Unexpected impairment of INa underpins reentrant arrhythmias in a knock-in swine model of Timothy syndrome

Figures

• 

  Fig. 1. TS1 pigs recapitulate cardiac human phenotype. a, Representative ECG traces of WT (top, in blue) and TS1 (bottom, in red) pigs showing marked QTc interval prolongation in TS1. b, Box plot shows the distribution of QTc interval in n = 27 WT animals (minimum 419 ms, median 482 ms, IQR: 469–500 ms, maximum 514 ms) and n = 28 TS1 animals (minimum 521 ms, median 586 ms, IQR: 552–615 ms, maximum 657 ms). Statistical analysis was conducted using the two-tailed Mann–Whitney U-test (P < 0.0001). c, Example of lethal ventricular arrhythmias in TS1 pigs, triggered by PVCs (red asterisks). d, Documentation of a sequence of three PVCs on the 24-h Holter ECG (above) and a PVC-triggered episode of VF recorded by implantable cardioverter defibrillator (below) in a young patient with TS1. Source data

• 

  Fig. 2. Biventricular simultaneous mapping and ventricular fibrillation induction. a, Graphs show the quantification of LAT, LRT and LRT range of the S2, S3 and S4 as close as possible to ventricular refractoriness during spontaneous sinus rhythm between WT (blue, n = 8) and TS1 (red, n = 10) animals. Each value is represented as mean ± s.d. Statistical analysis was conducted using two-way ANOVA with Šidák post tests. b, Surface ECG (left), unipolar electrograms (central) and electroanatomical maps (right) documenting ventricular fibrillation induction in a TS1 pig after three extrastimuli on sinus rhythm, a situation that never occurred in WT pigs. The first activation during S4 corresponded to the pacing site in the RV (*). The latest activated area was found in the LV (**), whereas the first spontaneously reactivating region was found in the proximal site in the RV (+). Notably, this early reactivated area (reentrant activity) corresponded to the area with the shortest LRT registered during the closest S4 that did not induce ventricular fibrillation. The black arrow indicates the pacing site in the RV endocardium. Source data

• 

  Fig. 3. Ultra-high-density sequential EAM discloses the pathological functional substrate in the TS1 swine model. Electroanatomical endo-epicardial maps and exemplificative unipolar electrograms (*) of a WT animal (top) and a TS1 animal (bottom) are represented. Compared to the WT, three premature stimuli caused a marked activation delay in TS1 (bottom left), whereas the LRT abbreviated heterogeneously, creating an increased dispersion of repolarization, as compared to WT (top right). Black arrows indicate the pacing site in the RV endocardium.

• 

  Fig. 4. Advanced mapping metrics obtained from ultra-high-density sequential EAM confirmed the presence of a substrate vulnerable to reentry in the TS1 swine model. Compared to WT (upper left), three premature stimuli caused the appearance of areas of slow conduction (CV <0.2 m s−1) in TS1 (upper right). In correspondence with such areas, the LRT abbreviates heterogeneously, as demonstrated by steep LRT gradients (central right), creating a substrate vulnerable to the development of functional reentry, as documented by extremely low RVI (lower right). Black arrows indicate the pacing site in the RV endocardium.

• 

  Fig. 5. RVI assessment in a TS1 model after administration of various pharmacological compounds, visualized using high-density EAM. This figure presents representative RVI maps for the same TS1 pig before (left panels) and after (right panels) pharmacological interventions. The color-coded regions, green and red, represent areas with a lower and a higher likelihood of reentrant arrhythmia, respectively, as per the RVI evaluation. Administration of sodium channel blockers mexiletine and ranolazine exacerbated the arrhythmic substrate, as indicated by a worsening of the global RVI. Conversely, verapamil administration enhanced the uniformity of cardiac activation and repolarization, leading to a slight improvement in the global RVI. However, the verapamil-induced benefits were reversed by the concomitant administration of metoprolol, a selective beta blocker. Administration of dextromethorphan (DXM) resulted in more homogeneous cardiac repolarization, but the effects on cardiac activation were minimal, thus resulting in only minor changes in the arrhythmic substrate, as evidenced by negligible changes in RVI. Finally, administration of ICA-105574 resulted in a significant reduction in…

• 

  Fig. 6. Cellular phenotype of TS1 cardiomyocytes. a, Representative traces (20-mV increments from −20 mV) of ICa in WT (blue) and TS1 (red) cardiomyocytes. From those measurements, peak I-V curve and speed of inactivation were calculated. b, Peak ICa I-V relationships for WT (blue, n = 22 cells from N = 5 animals) and TS1 (red, n = 15 cells from N = 3 animals) (P = 0.825). c, Analysis of ICa inactivation rates in WT (blue, n = 21 cells from N = 5 animals) and TS1 (red, n = 15 cells from N = 3 animals). For τ fast: P = 0.020 and P = 0.047 at +20 mV and +30 mV, respectively. d, Activation and availability curves of ICa in WT (blue, n = 14 cells from N = 3 animals) and TS1 (red, n = 20 cells from N = 3 animals) (P = 0.010 and P < 0.0001 at +20 mV and +30 mV, respectively). e, Recovery from inactivation of ICa for WT (blue, n = 15 cells from N = 3 animals) and TS1 (red, n = 18 cells from N = 3 animals) (P = 0.300). f, Representative traces of AP and simultaneous Ca2+ transient recordings at 0.5 Hz, 1 Hz and 2 Hz in WT (top) and TS1 (bottom) isolated cells. g, Quantification of APD90 at 0.5 Hz, 1 Hz and 2 Hz in WT (n = 38/33/17 cells from N = 14/14/10 animals, respectively) and TS1 (n =…

• 

  Fig. 7. EADs in TS1 and potential contribution of late-systolic Ca2+ waves/ripples to EAD genesis and maintenance. a–c, EADs in TS1. a, Representative examples of EADs with different take-off potentials in TS1 cardiomyocytes and quantification of take-off potentials of EADs in WT (blue, n = 5 EADs from n = 2 cells from N = 2 animals) and TS1 (red, n = 35 EADs from n = 11 cells from N = 9 animals) cells paced at 0.5 Hz, 1 Hz or 2 Hz (P = 0.159). Each value is represented as mean ± s.d. Statistical analysis was conducted using two-tailed nested t-test. b, Quantification of the incidence of EADs in WT (blue, n = 32/25/15 cells from N = 14/14/10 animals) and TS1 (red, n = 19/22/11 cells from N = 6/7/5 animals) cells paced at 0.5 Hz, 1 Hz or 2 Hz (P = 0.040, P = 0.007 and P = 0.423 at 0.5 Hz, 1 Hz and 2 Hz, respectively). Each dot in the graph represents the percentage of cells with EADs. Statistical analyses were conducted using two-tailed chi-square with Fisher exact test. NS, not significant, *P < 0.05, **P < 0.01, ***P < 0.001 and ****P < 0.0001. c, Simultaneous voltage (black line) and global Ca2+ recordings (red line, confocal image) from a TS1 cell presenting EADs at 0.5 Hz. The figure…

• 

  Fig. 8. INa alterations in TS1: regulation by Ca2+ and CaMKII. a–d, Peak INa in the absence of intracellular Ca2+. a, Traces of INa in WT (top, in blue) and TS1 (bottom, in red) cardiomyocytes. Graph shows peak INa I-V relationships for WT (blue, n = 17 cells and N = 4 animals) and TS1 (red, n = 23 cells and N = 4 animals) (P = 0.025 at −40 mV, P = 0.006 at −35 mV and P = 0.025 at −30 mV; two-way ANOVA with Šidák post test). b, INa activation and availability curves for WT (blue, n = 14 cells and N = 4 animals) and TS1 (red, n = 13 cells and N = 5 animals) (P = 0.025 for availability and P = 0.091 for activation; two-tailed nested t-test). c, INa inactivation rates WT (blue, n = 17 cells and N = 4 animals) and TS1 (red, n = 23 cells and N = 4 animals) (two-way ANOVA P = 0.005; P = 0.009 at −35 mV, P = 0.003 at −25 mV and P = 0.003 at −15 mV; two-way ANOVA with Šidák post test). d, Recovery from inactivation of INa for WT (blue, n = 12 cells and N = 5 animals) and TS1 (red, n = 20 cells and N = 6 animals) (P < 0.0001; two-tailed F-test). e, Late INa in the absence of intracellular Ca2+. Left, representative recordings of WT (blue) and TS1 (red) cardiomyocytes. Late INa current was…

• 

  Extended Data Fig. 1. CRISPR/Cas9-mediated generation, genetic characterization, and expression profiling of TS1 pigs. a, Exon 8 A homology. Sequences of the exon 8 A of the CACNA1C in humans (above) and pig (below) shows a 98% homology (102/104 [98%] homologous bases, shown in green), with only 2 non-identical bases (shown in red). The base of interest is shown in bold. b, Generation of TS1 clones. A primary fibroblast cell line was established from a skin biopsy of a tested boar. Cells were co-transfected using the CRISPR/Cas9 system, specific for the CACNA1C-exon8, and a ssODN carrying the TS1 mutation. Colonies positive to the RFLP-PCR and Sanger sequencing analyses were used as nuclear donors to generate cloned TS1 piglets. For the SCNT experiments, oocytes were aspirated from ovaries and were matured in vitro for 42 hours. The zona pellucida was removed, matured oocytes were enucleated and electrofused to TS1 cells. The resulting constructs were activated and in vitro cultured for six days. The resulting cloned TS1 blastocysts were finally surgically transferred into a synchronized recipient sow. Farrowing was induced in sows to deliver the TS1 cloned piglets. Figure was created with…

• 

  Extended Data Fig. 2. Examining gender-related variations in QTc interval and survival rates between TS1 and wild-type pigs. a, Gender differences in the QTc interval. No gender differences in the QTc interval were observed neither among TS1 pigs (in red tones; N = 28; p = 0.999) nor in WT pigs (in blue; N = 27; p = 0.983), but both TS1 males (red; minimum 522 ms, median 593, IQR: 545-615 ms, maximum 646 ms) and TS1 females (dark red; minimum 521 ms, median 574, IQR: 566-626 ms, maximum 657 ms) showed a significantly longer QTc interval as compared to their age- matched counterparts (in blue tones; males: minimum 419 ms, median 481, IQR: 467-494 ms, maximum 514 ms; females: minimum 449 ms, median 485, IQR: 465-508 ms, maximum 510 ms; p < 0.0001 for both). Each value is represented as median and IQR. Statistical analysis was conducted using one-way ANOVA with Šidák post-tests. b-c, Reduced survival in TS1 pigs. Statistical analysis was conducted using log rank test. b, Kaplan-Meier estimate of cumulative survival free from death shows reduced survival in TS1 pigs (red) as compared to WT (blue) pigs (p < 0.0001). c, Kaplan-Meier estimate of cumulative survival free from death stratified by gender showed…

• 

  Extended Data Fig. 3. Applying advanced imaging and histological techniques for comprehensive cardiac tissue analysis in TS1 pigs. a, Tissue characterization with late gadolinium sequences (left panel), extracellular volume (central panel) and T2 mapping (right panel) did not show any signs suggestive of cardiac fibrosis or inflammation in N = 9 TS1 pigs. b, Histological samples from N = 12 TS1 pigs confirmed the absence of inflammatory infiltrates and absence of fibrosis.

• 

  Extended Data Fig. 4. Second example of simultaneous biventricular mapping and ventricular fibrillation (VF) induction. Surface ECG (left), unipolar electrograms with local activation time depicted with white circles (central) and electroanatomical maps (right) documenting polymorphic VT induction in a TS1 pig after 3 extrastimuli. In analogy with Fig. 5 (see main text), the first activation during S4 corresponds to the pacing site in the right ventricle (*). Latest activated area was found in the left ventricle (**), while the first spontaneously reactivating region was found in the proximal site in the right ventricle (***), which corresponded to the area with the shortest LRT of the closest S4 that did not induce VF and was followed again by left ventricle activation (****). The black arrow indicates the pacing site in the RV endocardium.