Catecholaminergic polymorphic ventricular tachycardia (CPVT) is a familial arrhythmogenic disorder caused by unstable sarcoplasmic reticulum (SR) calcium (Ca²⁺) storage leading to exercise- or emotion-induced ventricular tachyarrhythmias and sudden cardiac death (1). The majority of CPVT cases are associated with dominant mutations in the cardiac ryanodine receptor gene (RyR2) with variable penetrance (CPVT1) (3), whereas the minority of cases result from recessive mutations in the cardiac calsequestrin isoform 2 (CASQ2) gene (5). Investigation of the mutations giving rise to CPVT provided important insights into the mechanisms underlying this disease and to the processes governing cardiomyocyte Ca²⁺ handling in general (6). Nevertheless, among the hurdles in studying this disease process has been the lack of appropriate human cardiac tissue models and the inability to study patient-specific disease variations.

The development of the groundbreaking human- induced pluripotent stem cells (hiPSCs) technology (7) may provide a possible solution to the aforementioned challenges. The iPSCs approach, pioneered by Takahashi and Yamanaka, allows the reprogramming of adult somatic cells into pluripotent stem cells by ectopic expression of a set of transcription factors (9). Subsequent studies showed the robust ability to generate hiPSCs and to coax their differentiation into cardiomyocytes (10). More recent proof-of-concept studies demonstrated the ability to establish patient- and disease-specific hiPSCs-derived cardiomyocytes (hiPSCs-CMs) that could recapitulate the abnormal cardiac phenotypes of Leopard and long-QT syndromes (12). In the current report, our goal was to establish an hiPSCs model of CPVT. We hypothesized that the generated CPVT-hiPSCs-CMs will recapitulate the disease phenotype in vitro, providing insights into disease mechanisms and offering a unique platform to evaluate patient-specific therapies.

The patient-specific hiPSCs clones were generated by retroviral reprogramming of dermal fibroblasts with Oct4, Sox-2, and Klf-4 followed by valproic acid treatment as previously described (13). Cardiomyocyte differentiation was induced using the embryoid body (EB) differentiating system. Briefly, undifferentiated hiPSCs were removed from the mouse embryonic fibroblast feeder-layer, dispersed into cell-clumps using collagenase type IV (300 U/ml [Life Technologies Corporation, Grand Island, New York]), cultivated in suspension for 10 days as EBs, and plated on gelatin-coated culture dishes (11).

Specimens were fixed with 4% paraformaldehyde, permeabilized with 1% Triton, blocked with 5% horse serum, and incubated with primary antibodies targeting: Oct4, Tra1-60 (Santa-Cruz Biotechnology, Santa Cruz, California), Nanog (Peprotech, Rehovot, Israel), SSEA-4 (R&D Systems, Minneapolis, Minnesota), troponin I, ryanodine receptor (Chemicon, Merck Millipore, Billerica, Massachusetts), and sarcomeric alpha actinin (Sigma-Aldrich, St. Louis, Missouri). The preparations were incubated with secondary antibodies and examined using a laser-scanning confocal microscope (Zeiss LSM-510-PASCAL). Alkaline phosphatase staining of hiPSCs colonies was performed using a Sigma detection kit.

Undifferentiated hiPSCs were dissociated with collagenase-IV and injected subcutaneously into SCID/beige mice. Tumor samples were collected at 6 to 8 weeks, fixed in 4%-formaldehyde, embedded in paraffin, sectioned, and stained with hematoxylin and eosin.

Genomic DNA was isolated by using the high-pure polymerase chain reaction (PCR) template preparation kit (Roche, Basel, Switzerland). The relevant DNA fragment of the RYR2 gene was amplified by PCR using 100-ng genomic DNA. Primers are detailed in (5). PCR products were then sequenced.

RNA was isolated by using the RNeasy Plus Mini Kit (Qiagen Inc., Valencia, California). Reverse transcription into cDNA was conducted with the high-capacity cDNA reverse-transcription kit (Applied Biosystems). Real-time polymerase chain reaction (RT-PCR) studies were performed by using the DreamTaq Master Mix (Ferments Molecular Biology Tools, Thermo Fisher Scientific, Waltham, Massachusetts). Primers are listed in (5). Each RT-PCR included 2 min at 93^°C followed by 27 to 35 cycles of 30 s at 93^°C, 60 s at 60^°C, and 40 s at 72^°C.

RT-PCR studies were conducted in triplicate by using the Fast SYBR Green Master Mix (Applied Biosystems, Life Technologies Corporation). Primers are detailed in (5). Samples were cycled by using the Fast ABI-7500 sequence detector. Conditions were as follows: 20 s at 95^°C followed by 40 cycles of 3 s at 95^°C and 30 s at 60^°C. Cycle threshold was calculated by using default settings for the real-time sequence detection software (Applied Biosystems).

Contracting EBs were enzymatically dispersed (1 mg/ml collagenase B [Roche]) and attached to fibronectin-coated glass coverslips. Action potentials (APs) were recorded from spontaneously contracting and quiescent hiPSCs-CMs at 32°C under spontaneous and paced (1 Hz) rhythms. The tyrode solution included (in mmol/l: NaCl 140; KCl 4; CaCl₂ 2; MgCl₂ 1; HEPES 10; and glucose 5 (pH 7.4; NaOH). The pipette solution contained (in mmol/l): K-aspartate 120; KCl 20; MgCl₂ 1; Na₂ATP 4; GTP 0.1; HEPES 10; and glucose 10 (pH 7.2; KOH). APs were recorded in the current-clamp mode with Axopatch 200B, Digidata 1322A, and pClamp 9 (Axon Instruments, Sunnyvale, California).

Microelectrodearray recordings (Multichannels Systems, Reutlingen, Germany) were performed as previously described (11). Local activation times were determined at each electrode and used for generation of color-coded activation maps.

Cells were loaded with 5μM fluo-4 fluorescent Ca²⁺ indicator (Molecular Probes, Life Technologies Corporation) to allow recordings of whole-cell [Ca²⁺]_i transients as previously reported (15). Experiments were conducted during continuous tyrode perfusion that included (in mmol/l) NaCl 140; KCl 5.4; CaCl₂ 1.8; MgCl₂ 1; HEPES 10; and glucose 10. [Ca²⁺]_i imaging was performed by using the Zeiss LSM-700 confocal system. To evaluate for store-overload–induced Ca²⁺ release (SOICR), we used a modification of the method described by Jiang et al. (17). In brief, repeated Ca²⁺ imaging from the hiPSCs-CMCs was performed at increasing extracellular [Ca²⁺]_o concentrations (0.1 to 4 mmol/l) in the presence of tetrodotoxin (10 μM). Tetrodotoxin was added to eliminate AP-dependent Ca²⁺ releases, thereby unmasking only SOICR events.

Continuous variables are reported as mean ± SEM. Categorical variables are expressed as frequencies. Categorical differences between groups were evaluated by using the chi-square test. Differences between group means were compared by using the unpaired Student t test. These statistical tests were applied to comparisons between the cells obtained from the 2 individuals studied and cannot necessarily be generalized to differences between patient populations. p < 0.05 was considered statistically significant.

Dermal fibroblasts were obtained from a 30-year-old woman diagnosed with CPVT1. The patient comes from a family with a significant history of sudden cardiac death (including the patient's father, brother, sister, and nephews) (19) and was implanted prophylactically with an implantable cardioverter-defibrillator. An episode of nonsustained ventricular tachycardia was documented during device follow-up and after an epinephrine stress test (Figure 66_gr1A). Genetic workup revealed a heterozygous missense mutation (M4109R) in the RyR2 gene in all affected family members (19).

Several CPVT-hiPSCs clones were generated during reprogramming of the patient's fibroblasts, 2 of which were propagated and used for cardiomyocyte differentiation and characterization (Figure 66_gr1B). Similar hiPSCs previously created from fibroblasts of a healthy individual (15) served as controls. Importantly, the M4109R mutation was identified in the CPVT-hiPSCs but not in the healthy control cells (Figure 66_gr1C).

The CPVT-hiPSCs colonies exhibited characteristic human embryonic stem cell morphology; stained positively for the pluripotency markers NANOG, OCT4, SSEA-4, and TRA1-60 (Figure 66_gr1B); displayed alkaline phosphatase activity (Figure 66_gr1B [right panel]); and maintained a normal karyotype (data not shown). Pluripotency of the CPVT-hiPSCs was confirmed by the presence of cell-derivatives of all 3 germ-layers in differentiating EBs (Figure 66_gr2A). Note the positive staining for nestin (ectodermal marker), alpha-fetoprotein (endoderm), and desmin (mesoderm) (Figure 66_gr2A). Similarly, injection of undifferentiated CPVT-hiPSCs into SCID/beige mice led to the formation of teratomas, containing advanced tissue derivatives of all 3 germ layers (Figure 66_gr2B). Finally, all clones showed silencing of the 3 retroviral transgenes (Figure 66_gr2C) and reactivation of endogenous pluripotency genes (OCT4, SOX2, NANOG, FOXD3, and REX1) (Figure 66_gr2D), indicating successful reprogramming.

We next used the EB differentiation system to coax the differentiation of the hiPSCs into the cardiac lineage. Both the healthy control and CPVT hiPSCs lines showed similar cardiomyocyte differentiation potential as judged by the number of contracting EBs generated during differentiation (Figure 66_gr3A) and by comparable quantitative expression of cardiac-specific genes by the differentiating cells (Figure 66_gr3B).

Gene expression analysis (Figure 66_gr3C) revealed the expression of cardiac-specific transcription factors (NKX2.5) and structural genes (MLC-2V, MYH-6, MYH-7 and cTnI) by the CPVT-hiPSCs-CMCs. Similarly, immunostaining studies targeting the sarcomeric proteins troponin I and alpha-actinin confirmed the cardiomyocyte phenotype of these cells (Figure 66_gr3D). These experiments also verified the presence of the major SR Ca²⁺ release channels (RyR2) in the CPVT-hiPSCs-CMs (Figure 66_gr3E). No differences were noted in RyR2 expression between healthy and diseased hiPSCs-CMs at the mRNA level (Figure 66_gr3F) and also in protein expression and subcellular localization as judged by immunostaining analysis (Figure 66_gr3E).

Intracellular (patch-clamp) and extracellular (multielectrode) electrophysiological recordings established the presence of cardiac-specific APs by the CPVT-hiPSCs-CMs at the cellular level (Figure 66_gr3G) and the development of a functional syncytium at the tissue level (Figure 66_gr3H). Similar to previous reports (10), 3 types of AP morphologies were recorded from the CPVT-hiPSCs-CMs (Figure 66_gr3G). The most dominant (57% of cells) was the ventricular-like morphology, with atrial-like (29%), and nodal-like (14%) types being less prevalent.

Detailed analysis of AP properties (amplitude, upstroke-velocity, durations, and resting membrane potential) revealed no significant differences between the healthy control and CVPT hiPSCs-CMs (5). In contrast, marked differences were noted in the cardiomyocytes' arrhythmogenic potential. Hence, clear delayed afterdepolarizations (DADs) developed in most CPVT-hiPSCs-CMs when paced at 0.5 to 1 Hz (Figure 66_gr4A). These DADs were observed in 69% (22 of 32) of the CPVT-hiPSCs-CMs studied (Figure 66_gr4C), with a mean amplitude of 13.0 ± 8.2 mV. In contrast, DADs were only seen in 11% (3 of 28; p < 0.01) of the healthy control hiPSCs-CMs ((Figure 66_gr4)B and Figure 66_gr4C) and their amplitude was significantly smaller (5.7 ± 3.9 mV; p < 0.01).

In the majority of CPVT-hiPSCs-CMs (17 of 22), afterdepolarizations were initiated during phase 4 of the AP (Figure 66_gr4A [left]). In the remaining 5 cells, afterdepolarizations were also noted during late phase 3 (Figure 66_gr4A [middle/right panels]). The severity of the afterdepolarizations in the CPVT-hiPSCs-CMs seemed to correlate with beating frequency (Figure 66_gr4D), and they could be identified both during the stimulation train ((Figure 66_gr5)A [top]) and immediately after termination of pacing (Figure 66_gr5A [bottom]). Finally, similar percentages of cells displaying afterdepolarizations (62% and 73%) were found in 2 different independently derived CPVT-hiPSCs clones, thus assuring that the aforementioned findings were not related to clonal variations.

Because arrhythmias in CPVT are related to exercise or emotional stress, we next evaluated the impact of activation of the adrenergic signaling pathway. Application of forskolin (5 μM), a direct activator of adenylate cyclase, significantly increased the frequency and amplitude of DADs in 83% (5 of 6) of the paced CPVT-hiPSCs-CMs studied and even led to the formation of triggered activity beats in 2 cells (Figure 66_gr5B [top]). We next studied quiescent CPVT-hiPSCs-CMs. Under baseline conditions, with cessation of pacing (Figure 66_gr5A [bottom]), low-amplitude DADs were generated at relatively long coupling intervals. Application of forskolin led to development of multiple DADs in these CPVT-hiPSCs-CMs (exhibiting higher amplitudes and shorter coupling intervals) and even to DAD-mediated triggered activity (Figure 66_gr5B [bottom]).

Application of the beta-agonist isoproterenol also had a proarrhythmic effect in the CPVT-hiPSCs-CMs. Thus, compared with baseline recordings (Figure 66_gr5C), isoproterenol (1 μM) administration resulted in the development of afterdepolarizations in 4 of the 6 cells studied, both during the stimulation train (1 Hz; Figure 66_gr5D), as well as in quiescent cells, after termination of pacing (Figure 66_gr5E). In the latter case, triggered activity was also noted.

A unique advantage of the hiPSCs technology is the potential to screen the effects of therapeutic interventions in a patient-specific manner. To test the feasibility of this concept, we evaluated the effects of flecainide, an antiarrhythmic agent suggested to display antiarrhythmic properties in CPVT (20). As shown in (Figure 66_gr5)F, flecainide 10 μM completely eliminated or significantly decreased afterdepolarizations in all of the patient-specific CPVT-hiPSCs-CMs studied (n = 6).

To verify the role of internal Ca²⁺ stores in the pathogenesis of the observed DADs, we treated the CPVT-hiPSCs-CMs with thapsigargin (10 μM), a specific inhibitor of the sarcoplasmic reticulum calcium ATPase pump. Blocking of Ca²⁺ reuptake into the SR gradually depletes intracellular Ca²⁺ stores. We next evaluated the ability of forskolin-treated cells to develop DADs before and after thapsigargin treatment. As shown in (Figure 66_gr5)G, thapsigargin significantly depressed postpacing afterdepolarizations, indicating the important role of internal Ca²⁺ stores in their formation.

To further evaluate calcium handling, we performed laser-confocal Ca²⁺-imaging of fluo4-loaded hiPSCs-CMs. As depicted in (Figure 66_gr6)A to (Figure 66_gr6)C, the CPVT-hiPSCs-CMs displayed significant whole-cell [Ca²⁺]_i transient abnormalities when compared with healthy control cells ((Figure 66_gr6)D to Figure 66_gr6E). These abnormalities were noted in 77% (23 of 30) of the CPVT-hiPSCs-CMs and were manifested by ≥1 of the following: 1) frequent local Ca²⁺ release events interspersed between the regular beats (Figure 66_gr6A); 2) development of large-storage release events (Figure 66_gr6B); and 3) broad double-humped transients (Figure 66_gr6C). In contrast, only 42% (14 of 41) of control cells displayed local Ca²⁺ release events (Figure 66_gr6E), and their amplitudes and durations were lower than those noted in the CPVT-cardiomyocytes. Importantly, none of the control cells developed the more severe whole-cell [Ca²⁺]_i transient irregularities.

In a similar manner to the electrophysiological recordings, adrenergic stimulation also worsened the degree of Ca²⁺ transient abnormalities in the CPVT-hiPSCs-CMCs. This finding was manifested by the development of new triggered activity events and/or broad double-humped transients in 60% (6 of 10) of the CPVT-hiPSCs-CMs after isoproterenol (1 μM) application (Figure 66_gr6F) and in 75% (6 of 8) of the cells after forskolin (5μM) administration. In contrast, none of the healthy control cells (n = 18 and 13, respectively) displayed such abnormalities following similar interventions (Figure 66_gr6G). Because beta-blockers represent the mainstay treatment for CPVT, we next evaluated their potential effects on the CPVT-hiPSCs-CMs. Application of propranolol (5 μM) reversed the observed abnormalities induced by isoproterenol in 78% (7 of 9) of the cells studied (Figure 66_gr6F).

We next evaluated the effect of Ca²⁺ overload on the degree of Ca²⁺ transient abnormalities in the hiPSCs-CMCs. To this end, the cells were perfused with increasing extracellular Ca²⁺ concentrations ([Ca²⁺]_o). As noted in (Figure 66_gr6)H, this resulted in significant worsening of the Ca²⁺ transient abnormalities (with development of large, wide, and double-humped transients) with elevated [Ca²⁺]_o in 75% (6 of 8) of the CPVT-hiPSCs-CMs evaluated.

It has long been appreciated that SR Ca²⁺ release can also occur in the absence of membrane depolarization through a mechanism referred to as SOICR (6). It was also suggested that some RyR2 mutations may markedly increase the occurrence of SOICR (6). To evaluate whether SOICR can be modeled by hiPSCs-CMs, we performed repeated Ca²⁺ imaging in tetrodotoxin-treated hiPSCs-CMs at increasing extracellular [Ca²⁺]_o concentrations (0.1 to 4 mM). As shown in (Figure 66_gr7)A, SOICR events could be identified in both the CPVT and healthy control hiPSC-CMs, with the number of Ca²⁺-oscillating cells increasing as function of [Ca²⁺]_o (Figure 66_gr7B). The threshold for generation of SOICR, however, was significantly altered by the M4109R RyR2 mutation. Consequently, the CPVT-hiPSCs-CMs began to develop SOICR at much lower [Ca²⁺]_o concentrations when compared with healthy control cells ((Figure 66_gr7)A and Figure 66_gr7B).

In the current study, we demonstrated the unique potential of hiPSCs for modeling inherited arrhythmogenic syndromes by establishing a patient-specific model of CPVT. By studying cardiomyocytes differentiated from the CPVT-hiPSCs, we were able to: 1) recapitulate the disease phenotype of the CPVT patient by showing the development of DADs in the patient-derived hiPSCs-CMs; 2) identify the role of internal Ca²⁺ stores in the pathogenesis of these afterdepolarizations; 3) reveal the contribution of the beta-adrenergic signaling pathway to these arrhythmias; 4) show the ability of the hiPSCs-CMs to develop SOICR with increasing Ca²⁺ concentrations and the significance of the M4109R RyR2 mutation in altering the threshold for such release events; and 5) demonstrate the potential of patient-specific CPVT-hiPSC-CMs for evaluating candidate drugs such as beta-blockers and flecainide.

Since the realization, more than a decade ago, that mutations in the cardiac ryanodine-receptor gene are responsible for CPVT1 (3), much has been learned about the disease pathogenesis. Experimental findings from heterologous expression systems, lipid bilayers, and knock-in and knockout mice models suggest that the basic mechanism underlying development of arrhythmias in this disorder involves a diastolic Ca⁺² leak from the mutant RyR2 (6). This may lead to development of DADs through activation of the membrane Na⁺/Ca²⁺ exchanger, which can eventually result in triggered activity; the pivotal arrhythmogenic mechanism in CPVT.

One of the challenges in confirming such a mechanism also clinically was the inability to directly study human cardiomyocytes from patients with CPVT. The advent of the hiPSCs technology may solve this cell-sourcing problem. This unique “disease-in-a-dish” concept is based on the capacity to derive isogenic hiPSCs from an affected individual and to coax their differentiation into the desired cell lineage (cardiomyocytes), thus portraying the patient-specific disease phenotype.

We have previously shown that hiPSCs-CMs express the necessary functional components associated with Ca²⁺ handling (15). This was demonstrated by development of spontaneous whole-cell [Ca²⁺]_i transients in the hiPSCs-CMs, which were dependent on both sarcolemmal Ca²⁺ entry via L-type Ca²⁺ channels and intracellular store Ca²⁺ release (15).

Here, using hiPSCs-CMs derived from a CPVT patient, we provide proof-of-concept evidence that the mechanism thought to underlie arrhythmogenicity in CPVT is also valid in the patient-specific cardiomyocytes. This was shown by the development of DADs in most CPVT-hiPSCs-CMs that became even more prominent after activation of the beta-adrenergic signaling pathway and could eventually give rise to triggered activity.

We next evaluated the role of intracellular Ca²⁺ stores in the formation of these DADs. By depleting internal Ca²⁺ stores with thapsigargin, we were able to suppress afterdepolarizations in the CPVT-hiPSCs-CMs, thus confirming the crucial role of these stores in DADs' pathogenesis. Moreover, in a similar manner to the electrophysiological recordings, laser-confocal Ca²⁺ imaging demonstrated significant Ca²⁺-handling irregularities in the CPVT-hiPSCs-CMs. These abnormalities, ranging from an increase in the incidence and magnitude of local Ca²⁺ release events to more severe global whole-cell Ca²⁺ irregularities and arrhythmogenesis, worsened with beta-adrenergic stimulation and Ca²⁺ overload.

It is believed that the diastolic Ca²⁺ wave that underlies the generation of DADs occurs when the SR Ca²⁺ content exceeds a threshold level. The mechanism for such a depolarization-independent SR Ca²⁺ release was termed SOICR (6), and this process was implicated in the arrhythmogenesis associated with intracellular Ca²⁺ overload, such as during digitalis intoxication. It was also suggested, in both heterologous expression systems (18) and in cardiomyocytes derived from a mouse model of CPVT (22), that several RyR2 mutations may markedly reduce the SR Ca²⁺ threshold required for SOICR. In the current study, we demonstrated the ability of the hiPSCs-CMs to recapitulate the SOICR phenomenon with the number of Ca²⁺-oscillating cells increasing with elevated Ca²⁺ concentrations. We also found that the M4109R RyR2 mutation decreases the threshold for such events, with SOICR occurring in the CPVT-hiPSCs-CMs at much lower Ca²⁺ concentrations then in healthy control cells.

Finally, feasibility studies were performed to demonstrate the potential of the CPVT-hiPSCs-CMs model for drug screening. These studies revealed that beta-blockers, which have been the mainstay treatment for CPVT, may also be beneficial in the patient-specific hiPSCs-CMs by reversing the Ca²⁺-handling irregularities observed after adrenergic stimulation. In addition to beta-blockers, potential new antiarrhythmic strategies for CPVT include agents targeting the abnormal diastolic Ca²⁺ leak (e.g., RyR2 stabilizers) (23) or those targeting the ionic mechanisms responsible for the generation of triggered activity (6). Flecainide was recently shown to possess antiarrhythmic properties in CPVT (20) through either a stabilizing effect on RyR2 (decreasing its opening probability) (20) or by increasing the threshold for triggered activity (through its Na⁺-channel blocking activity) (25). Although the current study was not designed to shed additional light on the specific antiarrhythmic mechanism of flecainide, it did reveal its ability to eliminate afterdepolarizations and triggered activity in the patient-derived hiPSCs-CMs. In addition, because this work involved a single CPVT patient, future studies will have to expand the results of this study to larger cohorts of patients.

Taken together, our data demonstrate the unique potential of hiPSCs-CMs for modeling CPVT. Using this approach, the disease phenotype presented by the patient at bedside could be readily recognized in the patient-derived cells in vitro as an anomalous electrophysiological signature. Importantly, this study also provided insights into arrhythmia mechanisms in this syndrome and highlighted the potential of the hiPSCs strategy for screening the effects of potential disease aggravators (adrenergic stimulation) and customized treatment options (beta-blockers and flecainide).