Electrophysiological abnormalities in induced pluripotent stem cell‐derived cardiomyocytes generated from Duchenne muscular dystrophy patients

Abstract Duchenne muscular dystrophy (DMD) is an X‐linked progressive muscle degenerative disease, caused by mutations in the dystrophin gene and resulting in death because of respiratory or cardiac failure. To investigate the cardiac cellular manifestation of DMD, we generated induced pluripotent stem cells (iPSCs) and iPSC‐derived cardiomyocytes (iPSC‐CMs) from two DMD patients: a male and female manifesting heterozygous carrier. Dystrophin mRNA and protein expression were analysed by qRT‐PCR, RNAseq, Western blot and immunofluorescence staining. For comprehensive electrophysiological analysis, current and voltage clamp were used to record transmembrane action potentials and ion currents, respectively. Microelectrode array was used to record extracellular electrograms. X‐inactive specific transcript (XIST) and dystrophin expression analyses revealed that female iPSCs underwent X chromosome reactivation (XCR) or erosion of X chromosome inactivation, which was maintained in female iPSC‐CMs displaying mixed X chromosome expression of wild type (WT) and mutated alleles. Both DMD female and male iPSC‐CMs presented low spontaneous firing rate, arrhythmias and prolonged action potential duration. DMD female iPSC‐CMs displayed increased beat rate variability (BRV). DMD male iPSC‐CMs manifested decreased I f density, and DMD female and male iPSC‐CMs showed increased I Ca,L density. Our findings demonstrate cellular mechanisms underlying electrophysiological abnormalities and cardiac arrhythmias in DMD.

potential duration. DMD female iPSC-CMs displayed increased beat rate variability (BRV). DMD male iPSC-CMs manifested decreased I f density, and DMD female and male iPSC-CMs showed increased I Ca,L density. Our findings demonstrate cellular mechanisms underlying electrophysiological abnormalities and cardiac arrhythmias in DMD.

K E Y W O R D S
arrhythmia, dilated cardiomyopathy, Duchenne muscular dystrophy, induced pluripotent stem cell-derived cardiomyocytes, X chromosome inactivation 1 | INTRODUCTION DMD caused by mutations in the dystrophin gene, is an X-linked muscle degenerative disease present in 1.7-2.1 of 10 000 male births. [1][2][3] Homozygous symptomatic females are rare and heterozygous females are mostly unaffected carriers, yet~20% are symptomatic manifesting carriers with later and milder disease progression than male patients. [4][5][6] In DMD patients dystrophin mutations lead to destabilization of the dystrophin glycoprotein complex (DGC) and sarcolemma instability resulting in functional deficits. 7 Dystrophin also plays a role in anchoring dystrophin-associated proteins to the sarcolemma, including nitric oxide synthase (nNOS). 8 Absence of functional dystrophin ultimately leads to degeneration and death of skeletal and cardiac muscle, being replaced by fibrous tissue. 9 DMD symptoms begin at early childhood, mainly including muscle fatigue. With age, the disease progresses and wheelchair assistance is needed in early teen years. Eventually, respiratory function is compromised and requiring mechanical support. Cardiac dysfunction usually manifests during late teenage or early twenties, but ultimately dilated cardiomyopathy (DCM) and congestive heart failure develop with ventricular and supraventricular arrhythmias.
Finally, patients progress to death because of respiratory or cardiac failure. 10 In female embryos, one of the X-chromosomes is randomly inactivated leading to mosaicism of maternal and paternal X-linked alleles explaining why heterozygous females are usually less affected or asymptomatic. iPSCs enable generation of human disease models to study underlying pathomechanisms in a cell type specific level. Previous studies investigating the effect of reprogramming on X chromosome status suggest three possible outcomes: (1) X chromosome inactivation (XCI) status is maintained throughout the reprogramming process; (2) inactivated X chromosome is reactivated; (3) inactivated X chromosome is eroded. [11][12][13][14][15][16][17][18] To investigate the electrophysiological abnormalities caused by dystrophin mutations in cardiomyocytes, we generated iPSCs from two DMD patients: a male and a female heterozygous manifesting carrier. X inactivation status and dystrophin gene expression were analysed in iPSCs and iPSC-CMs generated from healthy controls and patients of both sexes with regard to genetic, molecular and functional abnormalities. In support of the hypothesis that dystrophin-mutated iPSC-CMs from the mutant male and the heterozygous female carrier exhibit key features of DMD, mutated cardiomyocytes displayed molecular, genetic and electrophysiological abnormalities, including arrhythmias.

| Electrophysiological experiments
Action potentials were recorded from small cardiomyocyte clusters and single cells in whole-cell configuration. The pacemaker current I f was recorded from single cardiomyocytes (enzymatically dissociated) in the presence of 500 μM BaCl 2 to block I K1 . To record I f , the membrane was clamped at 15 seconds intervals, from a holding potential of −40 mV to −120 mV in 10 mV steps for 2 seconds pulse durations. 21 To record L-Type Ca 2+ current (I Ca,L ), the membrane was clamped at 200 ms depolarizing steps, from a holding potential of −70 mV ranging from −40 to +40 mV after a 20 ms −40 mV pulse for inactivation of Na + currents. 22 Axopatch 200B, Digidata 1322 or 1440 and pClamp10 (Molecular Devices, Sunnyvale, CA, USA) were used for data amplification, acquisition and analysis. For current recording, pipette capacitance compensation was adjusted, whole cell capacitance was measured and series resistance compensation was adjusted to 80%; finally, current density was calculated. Extracellular electrograms were recorded for 1000-1800 seconds from spontaneously contracting iPSC-CMs clusters by using the Micro Electrode Array (MEA) apparatus (Multi Channels Systems, Reutlingen, Germany).

| Statistical analysis
Results are presented as mean ± SEM. Comparisons between Male-DMD, Female-DMD and control iPSC-CMs were performed with one-way or two-way ANOVA followed by Holm-Sidak test using Sig-maPlot 12.0 software (Systat Software International, San Jose, CA, USA). A value of P < 0.05 was considered statistically significant.

| Clinical characterization of the DMD male patient and heterozygous symptomatic female patient
The male DMD patient suffered from muscular dystrophy from early childhood and was diagnosed with DCM at the age of 17. At age 25, he was hospitalized with respiratory insufficiency primarily because of combined respiratory and heart failure, and required tracheostomy and prolonged ventilation. At that time, his LVEF was 15% and the ECG showed sinus rhythm with narrow QRS and QE pattern in L1, AVL and V2-3. At age 30, he was respirator-dependent but his heart failure was reasonably controlled. tachycardia. An ICD was implanted, but despite heart failure therapy, her condition continued to deteriorate. Cardio-respiratory exercise test showed VO 2 max of 6 ml/kg/min (33% of normal value) resulting from primarily cardiovascular limitation, but the patient declined heart transplantation. At age 49, she was in NYHA IV. AV nodal ablation and upgrading to CRTD was required because of atrial fibrillation with rapid ventricular response. Echo-Doppler showed severe biventricular dysfunction with LVEF of 20% and severe tricuspid regurgitation. The muscle weakness progressed to complete inability to walk. The patient expired at age 51 because of end-stage heart failure associated with renal insufficiency.

| Molecular characterization of DMD iPSCs and iPSC-CMs
iPSCs were generated from dermal fibroblast. For iPSCs characterization, FACS and immunostaining validated pluripotent markers expression ( Figures S1 and S2). Teratoma assay demonstrated iPSCs ability to differentiate into the three germ layers ( Figure S1). Karyotype was preserved throughout the reprogramming process, and Sanger sequencing confirmed the mutations in the patients iPSCs ( Figure S1). For control, iPSCs and iPSC-CMs were generated from four different healthy female and male donors with varying genetic backgrounds, thus markedly strengthening the comparisons between the mutated and control populations.

| Genetic characterization of the female manifesting carrier
Previous studies suggested an inactive X chromosome may undergo reactivation (XCR) or erosion during human iPSCs (hiPSCs) derivation [23][24][25] while others proposed that XCI is maintained throughout the reprograming process. 26 Given that hearts of females carrying dystrophin mutations in humans and mice show random XCI and mosaic cellular dystrophin expression, 27 we aimed to evaluate the X chromosome status during reprogramming and differentiation. XIST is a long non-coding RNA on the X chromosome that acts as a major effector of the X inactivation process. 17,28 Expression analysis demonstrated the absence of XIST RNA in iPSCs, indicating that XCR or erosion of XCI occurred during reprogramming of DMD and control female dermal fibroblasts. In agreement with previous reports, 24,25,29 XIST expression patterns were maintained during differentiation of iPSCs into iPSC-CMs ( Figure 1G). As XIST does not fully correlate with XCI in iPSCs, 17,30 Figure 1E).  Figure 4D).

| Arrhythmias in DMD iPSC-CMs
In addition to the slow firing rate, DMD female and male iPSC-CMs
To neutralize the effect of slow firing rate in DMD cardiomyocytes on APD, we measured action potential parameters in paced single cell cardiomyocytes. In agreement with the results from spontaneously firing cardiomyocytes, APD 20 /APD 50 /APD 90 were prolonged in DMD female and male iPSC-CMs ( Figure 5A, 5-E), while other parameters were comparable to control ( Figure 5B). To identify the specific action potential phase responsible for APD prolongation, we calculated the ratio between APD at different stages of repolarization and APD 90 ; the APD contributing mostly to prolongation was APD 20 , implying involvement of early repolarization phase (ie APD 20 ; Figure 5B).

| Increased I Ca,L density in DMD iPSC-CMs
To determine the mechanism underlying APD 20 prolongation, we measured I Ca,L , the major depolarizing current in early repolarization. 33 As illustrated by the representative current traces ( Figure 6A) and mean I-V relations ( Figure 6B), I Ca,L density was larger in both DMD female and male than in control iPSC-CMs (female-at voltage steps from 0 mV to −20 mV and male-at 0 mV). No differences were observed in the current activation curve among the three groups ( Figure 6C). In agreement with larger I Ca,L density, qRT-PCR analysis demonstrated increased expression of the Ca 2+ voltagegated channel subunit alpha1 C (CACNA1C) in DMD female and male compared to control iPSC-CMs ( Figure 6D).  Figure S5).

| DISCUSSION
In support of the hypothesis that dystrophin-mutated iPSC-CMs

| Decreased automaticity and altered I f density
Contrary to DMD patients who usually manifest increased high heart rate, 32 DMD iPSC-CMs display low spontaneous firing rate Duplicates/triplicates of independent differentiations were used for each clone: control, DMD female and DMD male; n = 3 for each experimental condition. One-way ANOVA followed by Holm-Sidak post-hoc analysis. *P < 0.05, ***P < 0.001 compared to control. However, in DMD patients, increased heart rate is attributed to elevated sympathetic tone 36 ; hence, as this masking effect is absent in DMD iPSC-CMs, the in vitro study exposes a fundamental pathophysiology of the pacemaker function -attenuated automaticity. To elucidate the mechanism underlying this reduced automaticity in the male and female cardiomyocytes, we recorded I f , a major pacemaker current. 21 In agreement with the reduced automaticity in DMD male cardiomyocytes, I f density was smaller than control, albeit similar mRNA expression levels of HCNs.
In DMD female iPSC-CMs I f density was similar to control, while mRNA expression levels of HCNs were significantly increased. Overall, both DMD female and male iPSC-CMs demonstrate reduced I f density relatively to their respective mRNA expression levels of HCNs. The precise mechanism involved in altered HCN function is yet to be determined. No differences were observed in current activation between DMD and control iPSC-CMs.

| Arrhythmias including DADs and OPPs
In contrast to regular firing pattern in control iPSC-CMs, 52% and 17% of DMD female and male cardiomyocytes exhibited arrhythmias including DADs and OPPs, respectively. The common cause for DADs is intracellular Ca 2+ overload leading to Na + /Ca 2+ exchanger activation resulting in transient inward depolarizing current (I Ti ) and DADs initiation. 37 The proposed mechanisms underlying Ca 2+ over- (C-E) Action potential duration at 20/50/90% of repolarization (APD20/50/90); (F) Maximal diastolic potential (MDP); (G) Action potential peak; (H) Maximal rate of phase 0 depolarization (dV/dtmax). Two-way ANOVA followed 5 by Holm-Sidak post-hoc analysis. *P < 0.05, **P < 0.01, ***P < 0.001 ryanodine receptors (RyRs) located on the sarcoplasmic reticulum (SR), resulting in SR Ca 2+ leakage into the cytosol. 38 Notably, leaky RyRs, which can lead to Ca 2+ overload, were reported in mdx mice. 39 Furthermore, I Ca,L was increased in DMD iPSC-CMs compared to control iPSC-CMs which may contribute to Ca 2+ overload. The phenomenon of DADs and OPPs in DMD iPSC-CMs is in agreement with increased Ca 2+ influx and Ca 2+ overload, which can contribute to cardiac arrhythmias in DMD patients. 40

| Prolonged APD and enhanced I Ca,L density
Importantly, the prolonged APD in female and male DMD cardiomyocytes, potentially associated with fatal arrhythmias, 41 may account for clinical phenotypes observed in DMD patients, such as long QT interval (LQT). 42 The ratio between the changes in APD 20 and APD 90 strongly implies that most of the prolongation stems from early repolarization phase. As I Ca,L is the major depolarizing current ; DMD female, n = 11; DMD male, n = 12. One-way ANOVA followed by Holm-Sidak posthoc analysis. *P < 0.05 during early repolarization, 33 we measured its density which was higher in both DMD female and male iPSC-CMs. These results are consistent with previous findings of enhanced I Ca,L in adult cardiomyocytes of mdx mice. 43 Furthermore, CACNA1C expression was increased in both DMD female and male iPSC-CMs compared to control.

| Increased BRV in DMD female iPSC-CMs
DMD female (but not male) iPSC-CMs displayed increased BRV measures compared to control cardiomyocytes. While further investigation is required to determine the contribution of the mixed dystrophin expression to this finding, it is likely that heterogeneity of the cardiomyocytes expressing either WT or mutated dystrophin alleles, contributes to increased BRV indices in action potential recordings from iPSC-CM clusters.

| Advantages of DMD iPSC-CMs over the mdx mouse model
The common murine model for DMD research is the mdx mouse which carries a point mutation in exon 23 of the dystrophin gene. 43 However, the similarity of mdx hearts to human DMD hearts is debated; while mice initially develop hypertrophic cardiomyopathy and only at a later stage contractile dysfunction and systolic heart failure, human cardiomyopathy is characterized primarily by ventricular dilatation and reduced systolic function. 44 The use of patients' iPSC-CMs has been previously demonstrated as valid cellular in vitro model for various genetic heart diseases including DMD, [45][46][47][48] as well for LQT syndrome, 41 and PRKAG2 myopathy. 49 Furthermore, iPSC-CMs are an accessible human cellular model which also enables exclusion of in vivo masking factors such as the influence of the autonomous nervous system.

| Summary and conclusions
In conclusion, this study demonstrates that loss of dystrophin in patients' iPSC-CMs is sufficient to cause electrophysiological abnormalities including low spontaneous firing rate, arrhythmogenic firing patterns, abnormal action potential parameters, increased BRV and altered pacemaker currents density. Importantly, this is the first study demonstrating the impact of mixed dystrophin expression pattern on abnormal characteristics and functional abnormalities of DMD iPSC-CMs generated from a DMD female manifesting carrier.
These results may suggest greater predisposition to arrhythmias in early stages of DMD cardiomyopathy in females. Further studies are therefore required to characterize the natural history and define the arrhythmic risk in female patients with DMD mutation. This study also has implications for genetic therapies aimed at restoring dystrophin expression by gene editing or gene therapy; if genetic correction is incomplete, a subset of dystrophin deficient myocytes may increase electrophysiological heterogeneity which could be a substrate for arrythmogenesis.

ACKNOWLEDG EMENTS
We thank Mrs. Margarita Shulman and Martina Kasten for excellent technical assistance. FACS was performed in the Frankel Cardiovascular Regeneration Core Laboratory at the University of Michigan.

CONFLI CT OF INTEREST
The authors have nothing to declare.