubated with 0.5% Triton X-100. Rabbit/mouse anti- Na/K-ATPase/ SCN5A/ Calnexin antibody was used for immunofluorescence staining. After washed three times with PBS, the cells were incubated with a goat anti-rabbit/mouse antibody (Millipore) and measured by immunofluorescence microscopy.
Biotinylation of cell surface proteins
Cell surface proteins of HEK293 cells transfected with WT, WT/L812Q, L812Q were isolated using the Pierce Cell Surface Protein Isolation kit (Thermo Scientific) following manufacturer's instructions. Briefly, sulfo-NHS-biotin solution was prepared by dissolving one vial of sulfo-NHS-biotin into 48 ml of ice cold PBS. Then, cells at 90-95% confluence in a 10-cm dish were incubated with 10 ml of sulfo-NHS-biotin solution for 30 min at 4°C. After the reaction was quenched by 500 µl of quenching solution, cells were scraped and washed with TBS buffer to remove extra biotin solution. After centrifugation to remove supernatant, cell pellets were lysed in lysis buffer by sonication and clarified by centrifugation. Clear supernatant was reacted with immobilized NeutrAvidin gel slurry in columns (Pierce) to isolate surface proteins. Columns were washed and protein eluted in sample buffer containing DTT. Surface proteins were resolved on a SDS-PAGE gel and the samples were analyzed by western blot.
Western blots
Transfected HEK293cells were placed on six-well plates, total proteins were extracted using RIPA buffer, supplemented with protease inhibitor (invitrogen). Protein concentration was estimated by quantitative analyzer (GeneQuant pro RNA/DNA). Protein was then separated with an 8% to 10% SDS-PAGE (Invitrogen), transferred to a nitrocellulose membrane, then incubated with the SCN5A, Na/K-ATPase or β-ACTIN antibodies (Abcam). After washing three times with TBST, the membrane was incubated with a goat anti-mouse antibody (Bioworld). For the total proteins, relative protein expression was normalized to β-ACTIN, and for the transmembrane proteins, relative protein expression was then normalized to Na/K-ATPase levels.
Patch-clamp recordings
Whole-cell patch-clamp recording [21] was performed at room temperature (21-24°C) with an Axopatch 200B amplifier (Axon Instruments, Foster City, CA, USA). The extracellular solution contained (in mM): NaCl 140, CsCl 10, CaCl2 2, MgCl2 1, glucose 5, and HEPES 10, pH 7.4 with NaOH. Electrodes had resistances of 1 - 2 MΩ when filled with the filling solution containing (in mM): CsCl 110, NaCl 10, EGTA 11, MgCl2 1 and HEPES 10, pH 7.2 with CsOH. Measured series resistance was between 1 and 3 MΩ for all recordings, and was compensated by ≈80%. When the series resistance changed during the course of an experiment, data were discarded. The membrane potential was not corrected for a small liquid junction potential (∼ 6.7 mV). pClamp6 software (Axon Instruments, Foster City, CA, USA) was used for the experiments and data analysis. Nav1.5 current (INa) densities (pA/pF) were obtained by dividing the peak INa by the cell capacitance read from the Axopatch 200B amplifier. Na+ currents were elicited by 10 mV depolarizing steps between -80 to +60 mV from a holding potential of -120 mV. The pulse duration was 40 ms with an interval of 2 s. The plots of steady-state activation and inactivation were fitted by Boltzmann equation: y=1/ [1+exp(V-V0 5)/k], where V0.5 is the voltage at which sodium channels are half-maximally activated, and k was the slope factor. The recovery from inactivation at -120 mV was fitted by a bi-exponential equation: y=y0+Af[ 1-exp(-t/τf)]+As[1-exp(-t/τs)], where Af is the fraction of fast components, As is the fraction of slow components, τf is time constants of fast components of recovery and current Decay, τs is the time constants of slow components of recovery and current decay. To obtain the voltage-dependence of activation and inactivation, all data were recorded after establishing the whole-cell configuration for 5 minutes.
Homologous model for human Nav1.5 channel
PHYRE2 server [22] was used to build a model for human Nav1.5 channel, based on the structure of a bacterial voltage-gated sodium channel (NavAb) [23]. The four α subunits were modeled individually and superposed on the NavAb structure. Chimera [24] was used to minimize the tetrameric models for WT and mutant Nav1.5 channel.
Statistical analysis
Results are presented as means ± standard errors of the mean (SEM). Where indicated, a t-test was performed by using SPSS13.0 software (SPSS, Chicago, IL, USA). A p value less than 0.05 was considered statistically significant.
Results
Clinical studies
The patient, a 58-year-old man of Chinese ancestry, presented to the emergency room with the onset of chest distress and agonal respiration, following with vomiting, choking and violent pain on ventri- thighs of both lower extremities. The patient had a 15 year history of transit occurrence of paroxysmal chest pain and stuffiness. The ECG was characteristic of spontaneous type-1 BrS, displaying a prominent ST-segment elevation in the right precordial leads, V1-V2, associated with the presence of incomplete right bundle branch block (Fig. 1). The ECG showed that the patient remained in sinus rhythm, with a heart rate of about 83 beats a minute; prolongation of QT interval, entricular polymorphic tachycardia or fibrillation were not observed. The patient had never experienced syncope, amaurosis and disturbance of consciousness. No abnormal observation was found with an echocardiogram and chest X-ray. The patient had no concomitant risks such as diabetes, hypertension, or thromboembolic disease. The investigation of the patient's family history found no other case of BrS, sustained ventricular arrhythmias, or sudden death. The patient was diagnosed with BrS according to the type-1 BrS pattern in ECG and the clinical symptoms [25].
Electrocardiogram (ECG) recordings of the index patient in precordial leads V1-V6. Note the marked ST-segment elevation in the leads V1-V2. The patient remained in sinus rhythm.
Electrocardiogram (ECG) recordings of the index patient in precordial leads V1-V6. Note the marked ST-segment elevation in the leads V1-V2. The patient remained in sinus rhythm.
Mutational analysis
The SCN5A gene associated with BrS was screened in the present study. A novel heterozygous missense mutation in SCN5A gene was identified in the patient by PCR followed by DNA sequencing (Fig. 2A). There was a heterozygous T to A base change at position 2629 in exon 16, which presumably changed the coding from a leucine (L) (CTG) to a glutamine (Q) (CAG) at position 812, that was, from a hydrophobic amino acid of non-polarity to a hydrophilic amino acid of polarity. The L812Q mutation was located in the S4 segment of domain II (DII) of the Nav1.5 α-subunit protein (Fig. 2B). This mutation had not been registered in the SNP database of the National Center for Biotechnology Information (www. ncbi.nlm.nih.gov). No other mutation was found in the gene screen. The patient's parents and brother were all negative for the L812Q mutation (Fig. 2C).
Genetic analysis of the index patient with the SCN5A mutation. (A) Sequence analysis of the SCN5A gene reveals a heterozygous missense mutation in exon 16, resulting in a change of Leucine (L) to Glutamine (Q) at position 812. (B) Membrane topology of Nav1.5 channel protein. The location of the identified L812Q mutation is shown. (C) Family pedigree. The filled black square indicates the carrier of the L812Q mutation.
Genetic analysis of the index patient with the SCN5A mutation. (A) Sequence analysis of the SCN5A gene reveals a heterozygous missense mutation in exon 16, resulting in a change of Leucine (L) to Glutamine (Q) at position 812. (B) Membrane topology of Nav1.5 channel protein. The location of the identified L812Q mutation is shown. (C) Family pedigree. The filled black square indicates the carrier of the L812Q mutation.
Characterization of Nav1.5 L812Q mutant Na+ channels
In order to understand the pathophysiologic mechanisms underlying the BrS phenotype in the patient, Nav1.5 channels were transiently expressed in HEK 293 cells and the effect of the L812Q mutation on channel function was investigated by whole-cell patch clamp recordings. The L812Q mutation significantly affected the function of the Nav1.5 channel. Compared to WT Nav1.5 channel, the L812Q mutation greatly reduced the Na+ current amplitude. The averaged current-voltage relations plotted in Fig. 3A showed that the current density was drastically decreased by L812Q mutation. The maximum peak Na+ current density at -30 mV were 298 ± 18 pA/pF (n = 8) and 130 ± 24 pA/pF (n = 9) for Nav1.5 WT and L812Q mutant channel, respectively. However, the voltages for the threshold of activation and the maximum peak current as well as the reversal potential for the current were not significantly altered by the L812Q mutation. The voltage dependence of steady-state activation and inactivation for Nav1.5 WT and L812Q mutant channel were determined as shown in Fig. 3B. The activation curves were not significantly different between WT and L812Q mutant channels, with a half-activation voltage (V0.5) of -40.9 ± 0.2 mV and a slope factor (k) of 5.1 ± 0.3 mV for the Nav1.5 WT channel (n = 9) and a V0.5 of -41.2 ± 0.2 mV and a k of 4.7 ± 0.2 mV for the L812Q mutant channel (n = 7). In contrast to the activation curves, the steady-state inactivation curve was significantly negatively shifted by the L812Q mutation. For the Nav1.5 WT channel, the V0.5 and k for the steady-state inactivation curve were -72.7 ± 0.9 mV and 8.0 ± 0.8 mV (n = 8), respectively. For the L812Q mutant channel, the V0.5 was shifted to -92.4 ± 0.6 mV with a k of 6.2 ± 0.4 mV (n = 7). These results indicated about 20 mV hyperpolarizing shift of the inactivation curve by L812Q mutation, suggesting that L812Q mutation enhanced the channel inactivation. Furthermore, the hyperpolarizing shift of voltage dependence of inactivation reduced the Na+ window current as shown in Fig. 3C. The Na+ window current was estimated from the overlay of activation and inactivation curves [26]. The negative shift of steady-state inactivation curve with an unchanged activation curve significantly decreased the Na+ window current. The peak of window current occurred at -61.5 ± 0.5 mV (n=8) and -52.2 ± 0.8 mV (n=8) for L812Q and WT channel, respectively, indicating a more negative voltage range for the window current in the L812Q mutant than in the WT channel. The kinetics of recovery from inactivation was also measured. The fast and slow time constant for recovery from inactivation were 5.16 ± 0.26 and 110.48 ± 29.06 ms (n = 8), respectively, for L812Q mutant channel, not significantly different from that for the Nav1.5 WT channels (5.29 ± 0.17 and 109.44 ± 20.04 ms, n=8) (Fig. 3D) (Table 1), suggesting that the conformation of the inactivated channel protein might not be altered by the L812Q mutation. Thus, the observed hyperpolarizing shift of the voltage dependence of inactivation and reduction of Na+ window current would result in a decrease in the numbers of open Nav1.5 channels during the upstroke and the repolarization phases of the cardiac action potential, leading to a decrease of the inward depolarizing Na+ current.
Gating kinetics of WT and L812Q. The numbers in the parentheses represent number of patches. Af = the fraction of fast components; As = the fraction of slow components; k = slope factor; τf = time constants of fast components of recovery and current decay; τs = the time constants of slow components of recovery and current decay; V1/2 = the membrane potential for half maximal activation or inactivation; WT = wild type. **= P<0.01

Comparison of the electrophysiological characteristics of WT and L812Q mutant Nav1.5 channels. WT, L812Q mutant or both WT and L812Q mutant (WT/L812Q) channels were transiently expressed in HEK-293 cells. Co-expression of WT and L812Q mutant channels (WT/L812Q) was conducted as described in METHODS. (A) Na+ current density and voltage relationships obtained from the cells expressing WT, L812Q or WT/L812Q channels. The Na+ currents were measured at membrane potentials between -80 to +60 mV. The holding potential was -120 mV. The experimental pulse protocol was illustrated in the inset. (B) Steady-state activation and inactivation curves for WT, L812Q or WT/L812Q channels. The insets illustrate the protocols for the measurement of steady-state activation and inactivation. The parameters for fitting the curves are seen in RESULTS and Table 1. (C) The change of the Na+ window currents induced by Nav1.5 L812Q mutant channel. Overlap of steady-state inactivation curves and activation curves display the window currents of WT, L812Q and WT/L812Q channels. (D) The time course of recovery from inactivation for WT, L812Q and WT/L812Q channels. The curves are fitted using bi-exponential function. The time constants are seen in RESULTS and Table 1. The inset shows the pulse protocol used for the measurement of recovery from inactivation.
Comparison of the electrophysiological characteristics of WT and L812Q mutant Nav1.5 channels. WT, L812Q mutant or both WT and L812Q mutant (WT/L812Q) channels were transiently expressed in HEK-293 cells. Co-expression of WT and L812Q mutant channels (WT/L812Q) was conducted as described in METHODS. (A) Na+ current density and voltage relationships obtained from the cells expressing WT, L812Q or WT/L812Q channels. The Na+ currents were measured at membrane potentials between -80 to +60 mV. The holding potential was -120 mV. The experimental pulse protocol was illustrated in the inset. (B) Steady-state activation and inactivation curves for WT, L812Q or WT/L812Q channels. The insets illustrate the protocols for the measurement of steady-state activation and inactivation. The parameters for fitting the curves are seen in RESULTS and Table 1. (C) The change of the Na+ window currents induced by Nav1.5 L812Q mutant channel. Overlap of steady-state inactivation curves and activation curves display the window currents of WT, L812Q and WT/L812Q channels. (D) The time course of recovery from inactivation for WT, L812Q and WT/L812Q channels. The curves are fitted using bi-exponential function. The time constants are seen in RESULTS and Table 1. The inset shows the pulse protocol used for the measurement of recovery from inactivation.
Consistent with an autosomal dominant trait, the proband was found to be heterozygous
In order to understand the heterozygous state of Nav1.5 channel in the patient, WT and L812Q mutant Nav1.5 channels were co-expressed in HEK293 cells (WT/L812Q). Na+ currents recorded from the cells co-expressing WT and L812Q mutant channel mimicking the heterozygous condition appeared to be a mixture of the mutant and WT channel. The Na+ current density was less reduced by WT/L812Q co-expression (Fig. 3A). The maximum peak Na+ current density at -30 mV was 184 ± 19 pa/pF (n = 9) for WT/L812Q channels, in between WT and L812Q mutant channels (Fig. 3A). The steady-state activation curve for WT/L812Q overlapped with that for WT and L812Q mutant channels, with a V0.5 of -41.1 ± 0.3 mV and a k of 5.0 ± 0.3 mV (n = 8) (Fig. 3B) (Table 1). The steady-state inactivation curve was still negatively shifted, but to a lesser extent than the L812Q mutant channel, with a V0.5 of -83.5 ± 0.4 mV and a k of 8.4 ± 0.3 mV for the WT/L812Q channel co-expression (n = 8) (Fig. 3B) (Table 1). Consequently, the Na+ window current was increased to an extent between the WT and L812Q mutant channels (Fig. 3C). The time course of recovery from inactivation was not altered by WT/L812Q co-expression, overlapping with that for WT and L812Q mutant channels (Fig. 3D) (Table 1). These results suggested that at the heterozygous state, the L812Q mutation did not exert a dominant-negative effect on WT channels.
Impaired trafficking of Nav1.5 channel caused by L812Q mutation
Because the L812Q mutation greatly reduced Na+ current density, we further investigated the effect of the L812Q mutation on the trafficking of the channel to the plasma membrane. The top panel of Fig. 4A showed typical images for the subcellular localization of the WT Nav1.5 and L812Q channels. In agreement with the reduced Na+ current density, the surface expression of the L812Q mutant channels was decreased, compared to that of WT Nav1.5 channels. Co-expression of WT and L812Q mutant Nav1.5 channels also showed lowered membrane associated fluorescence and increased intracellular fluorescence compared to the WT Nav1.5 channels. Next, we used the assay of immunofluorescence to observe the co-localization of L812Q with Na/K-ATPase, or the co-localization of L812Q with calnexin (bottom panel of Fig. 4A). The results showed that more intracellular retention was associated with L812Q mutant channels.
Location of WT and L812Q mutant Nav1.5 channels expressed in HEK293 cells. (A) Top panel: Confocal microscopy images of the cells transiently expressing the channels as indicated in the figure. Note that the clear plasma membrane localization of the WT Nav1.5 channels is reduced with WT/L812Q co-expression and not observable with only expression of L812Q mutant channels. Scale bar indicates 37.5 µm. Bottom panel: immunofluorescence images of the cells transiently expressing the L812Q channels. The green stain is presented Na/K-APTase and calnexin respectively, the red stain is presented SCN5A-L81Q. (B) Western blot of lysates of HEK293 cells transiently transfected with WT, L812Q, or WT/L812Q channel. Top panel: the expression of WT, L812Q, or WT/L812Q channel on total proteins, β-actin are used as a control. The bottom panel: the expression of WT, L812Q, or WT/L812Q channel on transmembrane proteins, Na/K-ATPase as a control (n=5).
Location of WT and L812Q mutant Nav1.5 channels expressed in HEK293 cells. (A) Top panel: Confocal microscopy images of the cells transiently expressing the channels as indicated in the figure. Note that the clear plasma membrane localization of the WT Nav1.5 channels is reduced with WT/L812Q co-expression and not observable with only expression of L812Q mutant channels. Scale bar indicates 37.5 µm. Bottom panel: immunofluorescence images of the cells transiently expressing the L812Q channels. The green stain is presented Na/K-APTase and calnexin respectively, the red stain is presented SCN5A-L81Q. (B) Western blot of lysates of HEK293 cells transiently transfected with WT, L812Q, or WT/L812Q channel. Top panel: the expression of WT, L812Q, or WT/L812Q channel on total proteins, β-actin are used as a control. The bottom panel: the expression of WT, L812Q, or WT/L812Q channel on transmembrane proteins, Na/K-ATPase as a control (n=5).
Western blot assay was used to further evaluate the amount of WT and L812Q mutant Nav1.5 channels proteins at the total cell or cell surface. Fig. 4 showed that the expression of L812Q mutant channel at the plasma membrane surface was significantly decreased compared to that of WT Nav1.5 channels and total cell expression of the L812Q mutant was reduced compared to WT channels. The co-expressed WT/L812Q channel also showed decreased expression at the plasma membrane compared to homozygous WT Nav1.5 channels. Taken together, the results indicated that the L812Q mutation led to increased intracellular retention of the channel, reducing the expression at the plasma membrane surface.
Discussion
BrS is an inherited cardiac disorder with ECG characteristics of ST-segment elevation in the right precordial leads V1-V3 and right bundle branch block [2]. BrS accounts for ≥20% of sudden cardiac death in patients with structurally normal hearts and ≥4% of all sudden deaths [27,28]. Many researches show that a lot of genes are involved in development of BrS, such as SCN5A, GPD1L, CACNA1C, CACNB2, SCN1B, KCNE3, SCN3B. Mutations in the SCN5A gene encoding cardiac Nav1.5 channel have been identified that account for ≈20% to 30% of all BrS cases, 11-12% can be attributed to mutations in CACNA1C and CACNB2. Minor contributions to BrS cases are made from mutations in other genes [19,29]. Here, we report a novel mutation in the cardiac Nav1.5 channel, L812Q, in a proband with BrS. This missense mutation is in the DII-S4 transmembrane helices of Nav1.5 channel. Unlike most individuals diagnosed with BrS who have inherited the pathogenic variant from a parent, the proband has no family record of BrS. Sequencing of the DNA failed to identify the mutation in his parents and brother, indicating that this L812Q mutation is likely de novo. BrS caused by de novo mutation is rare, estimated at about 1% of all cases with the syndrome[30].
The human sodium channel Nav1.5, encoded by SCN5A gene, plays a critical role in cardiac excitability and the propagation of action potentials [31]. Mutations in the SCN5A gene can cause a number of inherited arrhythmia syndroms [32]. In our research, we have investigated the functional consequences of the L812Q mutation on the biophysical properties of Nav1.5 channel. Channels with the L812Q mutation exhibited severely reduced, by ∼60%, peak Na+ current density (Fig. 3A). In addition, the mutation resulted in a negative shift of the steady-state inactivation curve of the Na+ current by ∼20 mV without change of the voltage dependence of activation (Fig. 3B). Such a shift represents a decreased availability and an enhanced stabilization of the inactivated state of the Na+ channels, causing decrease of Na+ current during activation and the window Na+ current during the course of repolarization. Consequently, the reduction of the Na+ current during the depolarization and the repolarization phases of the action potential are likely to decrease action potential upstroke velocity, leading to slow atrial and ventricular conduction and disrupt the balance between the inward and outward currents during phase 1 of the action potential, likely shortening the plateau phase of the action potential. The effects of L812Q mutation on the function of Nav1.5 channel further proved that loss of function of Na+ channel can cause BrS.
To mimic the hyterozygous in vivo situation wherein the WT and mutated alleles coexist in the same patient, we performed co-expression experiments in which equal amounts of WT and L812Q mutation DNA were used to transfect HEK cells. We found that the density of Na+ currrent was reduced to ∼60% of the WT Na+ current density, and the voltage dependence of inactivation was also shifted towards negative voltages compared to the WT channel. Some research showed that when co-expressing trafficking-deficient BrS SCN5A mutations with WT Nav1.5 channel (0.5:0.5 in ratio) in heterologous expression systems, it is observed that the Na+ current was reduced by about 50% compared to the Na+ current produced by expressing only WT Nav1.5 channel [33,34,35]. Such reduction of Na+ current is due to the haploinsufficiency of WT Nav1.5 channel. However, other research showed that co-expressing mutations with WT channel, the Na+ current density was also drastically reduced by more than 40% compared to the control condition where 100% of WT channels were expressed, suggesting that those mutations exert a dominant negative effect on WT channels [36,37]. Furthermore, a recent study report that some BrS SCN5A mutations heterologously in the homozygous state, only minor biophysical defects, but not produce a BrS phenotype. However, when these mutations were co-expressed with WT Nav1.5 channel in patients, these atypical mutations could lead to a severe reduction in Na+ current densities similar to typical BrS mutations, i.e., they appear to exert a dominant-negative effect. This decrease in current density was the result of reduced surface expression of both mutant and WT channels [38]. Unlike the mutations that exert dominant negative effects on WT channels, the effect of L812Q mutation on channel function was less severe as the channel still induced significant Na+ current in a heterologous expression system in the homozygous state. Co-expression of the L812Q mutation with WT Nav1.5 channel decreased Na+ current density by about 40% (Fig. 3). The relatively benign reduction of the Na+ current induced by co-expression of L812Q mutation and WT channel is consistent with haploinsufficiency without producing dominant negative effect, suggesting that L812Q mutation did not exert a dominant-negative effect on WT channels. Interestingly, when we text the expression of L812Q channel protein on total proteins by using western blot, we found that the expression of L812Q was lesser than that of WT channel (Fig. 4), suggesting that some proteins might be degraded in cytoplasm caused by intracellular retention of L812Q channel protein.
L812Q occurs in the DII-S4 transmembrane helices of the Nav1.5 channel protein, which is one of the “voltage sensors” that carries a number of positively charged residues, lysine and arginine, to sense the changes in the transmembrane electrical field for activation gating of the channel [39]. Other mutations affecting residues in the DII-S4 (R808P, R811H, L812P, R814Q, and W822X) have been reported to be associated with BrS [40]. The functional consequences of these mutations include impairment of the channel trafficking and alterations of the channel gating process. Similar to these mutations, L812Q mutation has the same function (Fig. 3 and 4). The recent determined structures of bacterial sodium channels gave us good templates to model the mutations in their human relatives. The homologous models of wild type and mutant Nav1.5 (Fig. 5), generated based on the NavAb crystal structure [23], indicate that the L812Q mutation probably does not change the overall structure of the channel (overall RMS 0.096A between 11446 atom pairs). L812 is located in a hydrophobic pocket formed by S1/S4 from domain II and S5 from domain III. The hydrophobic properties of the pocket (M734, L813, V815, I1346, I1349, M1350), is conserved in NavAb as well as many other Nav channels, suggesting a probable functional importance of these interactions. Introduction of a hydrophilic bulky residue glutamine in the hydrophobic pocket will no doubt destabilize the interface between S4 from domain II and S5 from domain III, and affect the relevant motion. From the surface view, L812 is half buried and half exposed to a water cavity connected to the extracellular side, suggesting a function, perhaps, like a lid for the hydrophobic pocket. Mutations to glutamine will open the lid and introduce water molecules further into the crevice between S4 and S5, which may lower the energy barrier even more to facilitate the relative motion between two helices. The possible effect of L812Q mutation on the interaction between DII-S4 and DIII-S5 of the channel are very intriguing. Whether these interactions will affect the inactivation process of the Nav channel is highly worthy of further exploration in the future studies.
Mapping of L812Q residue onto a Nav1.5 homology model. (A) Top view of Nav1.5 structural model. The model is created by Phyre2 server [22] based on the crystal structure of NavAb (PDB:3RVZ) and energy minimized by chimera. Wild type Domains I through IV are colored in gold, cyan, purple and green, respectively. Domain II with L812Q mutation is colored in red. Leu812 is shown as black spheres. (B) Enlarged view of the local environment of L812Q. The side chains of residue 812 (shown in black) and the neighboring residues are displayed as stick representations (wild type in gold and mutant in cyan). (C) Surface view of L812Q. Residue 812 (black) is half buried and half exposed to a water cavity (wild type in gold and mutant in cyan).
Mapping of L812Q residue onto a Nav1.5 homology model. (A) Top view of Nav1.5 structural model. The model is created by Phyre2 server [22] based on the crystal structure of NavAb (PDB:3RVZ) and energy minimized by chimera. Wild type Domains I through IV are colored in gold, cyan, purple and green, respectively. Domain II with L812Q mutation is colored in red. Leu812 is shown as black spheres. (B) Enlarged view of the local environment of L812Q. The side chains of residue 812 (shown in black) and the neighboring residues are displayed as stick representations (wild type in gold and mutant in cyan). (C) Surface view of L812Q. Residue 812 (black) is half buried and half exposed to a water cavity (wild type in gold and mutant in cyan).
In summary, we have identified a Nav1.5 mutation, L812Q, located in the DII-S4 transmembrane region, in a BrS patient. This novel mutation not only causes a reduction of channel expression at the cell surface, but also reduces the Na+ current and significantly hyperpolarizes the voltage-dependent inactivation of the channels. Those biophysical modifications of Nav1.5 by L821Q predispose to the BrS of this patient. However, we find that L812Q is de novo mutation and does not exert a dominant-negative effect on WT channels, suggesting that the environmental factor, other genes or some hidden reasons may be involved in pathogenicity of the L812Q and those clinical features. Further studies are needed to answer these questions. Additionally, in the context of the recently published crystal structure of the bacterial NavAb channels, the evident functional effects of the mutation correlate with a lack of change of the overall structure, suggesting that the disruption of specific intra- and/or inter-domain interaction could underlie the mutation-dependent effects.
Acknowledgments
We thank Dr. David Steele for valuable comments on the manuscript. National Natural Science Foundation of China -Canada Cooperation Projects (81161120549).
Disclosure Statement
No potential conflicts of interest were disclosed.
References
L. Wang and X. Meng contributed equally to this research.