Rapid Electrical Stimulation Increased Cardiac Apoptosis Through Disturbance of Calcium Homeostasis and Mitochondrial Dysfunction in Human Induced Pluripotent Stem Cell-Derived Cardiomyocytes

Tachycardia is the most common arrhythmia, which can cause many negative outcomes, including heart failure or even sudden cardiac death (SCD) in some patients [1]. Although various treatment procedures have been undertaken to terminate tachycardia, including drug prescription and ablations [2], several studies have claimed that some patients still suffer higher risks of congestive heart failure, recurrence of tachycardia, and SCD even after the termination of tachycardia, indicating some remained damages induced by tachycardia [3, 4].

Despite the fact that several animal models have been created to investigate the mechanisms by which tachycardia cause heart dysfunction [5-7], little is known about the changes occurring at the cellular level and the underlying mechanisms. To handle this issue, several researchers have applied rapid electrical stimulation (RES) on cardiomyocytes. Due to the difficulties in obtaining human adult cardiomyocytes, cardiomyocytes used in these trials were primarily isolated from rats [8] and mice [9, 10]. However, considering the significant species differences and the inherent defect of isolated cells wherein they cannot be cultivated long-term, it remains debatable whether these observations are applicable to human beings.

Human induced pluripotent stem cell-derived cardiomyocytes (hiPSC-CMs) have been obtained and applied in many studies [11, 12], and have been demonstrated the similar properties to human myocardia in several aspects, including ion channel gene expressions [13, 14], sarcoplasmic reticulum (SR) calcium handling [15, 16], etc. Based on the abovementioned hurdles and the advantages of regenerative cardiomyocytes, we sought to investigate the changes brought about by tachycardia and the underlining mechanisms by applying RES on hiPSC-CMs.

Herein, compared to the no electrical stimulation (NES) control group and low-frequency electrical stimulation (LES) control group, RES disturbed calcium homeostasis and activated the mitochondrial apoptotic pathway(MAP), which may lead to cardiomyocyte loss and systolic dysfunction. Phosphorylated Ca²⁺/calmodulin-dependent protein kinase II (p-CaMKII) was upregulated following RES, and inhibition of CaMKII partly reversed the adverse effects of RES, including disturbed calcium transients, decreased L-type calcium currents and cardiomyocyte apoptosis. Therefore, CaMKII could potentially be chosen as a supplemental therapeutic target for tachycardia patients.

Human iPSC line NC5 cells (Help Stem Cell Innovations, NC2001) were seeded onto Matrigel (Corning, Corning, NY, USA, 354277) coated 6-well plates and cultured in the mTeSR^TM medium (StemCell Technologies, Vancouver, Canada, 05850) to 90% cell confluency. Next, NC5 hiPSCs were differentiated to cardiomyocytes as previously described by our group [17]. Eight weeks following the differentiation, cells were digested with 0.25% trypsin-EDTA (Life Technologies) and resuspended into single-cell suspension and reseeded on gelatin-coated plates for further study.

hiPSC-CMs were reseeded onto 6-well plates at appropriate concentrations. Electrical stimulation was conducted using a CDish100^TM culture dish and C-Pace100^TM culture pacer (IonOptix Corporation, Netherlands). For RES, 3 V, 0.5 ms and 3 Hz of electrical pulses were selected. Two subgroups—8 h/day (S-8h) and 24 h/day (S-24h)—were set to represent short-term and long-term tachycardia. To examine whether electrical stimulation itself can bring about any changes to cardiomyocytes, stimulation at 1 Hz frequency for 24 h/day was applied to a group of cells as the low electrical stimulation (LES) control group. Age-matched cardiomyocytes without electrical stimulation were used as no electrical stimulation (NES) controls. After 7 days of stimulation, cardiomyocytes were harvested for functional assessments.

Mitochondrial membrane potential (MMP) was measured with the fluorescent indicator JC-1 (Beyotime Institute of Biotechnology, China, C2006). In normal cells, JC-1 forms aggregates which fluoresce with an emission peak at 588 nm. Decreased membrane potential favors the monomers of JC-1, which has an emission peak at 530 nm. Cardiomyocytes were harvested using 0.25% trypsin. Approximately 100, 000 cells were suspended in 0.5 mL medium in each tube for JC-1 staining. Staining working solution was prepared following the manufacturer’s instructions. Five hundred microliters of the working solution were added into each tube and cardiomyocytes were incubated for 20 min at 37°C in the dark. Following incubation, cells were washed with JC-1 staining buffer twice and resuspended for flow cytometry analysis. The wavelengths of excitation and emission were 514 nm/529 nm for detection of the monomers form of JC-1, and 585 nm/590 nm for detection of the aggregates through FL-1 and FL-2 channels.

Hoechst staining (Beyotime Institute of Biotechnology, China, C0003) was used to evaluate morphological features of cardiomyocytes. Cells were washed with cold DPBS and fixed in prepared 4% PFA for 30 min. After fixation, cells were washed with cold DPBS twice and incubated with 5 µg/ml Hoechst at 37°C for 15 min in the dark. Following incubation, cells were washed again and apoptotic cells were identified using a fluorescence microscope (Axio Vert.A1, Zeiss). Normal cells presented homogeneous blue chromatin, while apoptotic cells were bright blue condensed or fragmented.

Annexin V/propidium iodide (AV/PI) staining was used, according to manufacturer’s instructions, to evaluate cell apoptosis. Briefly, approximately 1× 10⁶ cells were harvested with 0.25% trypsin without EDTA and resuspended in 300 µL binding buffer. Next, cells were incubated with AV/PI (Vazyme Biotech Company, China, A211-02) for 20 min in the dark and analyzed with a flow cytometer (BD FACS Calibur) within 1 hour. Data from 8, 000 cells/sample were collected, and quadrants were set according to viable, unstained populations in blank samples.

Following the manufacture’s instruction, Fluo-3 AM (Life technology, F1242) was used to measure calcium transients. In brief, 5 mM Fluo-3 AM in DMSO stock and equal volume of 20% (w/v) Pluoronic F-127 were diluted to 5 µM using culture medium to obtain the working solution. Cells were washed with DPBS twice before being incubated with the working solution for 90 min at 37°C. After incubation, cells were washed to remove any remaining dye and then incubated with normal culture medium for another 30 min. Next, cells were washed again and medium was changed to Tyrode’s solution. Fluo-3 AM in the cardiomyocytes was excited at a light wavelength of 488 nm and the emission was recorded at 520 nm using a E3CMOS camera (Touptek, China). The Ca²⁺ transients are expressed as ΔF/F₀, where ΔF is the change in fluorescence intensity relative to the minimal fluorescence (F₀) measured at each Ca²⁺ transient.

Whole-cell patch clamp was used to record the L-type calcium currents (I_Ca,L) of cardiomyocytes on the Axopatch 200B amplifier (Axon Instruments, CA, USA) at 37°C, as previously described [17]. Patch pipettes were prepared with a micropipette puller (P-1000, Sutter Instrument, Novato, CA, USA) and had a typical resistance of 3-6 MΩ. I_Ca,L was examined using the following intracellular solution: 130 mM CsCl, 20 mM TEA-Cl, 1.8 mM MgCl₂, 5 mM Na₂ATP, 5 mM EGTA, 10 mM glucose, and 10 mM HEPES (pH 7.3 with CsOH). The external bath solution composition was as follows: 120 mM choline chloride, 20 mM CsCl, 1 mM MgCl_2, 1.8 mM CaCl₂, 10 mM glucose, and 10 mM HEPES (pH 7.3 with CsOH). The program for recording I_Ca,L was from a holding potential of –40 mV to measure potentials from –40 to +60 mV, every 10 mV for 500 ms. Cell capacitance was used to normalize I_Ca,L and current density were expressed as pA/pF. Data were analyzed using Clampfit 10.6 (Axon Instruments).

Total RNA was extracted with the TRIzol Reagent (Invitrogen) following the manufacturer’s instructions. Amounting to 500 ng RNA was used for the synthesis of cDNA using a SuperScript Reverse Transcriptase (Bio-Rad, Hercules, CA, USA, 170-8891). qRT-PCR was performed using the AB 7900HT Real-Time PCR System (ABI, USA) with SYBR Green (Bio-Rad, 170-8882AP) in a total reaction series of 10 µL mixture. mRNA levels were based on the critical threshold (Ct) value and were normalized to Glyceraldehyde-3-phosphate dehydrogenase (GAPDH). The primers used in this work (for all online suppl. material, see www. karger.com/doi/10.1159/000490213) are listed in Table S1.

Western blotting was conducted as previously described by our group [17]. The primary antibodies used in the present study were as follows: GAPDH (1: 1000, Cell Signaling, 2118S), Bcl-2 (1: 1000, Cell Signaling, 3498), Caspase-3 (1: 1000, Cell Signaling, 9662), Bax (1: 1000, Cell Signaling, 2772), Serca2a (1: 1000, Abcam, ab2861), Phospholamban (1: 1000, Abcam, ab15000), p-CaMKII (1: 1000, Cell Signaling, 12716T), and Calsequestrin-2 (1: 1000, Abcam, ab108289). Secondary antibodies employed in this work included anti-rabbit IgG antibody (1: 5000, Cell Signaling, 7074P2) or anti-mouse IgG antibody (1: 5000, Cell Signaling, 7076).

Immunofluorescent staining was performed as previously described by our group [17]. Approximately 30, 000 hiPSC-CMs were reseeded onto gelatin-coated 24mm coverslips and electrical stimulation began at different parameters 3 days after the seeding. After the stimulation, cardiomyocytes were fixed and permeabilized before being blocked with 5% BSA. After the incubation with primary antibody cTnT (1: 200, Abcam, ab8295) at 4°C overnight, cells were washed and incubated with fluorescent-conjugated secondary antibody (1: 500, Abcam, ab150077) at room temperature in the dark for 1 hour. Nuclei were stained with DAPI (Life Technologies, P36931) for 10 min at room temperature before observed and captured by a Zeiss fluorescence microscope (Axio Imager A2, Zeiss).

All results were expressed as the Means ± standard deviation (SD). SPSS Statistics 19.0 (IBM Corp., Armonk, NY, USA) was used for Statistical analysis. One-way analysis of variance (ANOVA) was performed for testing significant differences among experimental groups. A p value less than 0.05 was considered statistically significant.

It was accepted that tachycardia could alter the mitochondrial metabolism [18, 19]. Recently, Karin A.L. Mueller et al. [20] reported a disturbed mitochondrion distribution in cardiomyocytes from tachycardia patient biopsies. Therefore, we first examined mitochondrial function at the cellular level. Mitochondrial membrane potential (MMP) is an indicator of mitochondrial function, and loss of MMP often suggests mitochondrial dysfunction. Flow cytometry analysis of JC-1 staining demonstrated a significantly decreased fluorescence intensity of aggregates and an increased fluorescence intensity of monomers (Fig. 1A). The ratio of monomers and aggregates significantly increased from 0.45±0.06 in the NES control group to 1.21±0.24 (p< 0.01) in S-8 h group and 2.17±0.32 (p< 0.001) in S-24 h group, indicating a dramatic loss in MMP and resultant mitochondrial dysfunction in the rapid stimulation groups (Fig. 1B). There was no significant increase in the LES control group (0.49±0.07, p=0.53) compared to the NES control group.

Biopsies from tachycardia patients [21, 22] and animal models [23] showed decreased cardiomyocytes and increased fibrotic tissues, but little is known about the underlying mechanisms. Considering that our previous study demonstrated increased apoptosis in HL-1 cells after rapid pacing [10], we next examined the morphological features of cell nuclei using Hoechst staining. Cardiomyocytes in S-8 h group and S-24 h group exhibited the typical morphological characteristics of apoptosis including condensed or fragmented nuclei after 7 days of RES compared to the NES control group (Fig. 2A). In order to further quantify the apoptotic rates, flow cytometry analysis of apoptosis was performed using the AV/PI staining (Fig. 2B). Similar to the results of Hoechst staining, RES significantly increased the apoptotic cell percentage in a time-dependent manner (7.69±1.22%, 13.56±1.74%, 31.85±3.27% for NES control, S-8 h and S-24 h groups, respectively, p< 0.001). In contrast, there was no significant difference between cardiomyocytes in the NES control group and the LES control group (8.06±0.86%, p=0.64).

Since several studies have emphasized the relationship between mitochondrial dysfunction and cell apoptosis [24, 25], we subsequently examined the gene expression of several molecules in MAP, including caspase3, anti-apoptotic protein Bcl-2 and pro-apoptotic protein Bax by western blotting and qRT-PCR. Fig. 2C shows the representative blots images of apoptotic-related proteins. Consistent with previous studies, the ratio of Bax/Bcl-2 was significantly higher in S-8 h group (0.41±0.07, p< 0.05) and S-24 h group (1.13±0.12, p< 0.001) (Fig. 2D) compared to the NES controls (0.23±0.04). Protein levels of cleaved caspase3 were also significantly higher in S-8 h group (1.53±0.14, p< 0.001) and S-24 h group (1.85±0.15, p< 0.001) compared to those in the NES control group (1.05±0.14, relative to GAPDH) (Fig. 2E). Gene expressions of these molecules were further examined and similar results were obtained: CASPASE3 and BAX were significantly upregulated in the RES groups, while BCL was significantly downregulated (Fig. 2F). Accordingly, RES may elicit damages to mitochondria and promote apoptosis, which may result in cardiomyocyte loss.

Several previous studies have claimed that cytoplasmic calcium has an important impact on mitochondrial condition [26-28]. Considering the mitochondrial dysfunction in our study and the importance of calcium to cardiomyocytes, we next examined calcium transients in cardiomyocytes with and without electrical stimulation using Fluo3-AM staining. Fig. 3A shows the typical dynamic evolution of calcium in each group, and the peak value of each curve represents the maximal increase in calcium relative to baseline. Compared to the NES control group, amplitudes of calcium transients in the RES groups were significantly decreased (0.36±0.04, 0.18±0.02, 0.10±0.02 for NES control, S-8 h and S-24 h groups, respectively, p< 0.001, Fig. 3B).

We next recorded L-type calcium currents (I_Ca,L) using patch clamp under voltage clamp mode. Typical traces of I_Ca,L and I-V curves of cardiomyocytes in each group are shown in Fig. 4A and 4B; currents were normalized to the capacitance and were expressed in pA/pF as current density. As shown in Fig. 4C, I_Ca,L amplitudes in RES groups dramatically decreased compared to the values observed in NES controls (-9.02±0.66 pA/pF, -2.88±0.26 pA/pF, -2.09±0.39 pA/pF for the NES control, S-8 h and S-24 h groups, respectively, p< 0.001). There was no significant difference between the NES control and LES control groups (-8.56±0.42 pA/pF, p=0.29).

Cytoplasmic calcium is maintained by a balance between the release of Ca²⁺ into the cytosol and the removal of Ca²⁺ by the associated action of some proteins in the cell. Ca²⁺ entry is facilitated by the voltage-sensitive L-type Ca²⁺ channels, which triggers the release of Ca²⁺ from SR through ryanodine receptors (RyR), raising cytoplasmic Ca²⁺ and causing contraction, known as calcium induced calcium release (CICR). Following cardiac contraction, Ca²⁺ is removed from the cytosol primarily in 3 ways: most of it is re-sequenced to the SR via the sarco-endoplasmic reticulum ATPase (Serca), some of it is extruded into the extracellular space through the Na⁺-Ca²⁺ exchanger (NCX) or Ca²⁺ ATPase pump, and some enters the mitochondria. In the present study, disrupted cytoplasmic calcium and decreased L-type calcium currents indicated disturbed calcium homeostasis. Herein, we then investigated the expression of calcium homeostasis-related proteins and genes using western blotting and qRT-PCR. We first examined the expression of Serca2a and its regulatory protein phospholamban (Pln), which interacts with Serca2a and promotes its pumping activity [29]; the representative blot image of Serca2a and Pln is shown in Fig. 5A. There was a significant reduction in the expression of Serca2a in S-24 h group (p< 0.05) compared to that in the NES control group; however, no significant reduction in expressions were observed among other groups (Fig. 5B). Cardiomyocytes in RES groups demonstrated significantly lower levels of Pln, which may inhibit the pumping function of Serca2a and the re-sequencing of cytoplasmic calcium. Calsequestrin2 (Casq2), a large-capacity Ca-binding protein which locates in the terminal cisternae of SR, plays a role as a buffer of free Ca²⁺ in the SR and maintains a relatively lower concentration of free Ca²⁺ within the SR, thereby dynamically mediating Ca²⁺ re-sequencing [30]. As seen in Fig. 5A and 5B, protein levels of Casq2 were dramatically decreased in S-8 h (p< 0.05) and S-24 h groups (p< 0.01), whereas no decrease was observed in LES control group. These results were further confirmed by qRT-PCR (Fig. 5C).

CaMKII is a ubiquitous kinase which is canonically activated by the increase of intracellular Ca²⁺. CaMKII has been reported to be activated in heart failure animal models [31] and patients [32] and is strongly associated with mitochondrial stress [33]. Activation of CaMKII has also been demonstrated to be a common intermediate of variable death stimuli which induce apoptosis in cardiomyocytes [34]. Hashambhoy YL et al. [35] reported that CaMKII is activated by excessive calcium, and overactivation of CaMKII increases CaMKII-dependent phosphorylation of RyR2 and thus increases the diastolic SR Ca²⁺ leak, which may further increase cytoplasmic calcium and form a vicious cycle. It was demonstrated in our study that the expression of phosphorylated CaMKII was significantly upregulated following RES (Fig. 5).

Considering that CaMKII inhibitors have been applied for heart failure patients [36, 37], we prescribed 1 µM KN93 based on published studies [38, 39] on S-24 h group cells during RES. As shown in Fig. 6, following KN93 treatment, the maximal ΔF/F₀ values of calcium transients were significantly increased to 0.244±0.028 (p< 0.001), indicating that disruption of cytoplasmic calcium was attenuated. Afterwards, L-type calcium currents were also recorded. I_Ca,L was significantly increased from -2.04±0.48 pA/pF in S-24 h group to -4.06±1.02 pA/pF in S-24 h+KN93 group (p< 0.001, see online suppl. material, Fig. S1); however, the currents in the KN93-treated group were still significantly smaller than those in the NES control group (-8.30±0.85 pA/pF, p< 0.001, see online suppl. material, Fig. S1).

After confirming the beneficial effects of KN93 on reversing disturbed calcium homeostasis, we investigated whether it also attenuated mitochondrial dysfunction and cardiomyocyte apoptosis induced by RES. As shown in Fig. 7A, compared to the S-24 h group, MMP loss of hiPSC-CMs was significantly attenuated in the KN93-treated cardiomyocytes. The ratio of monomers to aggregates decreased from 2.12±0.42 in S-24 h group to 1.06±0.20 in S-24 h +KN93 group (p< 0.01); however, the ratio in the KN93-treated group was still significantly higher compared to that of the NES control group (0.46±0.10, p< 0.01), indicating that mitochondrial dysfunction was partly attenuated. Apoptosis in hiPSC-CMs was also examined using Hoechst staining and AV/PI flow cytometry. As shown (see online suppl. material) in Fig. S2 and Fig. 7B, the percentage of apoptotic cells was decreased from 26.82%±3.19% in S-24 h group to 12.93%±1.60% in S-24 h +KN93 group (p< 0.001); however, the percentage in the KN93-treated group was still higher than the NES controls (8.58%±2.28%, p< 0.01). We further examined the expression of Caspase3, Bax and Bcl-2. As shown in Fig. 7C-F, all the changes caused by RES were significantly attenuated after treatment with KN93; however, cells did not fully recover to control levels, indicating that some damages caused by RES were irreversible.

Tachycardia is the most common arrhythmia in clinical practice and has become a huge burden for patients and society in terms of both economics and health. Heart failure induced by tachycardia has been increasingly observed; however, its mechanism remains unclear [1-3]. Many clinicians hold the view that damages induced by tachycardia can be reversed following the termination of tachycardia, but there have been reports of recurrence of tachycardia and sudden cardiac death even after the treatment of tachycardia, which indicates an incomplete recovery [3, 4]. Previously, many efforts have been made to investigate the mechanisms of tachycardia-induced cardiac damages in animal models. Paslawska U et al. [5] observed that swine, after 170 bpm pacing, presented reduced exercise capacity, LV dilatation, impaired LV systolic function and interstitial fibrosis; Kusunose K et al. [7] reported decreased systolic LV strain, loss of early diastolic recoil, and twist deformation. Apart from animal models, there were also various studies investigating tachycardia at the cellular level. Our previous study demonstrated that HL-1 cells, after rapid pacing, presented a greater extent of apoptosis, and the ER stress-mediated MAP and MAPK pathways participated in this process [10]. Using myocytes isolated from Wistar rats, Sepúlveda M et al. [8] discovered that rapid pacing induced CaMKII activation and myocyte apoptosis in a cascade reaction of ROS. However, the precise changes in calcium homeostasis remained unclear. Despite the findings in these studies, an inevitable hurdle is the species differences between animals and human beings. Moreover, cardiomyocytes in these studies were stimulated for 24 hours or even 1 hour, which is an extremely short time for mimicking clinical tachycardia. Owing to the great advances of regenerative medicine, we tried to mimic clinical tachycardia by applying 180 bpm RES on hiPSC-CMs for 1 week. We also divided the cells into two subgroups, S-8 h and S-24 h, to represent short-term and long-term tachycardia. A negative control group receiving 1 Hz stimulation was also established to define whether electrical stimulation itself can bring any changes to cardiomyocytes.

Previously, it was suggested that electrical stimulation promotes cardiac differentiation and maturation of hiPSC-derived cardiomyocytes [40-42]. Ruan et al. [41] reported that electrical stimulation promotes force maturation of induced pluripotent stem cell-derived cardiac tissue; however, the cardiomyocytes used in this study were only 2 or 3 weeks old. Damián Hernández et al. [42] reported that electrical stimulation promoted cardiac differentiation; however, the parameters of electrical stimulation were set at 1 Hz and for a very short period of approximately 15 min. The effects of stimulation when extended to a longer time or on older cardiomyocytes remain unknown. Consistent with previous findings [43, 44] that long-term cultivation promotes maturation of hiPSC-CMs, immunostaining of the cardiac marker cTnT presented clear and organized sarcomeres (see online suppl. material, Fig. S3), which indicated that 8-week-old cardiomyocytes used in our work were relatively mature, and the effects of applying RES on these cells have not been reported.

In our study, amplitudes of calcium transients and L-type calcium currents were significantly decreased following RES in a time-dependent manner (Fig. 3, 4). Genes related to the re-sequencing of calcium including SERCA, CASQ, and PLN were significantly downregulated at both the protein and mRNA levels following RES, these changes disturbed the normal decrease of free calcium in cytoplasm and may explain the disrupted cytoplasmic calcium occurred in the present study. Significant loss in MMP of mitochondria and increased cardiomyocyte apoptosis were observed following RES. Caspase3 and pro-apoptotic Bax were significantly upregulated in RES groups, while anti-apoptotic Bcl-2 was correspondingly downregulated, suggesting the activation of the MAP.

Previously, several studies have reported the relationship between cellular calcium and mitochondria. Vasington and Murphy [45] first demonstrated, at the beginning of the 1960s, that Ca²⁺ is accumulated by the mitochondria. Santulli G et al. [46] further reported that increased cytoplasmic calcium concentration caused mitochondrial Ca²⁺ overloading and dysfunction in a murine model of post-myocardial infarction heart failure. Ca²⁺ accumulation impairs mitochondrial function, which resulted in reduced ATP production and enhanced release of reactive oxygen species (ROS) [47], and ROS production is closely involved in heart failure [48]. Based on these studies and our present work, we believe that RES-induced cytoplasmic calcium disruption may cause mitochondrial dysfunction and activate MAP, which may lead to cardiomyocyte apoptosis and cell loss.

CaMKII, which can be activated by elevated cytoplasmic calcium, is a critical molecule involved in excitation-contraction coupling and calcium homeostasis. It has been previously reported that increased concentration of cytoplasmic Ca²⁺ triggered cell death in several experimental models by activating CaMKII [31, 32]. Upregulated CaMKII phosphorylation can also increase CaMKII-dependent phosphorylation of RyR2 and cause a diastolic SR Ca²⁺ leak and further increase cytoplasmic calcium, thus creating a vicious cycle [35]. Breaking this cycle by inhibition of CaMKII has been reported to benefit heart failure animals and patients [36, 37]. Consistently, KN93, a CaMKII inhibitor, used in our study significantly attenuated adverse effects caused by RES, including calcium transients and L-type calcium currents. Moreover, MMP was increased to levels comparable to that observed in NES controls, meanwhile apoptotic cell rates were significantly decreased. MAP molecules, Caspase3 and Bax were observed to be downregulated compared to their levels in S-24 h group, while Bcl-2 was upregulated. However, these recoveries were incomplete compared to the levels observed in the NES control group, indicating that some damages were irreversible.

Taking together, we demonstrated that RES disturbed calcium homeostasis and resulted in mitochondrial dysfunction and promoted cell apoptosis. Phosphorylated CaMKII was upregulated under RES and inhibition of CaMKII with KN93 partly attenuate the adverse effects of RES.

In summary, as shown in the illustrated Fig. (Fig. 8), rapid electrical stimulation on hiPSC-CMs disturbed calcium homeostasis, which led to mitochondrial stress and promoted cell apoptosis and caused electrophysiological remodeling in a time-dependent manner. CaMKII played a central role in the damages induced by RES, and pharmacological inhibition of CaMKII activity partly reversed the adverse effects on both structural and electrophysiological properties, which could be chosen as a possible supplemental therapeutic target for tachycardia patients.

This work was supported by grants from National Natural and Science Foundation of China (grant number 81400249 and 81470455 to ZW and MC, http://www.nsfc.gov.cn/), China Postdoctoral Science Foundation (grant number 2016 M601852 to JW, http://www.chinapostdoctor.org.cn/) and Natural Science Foundation of Jiangsu Province of China (grant number BK2016060 to JS, http://www.jstd.gov.cn/).

There is no conflict of interests.

L. Geng, Z. Wang and C Cui contributed equally to this work.