Hypokalaemia and bradycardia unmask the loss-of-function phenotype of a Brugada Syndrome SCN5A mutation

Abstract

Aims: Loss-of-function (LOF) mutations of the cardiac Na+ channel (SCN5A) are causatively associated with the Brugada Syndrome (BrS). However, the onset of Ventricular Fibrillation (VF) is a rare event, and critical factors favouring the pathological phenotype remain often elusive. This study explores how concomitant triggering conditions may impact on VF onset in a symptomatic proband carrying the S805L/SCN5A BrS mutation.

Methods and results: Clinical, in-vitro, numerical, and structural analyses were performed. A 67-year-old male was resuscitated after cardiac arrest, and clinical analysis upon hospitalisation revealed severe hypokalaemia (2.5 mEq/L). The ECG showed a coved type-I BrS pattern and the SCN5A mutation (S805L) was identified. Patch-clamp studies carried out in a heterologous expression system (HEK293 cells) revealed that WT/S805L channels exhibit two different phenotypes (normal and LOF); the main parameter controlling this distribution is the cell membrane potential. A protected/normal behaviour was observed at -80 mV; conversely, LOF occurred at more negative potentials (-100/-120 mV). Further analyses in isolated outflow tract ventricular cardiomyocytes showed that hypokalaemia (and bradycardia) induced diastolic potential hyperpolarisation, thus favouring the Na+ current LOF. Computational and molecular modelling confirmed our findings and revealed the structural determinant of this alteration.

Conclusion: WT/S805L Na+ channels exhibit either a LOF or a wild-type-like behaviour depending on the membrane potential. Since hypokalaemia and slow pacing rate induce cell hyperpolarisation and the associated LOF, they represent concurrent elements creating the scenario responsible for the VF and cardiac arrest. These results may represent an interpretative paradigm applicable to other BrS mutations.

Keywords: Arrhythmias; Brugada syndrome; Hypokalaemia; SCN5A.

Graphical Abstract

Figure 1

ECG phenotypes and familial inheritance of the S805L mutation. (A) ECG of the proband collected during cardiac arrest. (B) Type 1 BrS ECG pattern of the proband at hospital admission after resuscitation from cardiac arrest. (C) Schematic topology of the hNa_v1.5 channel; the S805L mutation is located at the extracellular limit of the DIIS4 TM segment. (D) Family tree of the proband (arrow); coloured background indicates the presence of BrS pattern in the ECG; black dots indicate the presence of the heterozygous S805L mutation. (E, F) Ajmaline test (1 mg/Kg in 5 min) was positive in the son and negative in the daughter.

Figure 2

The electrophysiological behaviour of the heterozygous S805L mutation depends on the cell HP. Current/voltage (I/V, A) and conductance/voltage (g/V, B), and activation (C) curves obtained using HPs of −120 (left), −100 (middle), and −80 (right) mV in HEK293 cells transfected with WT only (WT, ▽), WT and S805L (Hetero, ○) and S805L only (Homo, □) channel constructs. Sample currents (40 ms duration, range: −80 to +10 mV) are shown at the top of panel A. g/V curves were fitted by the Boltzmann equation (g(V)=g_max/(1 + exp(−(V−V½)/s))). Statistics were carried out on the peak of the I/V distribution (−20 mV) and on the g_max values. Peak currents, n, and statistical P values are provided in Supplementary material online, Table S2; g_max, n, and statistical P values are provided in Supplementary material online, Table S3. Statistic test: one-way ANOVA followed by post-hoc Fisher test; ^#  P < 0.05 vs. WT. Activation curves were fitted by the Boltzmann equation (y = 1/(1 + exp(−(V−V½)/s))). V½ and s values are provided in Supplementary material online, Table S4. Statistical curve comparisons (P values in the insets) were carried out using the Extra sum of squares F test.

Figure 3

Western blot analysis of S805L homo and WT channel expression in HEK293-transfected cells. (A) Representative blot of total protein extract of HEK293 cells expressing WT and S805L channels. Each lane was loaded with 5 µg of total protein extract obtained from an independent culture dish; top and bottom bands likely represent two different glycosylation states of the channel. (B, C) Densitometric analyses of 3 independent experiments of top + bottom (total, B) and top (C) bands; S805L signals were decreased (vs. WT) by 31.9% and by 40.2% (n = 11, 12), respectively. Data were normalized to endogenous actin and expressed as % of WT values. (D) Ratio of the top vs. bottom channel forms; for the S805L Homo channel the ratio is decreased by 25.1% (n = 11, 12). Box plot: middle line, mean value; extremities, SEM; whiskers, maximum and minimum values. Statistics: two-sample t-test.

Figure 4

The S805L mutation induces a GOF of the voltage-dependent inactivation. (A) Experimental data and Boltzmann fitting of mean fractional inactivation values for the WT (n = 34 cells), Hetero (n = 20), and Homo (n = 12) conditions, respectively. V_½ and s values are presented in Supplementary material online, Table S4; Homo and Hetero curves are significantly different from the WT (statistics are shown in the inset). The Extra sum-of-squares F test was used for statistical comparison. (B) The bell-shaped curve (left panel) corresponds to the difference between the Hetero and the WT inactivation curves shown in panel A and represents the GOF acquired by mutant channels. Diamonds represent the GOF factors at the holding value of −80, −100 (indicated by large arrows), and −120 mV. The bar-graph (right panel) illustrates the recruitability of WT (orange) and Hetero (green). The presence of the arrows in the green area indicates the GOF of Hetero channels.

Figure 5

Hypokalaemia and stimulation rate modulate the diastolic potential (E_diast) of RVOT cardiomyocytes. (A) Sample APs recorded in a cardiomyocyte stimulated at 1 (left) and 4 (right) Hz, and sequentially exposed to 2.5, 3.5, and 5.0 mM extracellular K⁺ concentrations (K ⁺  _out). The top part (>−65 mV) of the AP traces has been removed for clarity. Dashed lines correspond to steady-state E_diast levels recorded at 1 Hz. (B) Single E_diast values measured in n = 9 RVOT myocytes stimulated at 1 (left) and 4 (right) Hz, and sequentially exposed to 2.5, 3.5, and 5.0 mM extracellular K⁺ concentrations (as in panel A); mean ± SEM E_diast (see also Supplementary material online, Table S5) are presented on the right panel. Left and middle panels, statistics: RM one-way ANOVA followed by post-hoc Fisher test. Right panel, statistics: RM Two-way ANOVA; internal comparison reveals a significant interaction between K⁺_out and rate.

Figure 6

Human AP simulations and structural 3D prediction support the pathogenic role of the S805L mutation. (A) Computed human ventricular APs (top) and the corresponding WT and Hetero I_Na traces (bottom) simulated in basal (left: 1 Hz, 5 mM K⁺_out) and hypokalemic and bradycardic (right: 0.66 Hz, 2.5 mM K⁺_out) conditions. APs are presented using a dual time scale to better appreciate the different shapes; the areas identified by dotted rectangles are enlarged in the inset to better illustrate the difference in the time required to reach the peak of the upstroke phase (AP_ud). (B, left) Position of the S805 residue in the DIIS3-S4 loop. Relevant residues at the interface between DIIS4 and DIIS5 are shown in stick representation and labelled. H-bond is shown as dashed a line. (B, right) Structure superimposition of voltage-gated Na⁺ channel Na_v1.5 6LQA (green) and 7DTC (cyan).