Sodium current deficit and arrhythmogenesis in a murine model of plakophilin-2 haploinsufficiency

Abstract

Aims: The shRNA-mediated loss of expression of the desmosomal protein plakophilin-2 leads to sodium current (I(Na)) dysfunction. Whether pkp2 gene haploinsufficiency leads to I(Na) deficit in vivo remains undefined. Mutations in pkp2 are detected in arrhythmogenic right ventricular cardiomyopathy (ARVC). Ventricular fibrillation and sudden death often occur in the 'concealed phase' of the disease, prior to overt structural damage. The mechanisms responsible for these arrhythmias remain poorly understood. We sought to characterize the morphology, histology, and ultrastructural features of PKP2-heterozygous-null (PKP2-Hz) murine hearts and explore the relation between PKP2 abundance, I(Na) function, and cardiac electrical synchrony.

Methods and results: Hearts of PKP2-Hz mice were characterized by multiple methods. We observed ultrastructural but not histological or gross anatomical differences in PKP2-Hz hearts compared with wild-type (WT) littermates. Yet, in myocytes, decreased amplitude and a shift in gating and kinetics of I(Na) were observed. To further unmask I(Na) deficiency, we exposed myocytes, Langendorff-perfused hearts, and anaesthetized animals to a pharmacological challenge (flecainide). In PKP2-Hz hearts, the extent of flecainide-induced I(Na) block, impaired ventricular conduction, and altered electrocardiographic parameters were larger than controls. Flecainide provoked ventricular arrhythmias and death in PKP2-Hz animals, but not in the WT.

Conclusions: PKP2 haploinsufficiency leads to I(Na) deficit in murine hearts. Our data support the notion of a cross-talk between desmosome and sodium channel complex. They also suggest that I(Na) dysfunction may contribute to generation and/or maintenance of arrhythmias in PKP2-deficient hearts. Whether pharmacological challenges could help unveil arrhythmia risk in patients with mutations or variants in PKP2 remains undefined.

Figure 1

Characterization of PKP2-Hz hearts. (A) Picrosirius red staining to determine collagen abundance. Images obtained from left (LV; left) or right ventricle (RV; right) of mice wild type (WT; top) or PKP2-Hz (bottom). Extent of Picrosirius-red staining determined from fraction of Sirius red-positive pixels within a field. Data for each field sampled in a WT heart normalized to average value of 12 random fields in same heart. Data for PKP2-Hz hearts measured relative to average value in littermate WT hearts, processed in parallel. No difference observed between genotypes when compared with respective ventricle: RV-WT: 1.0 ± 0.16, RV-PKP2-Hz: 0.77 ± 0.08; pNS. LV-WT: 1.0 ± 0.15, LV-PKP2-Hz 1.25 ± 0.16; pNS (ANOVA). N = 4, n = 48. Calibration bar, 50 µm. (B) Western blots for intercalated disc proteins. For quantification (right panel), band densities were corrected by total protein, and PKP2-Hz data measured relative to WT on same exposure. *P< 0.05 compared with WT. All other comparisons, NS. (C and D) Immunofluorescence images for PKP2 (green) and N-Cad (red; C; Calibration bar, 50 µm) and for Na_V1.5 (green) and N-Cad (red; D; Calibration bar 25 µm) in WT and PKP2-Hz hearts. No difference was apparent between groups.

Figure 2

T-EM images of intercalated disc in PKP2-Hz adult heart. Outlined region in left panels, enlarged for insets (A–F). Each frame corresponding to different plane of visualization. Notice membrane invaginations, extending into the intracellular space and detaching from a healed membrane in some sectional planes.

Figure 3

Sodium current properties recorded from adult cardiac myocytes isolated from PKP2-Hz hearts. (A) Average peak sodium current density as a function of voltage command. Peak current amplitude at −45 mV: WT: −36.1 ± 2.4 pA/pF; n = 12. PKP2-Hz: −26.3 ± 2.6 pA/pF; n = 16. P< 0.01. (B) Voltage dependence of steady-state activation curves. Voltage for half-maximal activation (V_1/2): −57.4 ± 0.9 mV for WT and −58.9 ± 0.7 mV for PKP2-Hz (pNS). (C) Voltage dependence of steady-state inactivation. Voltage for half-maximal inactivation (V_1/2): −89.6 ± 1.4 mV for WT and −95.3 ± 1.1 mV for PKP2-Hz (P< 0.005). (D) Time course of recovery from inactivation. Time constant (one exponential function): 8.5 ± 0.9 ms for WT and 11.9 ± 0.7 ms for PKP2-Hz (P< 0.002).

Figure 4

PKP2 deficiency and use-dependent, flecainide-induced I_Na block. (A–C) Data from adult ventricular myocytes dissociated from either PKP2-Hz (red) or WT littermates (black). (A) Time course of I_Na, relative to current density at patch break. After 60s of regular pacing, voltage clamp pulses were interrupted, and Flecainide (1 μmol/L) added (arrow). Repetitive pulses were re-initiated 3min after addition of drug. SEM noted for each time point. (B) Compiled data for fraction of current decrease consequent to use-dependent flecainide block. WT, n = 8, 9, 8 and PKP2-Hz n = 10, 9, 9 for 1, 10 and 100 μmol/L flecainide, respectively. *P< 0.05 for each concentration. (C) Fast and slow time constants of use-dependent flecainide block (*P< 0.05).

Figure 5.

Epicardial activation maps in PKP2-Hz and control hearts. (A) Examples of activation maps for left (LV; top) and right ventricle (RV; bottom) of WT (left) and PKP2-Hz hearts in the absence or presence of flecainide. Colours indicate activation times. Isochrones outlined by black lines. Area depicted within each frame: 5.4 by 3.6 mm. Site of stimulation indicated by square-pulse symbol. (B and C) Average longitudinal conduction velocity (CV), measured in right (RV; B) or left ventricle (LV; C) of either WT (black symbols) or PKP2-Hz mice (red symbols). *P< 0.05.

Figure 6.

Electrocardiographic features of PKP2-Hz mice at baseline, and in response to flecainide. (A) Left, example of ECG traces from WT (top) and PKP2-Hz mouse, (bottom). Recordings obtained at baseline (left) and 10 min after flecainide (40 mg/kg i.p., right). Right: VT in PKP2-Hz mice. Recordings obtained from two animals, 10′ (top) and 9′ (bottom) after flecainide i.p. (B) Graph bars show average P wave duration, PR interval, QRS duration, and QTc interval measured at baseline, 5′ and 10′ after flecainide injection. A total of 11 WT and 12 PKP2-Hz animals were studied. Criteria for exclusion from data set: for P duration and PR interval, subject excluded from data set if ECG showed at least one of the following (number of excluded subjects from a given time point, in parentheses): (1) ventricular tachycardia, VT (4 PKP2-Hz at 10′), (2) second-degree AV block (5 WT, 3 PKP2-Hz at 10′), (3) atrial tachycardia, diagnosed for altered P polarity and shorter PR interval (one PKP2-Hz at 5′ and at 10′), or (4) P wave inversion, likely signalling a low atrial rhythm (two PKP2-Hz at baseline and at 5′). First two criteria also applied to QTc data set. For QRS duration, animals in VT were excluded; animals in second degree AV block were included, with parameters measured only from conducted sinus beats. n values indicated in each bar. *P< 0.05; **P≤ 0.01. Additional data in Supplementary material online, Tables S3—S5 and Figure S3.