Ion Channel Expression and Characterization in Human Induced Pluripotent Stem Cell-Derived Cardiomyocytes

Background Human induced pluripotent stem cell-derived cardiomyocytes (hiPSC-CMs) are providing new possibilities for the biological study, cell therapies, and drug discovery. However, the ion channel expression and functions as well as regulations in hiPSC-CMs still need to be fully characterized. Methods Cardiomyocytes were derived from hiPS cells that were generated from two healthy donors. qPCR and patch clamp techniques were used for the study. Results In addition to the reported ion channels, INa, ICa-L, ICa-T, If, INCX, IK1, Ito, IKr, IKs IKATP, IK-pH, ISK1–3, and ISK4, we detected both the expression and currents of ACh-activated (KACh) and Na+-activated (KNa) K+, volume-regulated and calcium-activated (Cl-Ca) Cl−, and TRPV channels. All the detected ion currents except IK1, IKACh, ISK, IKNa, and TRPV1 currents contribute to AP duration. Isoprenaline increased ICa-L, If, and IKs but reduced INa and INCX, without an effect on Ito, IK1, ISK1–3, IKATP, IKr, ISK4, IKNa, ICl-Ca, and ITRPV1. Carbachol alone showed no effect on the tested ion channel currents. Conclusion Our data demonstrate that most ion channels, which are present in healthy or diseased cardiomyocytes, exist in hiPSC-CMs. Some of them contribute to action potential performance and are regulated by adrenergic stimulation.


Introduction
Since the successful reprogramming of adult somatic cells to induced pluripotent stem (iPS) cells and generation of functional cardiomyocytes from human iPS cells (hiPSC-CMs) [1][2][3][4], hiPSC-CMs have been demonstrated to have electrophysiological and pharmacological properties including action potentials and responses to antiarrhythmic drugs which are similar to those of native cardiomyocytes [4][5][6]. In addition, emerging evidences indicate that the hiPSC-CMs derived from patients with genetic heart diseases recapitulated the phenotype of the disease [7][8][9][10][11]. For some functional studies, especially the electrophysiological studies, hiPSC-CMs have also important advantages over heterologous expression systems like Xenopus oocytes, human embryonic kidney (HEK) cells, and Chinese hamster ovary (CHO) cells lacking important constituents of cardiac ion channel macromolecular complexes that might be necessary for the normal electrophysiological characteristics. In addition, transgenic animals possess cardiac electrophysiological properties crucially different from those in humans. Therefore, it has been widely accepted that hiPSC-CMs are providing new opportunities for both mechanistic and therapeutic studies on some heart diseases. For both aims, efficient induction and purification of cardiomyocytes are required. High differentiation efficiencies of cardiomyocytes have been reported [12][13][14]. However, these methods still did not provide pure cardiomyocytes, and thus, more methods for improving the purity of iPSC-CMs including selecting markers have been established [15]. The stem cell-derived cardiomyocytes could be selected and purified by using Percoll gradient fractionation [16], drug selection of cells engineered with CM promoter-driven selectable markers [17], and fluorescence-activated cell sorting (FACS) selection of CMs by mitochondrial fluorescent dye labelling [18], by surface expression of signal regulatory protein [19], or by activated leukocyte cell adhesion molecule (ALCAM, also referred to as CD166) [20,21].
Another requirement for the applications of hiPSC-CMs is that they should have functional properties similar to those of native human cardiomyocytes. However, a major obstacle for the application of hiPSC-CMs is the distinct differences between hiPSC-CMs and mature cardiomyocytes. For example, the hiPSC-CMs are smaller, displaying abnormal properties including spontaneous beating and heteromorphologic action potentials with reduced diastolic potential, compared with mature cardiomyocytes [22]. These are indicative of an immature phenotype of hiPSC-CMs. The reason for the immaturity is not clear. The abnormal electrical property probably results from the changed expression profile of ion channels. Ion channels in hiPSC-CMs have been till now only partially characterized. At present, it is known that the major cardiac ion channels such as Na + (SCN5A), L-type Ca 2+ , I to , I Kr , I Ks , and I f are functionally expressed in hiPSC-CMs, but I K1 , which is important for resting potential in mature cardiomyocytes, is lacking or is expressed at a reduced level [9,22]. Many other ion channels, which have been demonstrated to exist either in healthy or in diseased cardiomyocytes, have not been studied in hiPSC-CMs. Therefore, we designed this study to investigate the expression and function as well as the regulation of ion channels in hiPSC-CMs with a focus on those that have not yet been checked or have not been studied in detail. 2.2. Generation of Human iPS Cells. The human iPS cells (hiPSCs) were generated from primary human fibroblasts derived from skin biopsies of the two healthy donors. The hiPSC line from donor 1 (female, here abbreviated as D1) was generated using lentiviral particles carrying the transactivator rtTA and an inducible polycistronic cassette containing the reprogramming factors OCT4, SOX2, KLF4, and c-MYC as previously described [23,24]. The hiPSC line from donor 2 (male, here abbreviated as D2) was generated using the integration-free episomal 4-in-1 CoMiP reprogramming plasmid (Addgene, #63726) with the reprogramming factors OCT4, KLF4, SOX2, and c-MYC and short hairpin RNA against p53, as described previously with modifications [25].

Material and Methods
2.3. Generation of hiPSC-CMs. The hiPS cell lines were differentiated into cardiomyocytes as described in our previous studies [26,27]. Cardiomyocytes 30 to 60 days after the onset of differentiation were dissociated from 24-well plates and used for PCR analysis or plated as single cells on Matrigelcoated 3.5 cm petri dishes for patch clamp measurements.

Polymerase Chain Reaction
Assays. The preparation of total RNA using the RNeasy mini kit (Qiagen, Hilden, Germany) including DNAse treatment was performed following the protocol. The cDNA was amplified by qPCR on a Stratagene MX 3005P real-time cycler (Stratagene, USA) using a PCR mix with hot start Taq DNA polymerase and SYBR Green (Sibir Rox Hot Mastermix, BIORON, Germany; Cat number 119405) in the presence of sense and antisense primers (400 nM each). Relative mRNA expression level was calculated as the expression of the mRNA of the gene of interest relative to GAPDH in samples from treated or untreated (control) cells which was calculated by the ΔΔCT method, based on the threshold cycle (CT), as fold change = 2 −Δ(ΔCT) , where ΔCT = CT gene of interest − CT GAPDH and Δ(ΔCT) = ΔCT treated − ΔCT control [20]. Results are shown as mean ± SEM from the measurements of 3 biological replicates and 2 technical replicates. The GenBank NCBI reference sequence number (RefSeq number) and catalog number of all the investigated genes are listed in Table S1. 2.5. Patch Clamp. The action potential (AP) and ion channel currents were measured by standard patch clamp recording techniques in the whole-cell configuration. All the experiments were carried out at room temperature (22-25°C). Different protocols were used for measuring different currents. To isolate one type of ion channel current from others, a specific channel blocker or solution was used. To minimize the effects of rundown of recorded currents on the results of experiments, we carefully monitored the time-dependent change of currents. Recordings were started when the current became stable, usually within 3 to 5 minutes.
APs were recorded in a current clamp mode. To elicit APs, a holding current of −40 pA was injected through the patch pipette into the cell and brief current pulses (2 ms, 1 nA) were applied at 1 Hz to trigger APs.
For pH-sensitive current measurements, the pH value of extracellular solution was reduced to 6.0.
2.6. Statistics. All the experiment data were analyzed using InStat© (GraphPad, San Diego, USA) and SigmaPlot 11 (Jandel Scientific). Parametric or nonparametric data were determined by analyzing the data with the Kolmogorov-Smirnov test. For nonparametric data, the Kruskal-Wallis test with Dunn's multiple comparison posttest was used. The paired t-test was used for comparison before and after application of a drug. Unpaired Student's t-test was used for comparisons of two independent groups with normal distribution. All the data are shown as mean ± SEM. p < 0 05 (two-tailed) was considered significant.
Next, we checked the functional responses of hiPSC-CMs to some physiological factors like pH value and adrenergic and cholinergic stimulation. The spontaneous cell beating was accelerated by the application of 10 μM isoprenaline (Iso). In contrast, a reduction of the pH in the perfused solution decreased the beating frequency. The effect of Iso could be reversed by 30 μM carbachol (CCh) (Figure 1(a)). CCh alone, however, failed to change the cell beating frequency (Figure 1(b)). When hiPSC-CMs were stimulated at a fixed frequency of 1 Hz, Iso reduced V max (the maximal depolarization velocity) of action potentials (APs) and shortened APD50 and APD90 (repolarization of AP at 50 and 90%) but did not influence the RP (resting potential) and APA (amplitude of AP). Acidosis reduced V max and shortened APD50. CCh alone showed no effects on all the AP parame- Then, we checked the possible variations of APs in cells measured at different time points or from different subjects. We observed very similar AP parameters in cells of two time groups (days 30 to 40 and days 50 to 60) and from the two donors (D1 and D2) ( Figure 2).

Ion Channel Currents in hiPSC-CMs.
To check whether the detected ion channels are functionally expressed in the cell membrane, we measured the individual ion channel currents with the help of special recording protocols, solutions, and channel blockers. The ion channel currents I Na , I Ca-L , I Ca-T , I NCX , I f (funny current), I KACh (acetylcholine-activated K + current), I to , I Kr , I Ks , I K1 (inward rectifier K + channel current), I SK1-3 , I SK4 , I KATP , I K-pH (pH-sensitive currents), I Cl-vol (volume-regulated Cl − channel current), I Cl-Ca (Ca 2+ -activated Cl − channel current), and TRPV1 channel currents could be measured, but I f and I K1 were much smaller than those in native cardiomyocytes reported by previous studies. Na + -activated K + currents (I KNa ) were detected but very small in hiPSC-CMs (Table 1 and Figure 4). When the currents measured during 30 to 40 days after the onset of differentiation were compared with those recorded during 50 to 60 days, significant changes in I K1 , I SK4 , and I Cl-vol were observed (Table 1 and Figure S2). No significant differences in the currents were detected in cells from the two subjects D1 and D2 ( Figure S3).
To assess the channel gating kinetics in cells with different differentiation time points, the activation, inactivation, and recovery from inactivation of I Na , I Ca-L , and I to were analyzed. No significant differences were detected between the two differentiation time groups (Figures S4-S6).

Contributions of Ion Channels to APs in hiPSC-CMs.
To assess the functional roles of the individual ion channels, we tested the influence of different types of channels on APs when they were blocked or activated ( Figure 5). No significant effect on the resting potential was observed ( Figure 5(a)). Diltiazem (I Ca-L blocker, 100 M), NiCl 2 (I NCX blocker, 5 mM), and high concentration (50 mM) of [Na + ] i (I KNa activator) reduced AP amplitude ( Figure 5(b)). Lidocaine, NiCl 2 , and high concentration of [Na + ] i reduced V max ( Figure 5(c)). When the frequency was fixed at 1 Hz, lidocaine, nifedipine, diltiazem, nicorandil (I KATP activator), NiCl 2 , and the hypotonic (I Cl-vol activator) and high [Na + ] i solution shortened APD50 and APD90. The I to blocker 4-AP, I Kr blockers E-4031 and sotalol, and I Ks blocker chromalol 293B prolonged APD50 and APD90. The I K1 blocker BaCl 2 , I SK blocker apamin, I SK4 blocker TRAM 34, I Cl-Ca blocker niflumic acid, and TRPV1 blocker capsaicipine as well as K ACh activator carbachol (CCh) did not significantly change APs (Figures 1 and 5).

Regulation of Ion Channels by Adrenergic and Cholinergic
Stimulation. Finally, we investigated the responses of the ion channel to adrenergic and cholinergic stimulation. Application of 10 μM Iso to hiPSC-CMs enhanced I Ca-L , I f , and I Ks but reduced I Na and I NCX without an effect on I K1 , I KATP , I to , I Kr , I SK1-3 , I K-Na , I Cl-Ca , I SK4 , and I TRPV1 . No significant effect of CCh (10 μM) on the currents was detected ( Figure 6).

Discussion
In this study, we investigated the ion channel expression, function, and regulation by adrenergic and cholinergic stimulation in hiPSC-CMs. The novelties of this study are (1) detecting both the gene expression and currents of K ACh , K Na , Cl Ca , Cl vol , and TRPV channels in hiPSC-CMs; (2) examining the possible functional contributions to the action potential of I SK1-3 , I SK4 , K Na , I Cl-Ca , I Cl-vol , I K-pH , and I TRPV1 ; and (3) testing the effects of adrenergic and cholinergic stimulation on I SK1-3 , I SK4 , K Na , I Cl-Ca , I TRPV1 , I Na , I to , I NCX , and I KATP .
We have successfully generated hiPS-CMs with the majority of cells showing ventricular-like APs [26]. The successful differentiation of hiPS cells into cardiomyocytes was confirmed by the decrease in expression of the pluripotency gene, the increase in cardiac genes ( Figure S1), and the spontaneously cell beating.
To characterize the physiological property of hiPSC-CMs, we checked their responses to some physiological factors like pH value and adrenergic and cholinergic stimulation. Application of 10 μM isoprenaline (Iso) accelerated the beating frequency of the cells. The effect of Iso could be reversed by 30 μM carbachol (CCh) (Figure 1(a)). These data showed the similarity between hiPSC-CMs and native cardiomyocytes [28][29][30]. Acidosis reduced frequency and V max (the maximal depolarization velocity) and shortened APD50 and APD90 (repolarization of AP at 50 and 90%) but did not influence the RP (resting potential) and APA (amplitude of AP) (Figures 1(c)-1(h)). To date, these effects of acidosis on AP parameters have not been shown in hiPSC-CMs.  The ion channel expression and functions in native ventricular cardiomyocytes have been well documented [31][32][33]. Major ion channels, like SCN5A, L-type Ca 2+ channel, I to , I Kr , I Ks , and I K1 , play critical roles in the electrical activities and functions of healthy and diseased cardiomyocytes. Some channels like K ATP , I f , SK 1-3 , and TASK-1 are associated with some heart diseases. Therefore, it is important to know the similarity and also the difference of ion channels between the hiPSC-CMs and the native cardiomyocytes. To get the information about ion channel expression, function, and regulation in hiPSC-CMs, we employed qPCR and patch clamp techniques to study ion channels in our hiPSC-CMs from the two donors with different differentiation times.
First, we checked the major group, the most frequently investigated ion channels, such as the Na, Ca, transient outward current (I to ), delayed rectifier K current (I Kr , I Ks ), inward rectifier current (I K1 ), Na/Ca exchanger (I NCX ), and funny current (I f ) channels. Although those channels have been investigated in hiPSC-CMs [9,34,35], at least two rationales led us to check them in our hiPSC-CMs: (1) cell lines generated with different protocols may show different properties including ion channel expression and functions and (2) some of those channels were not investigated in detail and some important data like the regulation of some channels by adrenergic and cholinergic stimulation are still lacking. In our recent study, we investigated the effects of LPS on I Na , I Ca-L , I to , I Kr , I Ks , I NCX , I SK1-3 , I K-pH , and I KATP [27]. But we did not examine their functional effects on action potentials. The effects of adrenergic and cholinergic stimulation on those channels were not checked either. In the current study, we focused on their functional contributions to AP and regulations by adrenergic and cholinergic stimulation. Indeed, we have added some new data. For example, we observed a reduction of I Na induced by Iso, which was not shown in hiPSC-CMs and different from some reported data in native cardiomyocytes showing an increase in I Na by Iso [36]. Furthermore, we observed an inhibition of I NCX by Iso, which has not been shown at least in hiPSC-CMs so far.
Next, we assessed some ion channels that have not been investigated in detail or have not been reported in hiPSC-CMs but have been shown to exist in native cardiomyocytes.
The functional T-type calcium channel exists in the conduction system of human heart but not in ventricular myocytes under physiological condition [37]. However, they can be expressed in atrial or ventricular myocytes in some diseases like heart failure [37]. One study reported the existence of T-type calcium channels in hiPSC-CMs by showing the currents evoked by a pulse from −90 to −40 mV [38]. But no clear evidence was provided in that study. In the current study, we detected the mRNA of the three isoforms of the T-type calcium channel CACNA1G (T-type Ca 2+ channel, Cav3.1), CACNA1H (T-type Ca 2+ channel, Cav3.2), and CACNA1I (T-type Ca 2+ channel, Cav3.3). In addition, we measured the Ca 2+ current in Na + -free solution plus 20 μM TTX, which is sensitive to mibefradil, a T-type calcium channel blocker. Therefore, the current study provides clearer evidences for the presence of T-type calcium channels in hiPSC-CMs (Figures 3 and 4(c) and Table 1).
The acetylcholine-activated potassium (K ACh ) channel is involved in cholinergic regulation of the heart rate. Although the atrial-selective ion current I KACh has recently been reported in human embryonic stem cell-derived atrial-like CMs [39], I KACh has not been studied in hiPSC-CMs. Therefore, we checked them in our cell line. Two isoforms of I KACh channels KCNJ3 and KCNJ5 (Kir3.1 and Kir3.4) could be detected by qPCR (Figure 3). Patch clamp measurements displayed carbachol-activated K + currents (Figure 4(e) and Table 1), suggesting expression of functional K ACh channels.
The sodium-activated potassium (K Na ) channel has been shown to be present in cardiomyocytes from different species [40,41]. The K Na channel can only be activated by high concentration of intracellular Na + , which cannot be reached under physiological conditions. Therefore, the physiological meaning of K Na channels is not clear. The molecular identity  CACNA1C  CACNA1G  CACNA1H  CACNA1I  HCN2  HCN4  KCND3  KCNQ1  KCNH2  KCNJ2  SLC8A1  KCNN2  KCNN3  KCNN4  KCNJ11  ABCC8  KCNK3  TRPV1  TRPV2  TRPV3  TRPV4  KCNJ3  KCNJ5  KCNT1  LRRC8A  LRRC8B  LRRC8E  ANO1  BEST1  BEST2  ADRA1A  ADRA2A  ADRB1  ADRB2  CHRM2  CHRM3   of K Na channels in cardiomyocytes is not clear either. Whether this type of channels exists in hiPSC-CMs is not known. We detected in our hiPSC-CMs the presence of the KCNT1 (Slo2.2) channel, a K Na channel that is widely expressed in the brain. But K Na currents are very small (Figure 4(o) and Table 1), suggesting a minor contribution to the electrical activity of hiPSC-CMs.
Various types of Cl − currents have been reported in cardiomyocytes from different regions of the heart and in different species, including the stretch-or volume-activated Cl − current (I Cl-vol ) and the Ca 2+ -activated Cl − current (I Cl-Ca ) [42,43]. Whether these channels exist in hiPSC-CMs is unknown. In current study, both the PCR analysis and patch clamp measurements exhibited the functional expression of the volume-and Ca 2+ -activated chloride channels in hiPSC-CMs (Table 1 and Figures 3 and 4(r) and 4(s)).
Transient receptor potential (TRP) channels are a big family of ion channels, which have been reported to exist in the heart [44]. To our knowledge, functional TRP channels have not been reported in hiPSC-CMs. In this study, we found out the expression of TRPV1-4 at mRNA level ( Figure 3) and the presence of capsaicipine-(TRPV1 blocker) sensitive currents in hiPSC-CMs, suggesting that at least TRPV1 is functionally expressed. Because of the limitation-  specific blocker, we could not identify the currents conducted by TRPV2-4 channels. APs in cardiomyocytes are critical for normal cell function, especially the normal rhythm. Any changes in APs may be pathogenic. Although some ion channels, especially the major ion channels, have been assessed in hiPSC-CMs, the contributions of many channels to action potentials have not been examined yet. To further check the physiological and pathological importance of the detected ion channels in hiPSC-CMs, we investigated their influence on APs ( Figure 5). No significant effects of the tested channel blockers or activators on RP were observed ( Figure 5(a)). It is known that I K1 , which is an important determinant of RP in native cardiomyocytes, is small in hiPSC-CMs. The RP in hiPSC-CMs should be determined by some other ion channel currents. One study demonstrated that I Kr is the main determinant [45]. Our data suggest that the resting potential is probably determined by multiple ion channels. Thus, the inhibition or activation of only one channel has a negligible effect on RP. The I Ca-L blocker diltiazem and I NCX blocker NiCl 2 reduced significantly APA ( Figure 5(b)), suggesting that I Ca-L and I NCX contributed to the amplitude of APs. The I Na blocker lidocaine reduced V max significantly ( Figure 5(c)), indicating that I Na determined the depolarization speed. NiCl 2 and high [Na + ] i also reduced V max , probably due to their effects on I Na . Potassium channel blockers (4-AP, sotalol, E-4031, and chromalol 293B) prolonged APD, whereas, lidocaine, nifedipine, diltiazem, NiCl 2 , and the K ATP activator nicorandil shortened APD. All these data confirmed the expected contributions of I Na , I Ca-L , I to , I Kr , I Ks , I NCX , and I KATP to the repolarization of APs in hiPSC-CMs and are indicative of the similarity between hiPSC-CMs and native cardiomyocytes. Lidocaine and NiCl 2 shortened APD50 more strongly than they did on APD90, suggesting minor effects of I Na and I NCX on the late phase of APD. In addition, hypotonic solution and high [Na + ] i shortened APD, meaning that I Cl-vol and I KNa may influence the repolarization. But their effects on other ion channels cannot be excluded due to lack of specific blockers for both channels. Furthermore, we found that the SK1-3 channel blocker apamin, SK4 channel blocker TRAM34, I K1 blocker BaCl 2 , I Cl-Ca blocker niflumic acid, and TRPV1 channel blocker capsaicipine showed no significant effects on AP parameters, although the respective currents could be recorded. These data may hint at a minor contribution of the respective currents to APs in hiPSC-CMs. Acidosis reduced V max and shortened APD50 significantly (Figures 1(f) and 1(g)) probably due to inhibition of inward currents or enhancement of some outward currents but not through the detected pH-sensitive TASK-1 channel currents because the reduction of I TASK-1 should prolong APD. Adrenergic and cholinergic regulation is of importance for normal heart function. In some studies with hiPSC-CMs, the effects of isoprenaline (changes in beating frequency and APD) were checked and responses similar to those in native cardiomyocytes were shown [8]. Adrenalin has been shown to enhance I Ks in hiPSC-CMs [8]. Iso enhanced I Ca-L in hiPSC-CMs [38]. Both are in agreement with the reported effect in native cardiomyocytes [33]. But the regulation of many other ion channels by adrenergic and cholinergic stimulation has not been examined in hiPSC-CMs. We checked the responses of the detected ion channels to both adrenergic and cholinergic stimulation. We observed that Iso increased I Ca-L , I f , and I Ks , similar to the reported effects of Iso in native myocytes. Iso reduced I Na in our hiPSC-CMs ( Figure 6); this is different from its effect on I Na in native cardiomyocytes [36]. Furthermore, we found that Iso showed no effects on I SK1-3 , I SK4 , K ATP , I KNa , I Cl-Ca , and I Cl-vol but reduced I NCX (Figure 6), which are different from the reported data in native cells. CCh has been shown to reduce the beating frequency of hiPSC-CMs [46]. This suggests possible effects of CCh on ion channels, which have not been tested in hiPSC-CMs. In the current study, the effects of CCh on different channel currents were examined. Although no direct effect of CCh on all the tested channel currents and no effect of Iso on most of the tested currents were detected, those data may also provide relevant information for future studies with respect to ion channels in hiPSC-CMs. This and previous studies demonstrated clearly that hiPSC-CMs responded to adrenergic and cholinergic stimulation. Transcriptome analysis revealed the expression of adrenoceptors in iPS cell-derived cardiomyocytes [47,48]. In the normal human heart, beta 1 adrenoceptors predominate over beta 2 with the ratio being approximately 4 : 1 [49,50]. Our data showed the presence of alpha 1, alpha 2, beta 1, and beta 2, and the mRNA level of beta 2 is higher than that of beta 1 in hiPSC-CMs (Figure 3), different from native human cardiomyocytes. Regarding muscarinic receptors, the predominant cardiac and ventricular isoform is M2. However, there are indications that other family members might be present, in particular M1 and M3 receptors [51]. We detected M2, M3, and M4 with M3 having the highest expression (Figure 3), again different from native cardiomyocytes.
It has been shown that the AP morphology and some ion channel currents varied with cell ages (differentiation times) [38,52,53]. We checked the AP parameters and some ion channel currents in cells measured during 30 to 40 days and 50 to 60 days after the start of differentiation. Although PCR data showed an increase in most ion channels, patch clamp measurements showed that only some channel currents like I Na , I SK4 , and I Cl-vol were significantly changed. More importantly, AP parameters exhibited no timedependent changes. In addition, we observed similar APs and ion channel currents in cells from both healthy donors, especially in older cells (Figures 2 and S3). Even though only two cell lines (one from each donor) were investigated in this study and the variation of ion channels in different cell lines cannot be excluded, the AP and current data from the two donors showed that the variation between D1 and D2 is not large. The qPCR data of some ion channels did not correlate with the current data. One reason could be that the gene of an ion channel was transcribed, but there was no equivalent protein in the membrane. Another reason could be that the channels were in the membrane, but the channel open probability was changed by some intracellular factors.
Our data displayed that hiPSC-CMs possess similarities but also differences when compared with the reported data from mature cardiomyocytes. However, quantitative comparison should be performed with caution. The reported data varied in different research groups, and more importantly, different groups performed the studies under different  experimental conditions like temperatures, solutions, recording protocols, and tissues species. It would be more plausible to compare the data from hiPSC-CMs and native human cardiomyocytes under the same experimental conditions. Due to limited availability of human ventricular cardiomyocytes, we could not compare our data from hiPSC-CMs with that from the native human cardiomyocytes under the same condition. Nevertheless, this study added some novel data about the functional ion channels and their regulation by adrenergic and cholinergic stimulation in hiPSC-CMs.

Conclusions
This study demonstrated that the ion channels that exist in mature cardiomyocytes are also present in hiPSC-CMs with similarities or differences in some aspects. The novel data from this study may provide useful information for future studies regarding ion channels in hiPSC-CMs.