Systematic expression analysis of genes related to generation of action potentials in human iPS cell-derived cardiomyocytes

Journal of Pharmacological Sciences 140 (2019) 325e330

Contents lists available at ScienceDirect

Journal of Pharmacological Sciences journal homepage: www.elsevier.com/locate/jphs

Full Paper

Systematic expression analysis of genes related to generation of action potentials in human iPS cell-derived cardiomyocytes Masami Kodama a, b, 1, Kazuharu Furutani c, d, 1, Reiko Kimura a, Tomoko Ando a, Kazuho Sakamoto e, Shushi Nagamori f, Takashi Ashihara g, Yoshihisa Kurachi c, Yuko Sekino h, Tetsushi Furukawa a, Yasunari Kanda i, Junko Kurokawa a, e, * a

Department of Bio-informational Pharmacology, Medical Research Institute, Tokyo Medical and Dental University, 1-5-45 Yushima, Bunkyo-ku, Tokyo, 113-8510, Japan Institute for Quantitative Biosciences, The University of Tokyo, Tokyo, 113-0032, Japan c Department of Pharmacology, Graduate School of Medicine, Osaka University, Suita, Osaka, 565-0871, Japan d Department of Physiology and Membrane Biology, University of California, Davis, CA, 95616, USA e Department of Bio-informational Pharmacology, School of Pharmaceutical Sciences, University of Shizuoka, Shizuoka, 422-8526, Japan f Laboratory of Bio-Molecular Dynamics, Department of Collaborative Research, Nara Medical University, Nara, 634-8521, Japan g Department of Medical Informatics and Biomedical Engineering, Shiga University of Medical Science, Shiga, 520-2192, Japan h Laboratory of Chemical Pharmacology, Graduate School of Pharmaceutical Sciences, University of Tokyo, Tokyo, 113-0033, Japan i Division of Pharmacology, National Institute of Health Sciences, Kawasaki, Kanagawa, 210-9501, Japan b

a r t i c l e i n f o

a b s t r a c t

Article history: Received 10 May 2019 Received in revised form 4 June 2019 Accepted 12 June 2019 Available online 21 June 2019

Human induced pluripotent stem cell-derived cardiomyocytes (hiPSC-CMs) are a valuable tool to characterize the pharmacology and toxic effects of drugs on heart cells. In particular, hiPSC-CMs can be used to identify drugs that generate arrhythmias. However, it is unclear whether the expression of genes related to generation of CM action potentials differs between hiPSC-CM cell lines and the mature human heart. To address this, we obtained accurate gene expression profiles of commercially available hiPSC-CM cell lines with quantitative real time RT-PCR analysis. Expression analysis of ten cardiac proteins important for generation of action potentials and three cardiac proteins important for muscle contractility was performed using GAPDH for normalization. Comparison revealed large variations in expression levels among hiPSC-CM cell lines and between hiPSC-CMs and normal human heart. In general, gene expression in hiPSC-CM cell lines was more similar to an immature, stem-like cell than a mature cardiomyocyte from human heart samples. These results provide quantitative information about differences in gene expression between hiPSC-CM cell lines, essential for interpreting pharmacology experiments. Our approach can be used as an experimental guideline for future research on gene expression in hiPSCCMs. © 2019 The Authors. Production and hosting by Elsevier B.V. on behalf of Japanese Pharmacological Society. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/ licenses/by-nc-nd/4.0/).

Keywords: iPS cells Quantitative real time-PCR Reference gene Cardiac ion channels Action potential

1. Introduction

Abbreviations: PCR, polymerase chain reaction; RNA, ribonucleic acid; AP, action potential; CM, cardiomyocyte; hiPSC-CMs, human induced pluripotent stem cellderived cardiomyocytes. * Corresponding author. Department of Bio-informational Pharmacology, School of Pharmaceutical Sciences, University of Shizuoka, 52-1 Yada, Suruga-ku, Shizuoka, 422-8526, Japan. Fax: þ81 54 264 5779. E-mail address: [email protected] (J. Kurokawa). Peer review under responsibility of Japanese Pharmacological Society. 1 These authors contributed equally to this work.

Human induced pluripotent stem cells (hiPSCs) can be differentiated into human cardiomyocytes that are a valuable tool for in vitro pharmacological screening assays.1e3 Protocols to differentiate hiPS cell-derived cardiomyocytes (hiPSC-CMs) have been created to induce further maturation and/or to obtain large numbers of hiPSC-CMs.4e6 However, hiPSC-CMs do not fully mature. hiPSC-CMs remain fetal-type with limited electrophysiological properties, resulting in diverse electrophysiological phenotypes among several hiPSC-CM cell lines. Even commercially

https://doi.org/10.1016/j.jphs.2019.06.006 1347-8613/© 2019 The Authors. Production and hosting by Elsevier B.V. on behalf of Japanese Pharmacological Society. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

326

M. Kodama et al. / Journal of Pharmacological Sciences 140 (2019) 325e330

available cell lines have some diverse properties in electrophysiology, calcium signaling, sarcomere structures and so on.7e9 Singlecell based characterization of electrophysiological properties of hiPSC-CMs revealed cell-to-cell variation of action potentials that may result in varied pharmacological responses.7 Furthermore, we have systematically shown that low expression of IK1 channels in a commercially available cell line, iCell-cardiomyocytes (CDIJ Fujifilm, Japan), changes the effects of a HERG blocker, E-4031, on action potential shapes.4 Thus, close monitoring of electrophysiological maturation of hiPSC-CMs is crucial for their further pharmaceutical application. Previous studies have compared gene expression of different hiPSC-CM cell lines created using custom, in-house protocols for cardiac differentiation and cell maintenance.10,11 In-house protocols have issues with reproducibility.1 Comparison of gene expression profiles from hiPSC-CMs in different experimental conditions is needed to determine the robustness of pharmacological and toxicological testing methods using hiPSC-CMs.4 In this context, commercially available hiPSC-CM cell lines have an advantage over in-house differentiated cardiomyocytes in that their experimental conditions are more easily standardized. However, there has been no direct comparison of gene expression among commercially available hiPSC-CM cell lines. To provide useful information for in vitro screening assays investigating the electrophysiology of hiPSC-CMs, we systematically compared expression of genes related to electrophysiological activities in commercially available hiPSC-CM cell lines with those in human heart. RNA samples were prepared according to the manufacturer's protocols to minimize inter-laboratory variations. To obtain quantitative gene expression data, housekeeping genes were selected specifically in hiPSC-CMs and human heart. The results with the detailed protocol could be useful for electrophysiological characterization of commercially-available hiPSC-CM cell lines and to evaluate various pharmacological and toxicological testing methods. 2. Materials and methods 2.1. Human iPSC-derived cardiomyocytes The hiPSC-CMs used in this study were purchased from FUJIFILM Cellular Dynamics International, Inc. (iCell Cardiomyocytes, Tokyo, Japan, lot. 1093227, lot. 1096654, lot. 1094831), Ncardia (Cor.4U Cardiomyocytes, Cologne, Germany, lot. fCMC130930_Ax017, lot. CB191CL_V_1M, lot. CB202CL_V_4M), Takara Bio Inc. (Cellartis hiPSCM, Shiga, Japan, lot. 131217-P11012, lot. 140211-P11012, lot. AFY0011S) and ReproCELL Inc. (ReproCardio2, Kanagawa, Japan, lot. A01N016, lot. A01N002, lot. A01N030). These hiPSC-CMs were cultured following each manufacturer's instruction on 6- or 12-well plastic plates, and the cells were collected within the period appropriate to multi-electrode array (MEA) analyses (4e14 days).1,2

Total RNA of normal human heart pooled from 3 male Caucasians (ages: 30e39; cause of death: trauma) was purchased from Clontech Laboratories (Cat. # 636532, lot. 01404545A, Mountain View, California, USA) and was reverse transcribed into cDNA in the same manner as total RNA from hiPSC-CM cells. 2.3. Real-time quantitative PCR (qPCR) analysis The SYBR based real-time qPCRs were performed using the ABI 7300 Real-Time PCR System (Thermo Fisher Scientific Inc., Waltham, Massachusetts, USA) and/or ABI 7900HT Fast Real-Time PCR System (Thermo Fisher Scientific Inc.). Each PCR reaction mixture contained 5 ml of Power SYBR Green PCR Master Mix (Applied Biosystems, Foster City, USA), 1 ml of the diluted cDNA reaction mixture (corresponding to 1 ng of starting amount of RNA) and 5 mM of each primer in a total reaction volume of 10 ml. Primers (Table 1) were designed according to generally conserved primer sets unless otherwise noted.12,13 PCR reactions were performed by One-Step PCR of the following condition: 2 min at 50  C and 10 min at 95  C followed by 40 cycles of the thermal cycling of 15 s at 95  C and 60 s at 60  C in a 96- or 384-well reaction plate. The homogeneity of the PCR was assessed by dissociation curve analysis at the end of real-time qPCR experiment by a gradual increase in temperature up to 95  C and ethidium bromide staining of agarose gel after electrophoresis. Each PCR reaction was performed in triplicate (technical replicates). To evaluate gene expression of SCN5A, MYH7, MYH6, ATP2A2 and GJA1, Taqman Gene Expression Assays was performed using StepOnePlus Real-Time PCR System (Applied Biosystems) according to the manufacturer's instruction. Taqman probes used in this study are described in Supplemental Table 1. For normalization of these Taqman-based real-time qPCR, the gene expression of GAPDH was also analyzed by Taqman-based real-time qPCR. Each PCR reaction was performed in triplicate (technical replicates). Relationship between Ct values and serial dilution of cDNA was calculated by linear regression to find the standard curve for each primer. The real-time qPCR efficiency (E) of one cycle in the exponential phase was determined based on the standard curve according to the equation:

E ¼ 10ð1=slopeÞ and was shown in Table 2. The ratio of mRNA of a target gene expressed in a hiPSC-CM cell line versus human heart in comparison to a reference gene was calculated by the following equation based on the Pfaffl method.14


DCttarget ðhuman hearthiPSCÞ Etarget ratio ¼  DCtref ðhuman hearthiPSCÞ Eref 3. Results

2.2. Total RNA isolation and cDNA synthesis 3.1. Identification of reference gene for data normalization Total RNA was extracted from hiPSC-CM sheet using the RNeasy Mini kit (Qiagen, Valencia, USA) with the addition of an on-column DNase I digestion according to the manufacturer's instructions. Concentration of RNA was determined using the NanoDrop ND1000 spectrophotometer (NanoDrop Technologies, Wilmington, DE, USA) and the integrity was verified by the ratio of absorbance at 260 nme280 nm > 1.90. First-strand cDNA was synthesized from 200 ng of total RNA in a total volume of 20 ml using the High Capacity cDNA Reverse Transcription kit (Applied Biosystems, Foster City, USA) following the manufacturer's protocol.

In order to perform real-time qPCR for relative quantification of gene expression between hiPSC-CMs and human heart, we first validated housekeeping genes as candidate reference genes for data normalization. Based on a literature search, nine commonly used control genes (GAPDH, B2M, HRPT1, RPL13A, SDHA, YWHAZ, ACTB, HMBS and TBP) were selected (Table 1). The possibility that these genes might be co-regulated is reduced, because the products of these genes are associated with a wide variety of biological functions. The cDNAs synthesized from total RNA of four different hiPSC-CMs and human heart were used as template. Homogeneity

M. Kodama et al. / Journal of Pharmacological Sciences 140 (2019) 325e330

327

Table 1 Primers used for SYBR based real-time qPCR. Primers for GAPDH, B2M, HRPT1, RPL13A, SDHA, and YWHAZ are described in the literature.13 The primer set for CACNA1C was described in Ref.12 Symbol

Gene name

Primer

Product size (bp)

Housekeeping genes GAPDH

GAPDH glyceraldehyde-3-phosphate dehydrogenase

87

B2M

beta-2-microglobulin

HRPT1

hypoxanthine phosphoribosyltransferase 1 ribosomal protein L13a

F: TGCACCACCAACTGCTTAGC R: GGCATGGACTGTGGTCATGAG F: TGCTGTCTCCATGTTTGATGTATCT R: TCTCTGCTCCCCACCTCTAAGT F: TGACACTGGCAAAACAATGCA R: GGTCCTTTTCACCAGCAAGCT F: CCTGGAGGAGAAGAGGAAAGAGA R: TTGAGGACCTCTGTGTATTTGTCAA F: TGGGAACAAGAGGGCATCTG R: CCACCACTGCATCAAATTCATG F: ACTTTTGGTACATTGTGGCTTCAA R: CCGCCAGGACAAACCAGTAT F: GCACAGAGCCTCGCCTTTG R: ATCCATGGTGAGCTGGCG F: AGGATGGGCAACTGTACCTG R: ATGGTAGCCTGCATGGTCTC F: CGGCTGTTTAACTTCGCTTC R: CCAGCACACTCTTCTCAGCA F: AAGGCTACCTGGATTGGATCAC R: GCCACGTTTTCGGTGTTGAC F: TGTCACGGATGAATGCCCAA R: CAAACACAGCTTGCCGTCTC F: TCAACTGCGAGATACCAACATG R: CTGGCTGCTCCGTGTCCTT F: CGCCTGAACCGAGTAGAAGA R: TGAAGCATGTCGGTGATGAG F: ACGCCAAGGCACCTGAAAC R: TGGATGGGAAGGAGGATGAA F: TGTGCATCTCAGCAATGTCA R: TGATGCCAATGCTCTCACTC F: GGCTGTCATCTTCCTCATTGG R: CGGTGGCCAGCAAACC F: ACAGCTGCCAAGGCTACCTA R: GCTTTTGACGTGCTTGTTGA

136

RPL13A SDHA

ACTB

succinate dehydrogenase complex flavoprotein subunit A tyrosine 3-monooxygenase/tryptophan 5-monooxygenase activation protein zeta actin beta

HMBS

hydroxymethylbilane synthase

TBP

TATA-box binding protein

YWHAZ

Target genes CACNA1C

SLC8A1

calcium voltage-gated channel subunit alpha1 C potassium voltage-gated channel subfamily J member 2 potassium voltage-gated channel subfamily H member 2 potassium voltage-gated channel subfamily Q member 1 hyperpolarization activated cyclic nucleotide gated potassium channel 4 solute carrier family 8 member A1

ATP1A1

ATPase Naþ/Kþ transporting subunit alpha 1

PLN

phospholamban

KCNJ2 KCNH2 KCNQ1 HCN4

Table 2 Efficiencies of real-time qPCR. Efficiencies of real-time qPCR (E) were calculated from standard curves and are shown here along with the R2. Data from Taqman gene expression assays were also calculated from standard curves, indicated with (Taqman) following gene symbols. Gene symbols Housekeeping genes GAPDH ACTB B2M HRPT1 RPL13A HMBS SDHA YWHAZ TBP Target genes SCN5A (Taqman) CACNA1C KCNJ2 KCNH2 KCNQ1 HCN4 SLC8A1 ATP1A1 ATP2A2 (Taqman) MYH6 (Taqman) MYH7 (Taqman) PLN GJA1 (Taqman) GAPDH (Taqman)

E

R2

1.926 1.928 1.900 1.898 1.864 1.874 1.835 1.816 1.511

1.000 0.993 0.991 0.987 0.985 0.985 0.984 0.996 0.890

1.857 1.907 1.926 1.949 2.101 1.941 1.896 2.040 1.955 1.836 2.038 2.034 1.903 1.793

0.999 0.989 0.975 0.995 0.977 0.967 0.998 0.998 0.999 0.964 0.999 0.996 0.955 0.997

86 94 126 86 94 63 82 121

184 128 71 115 191 62 191

of real-time qPCR products was confirmed by the presence of a single peak in the dissociation curve and the presence of a single band at the expected size after agarose gel electrophoresis and EtBr staining. The cycle threshold (Ct) value in the real-time qPCR reaction, defined as number of cycles at which the fluorescent signal crosses a threshold during the early exponential phase of the reaction, depends on the gene expression level and amplification efficiency of PCR. We obtained Ct values for the nine housekeeping genes to compare the expression level of these genes between hiPSC-CMs and human heart (Fig. 1). The selected genes had different Ct values ranging from 17.4 to 30.3 cycles, and were roughly classified into three groups. ACTB and GAPDH belonged to the group with the smallest Ct values, with mean Ct below 21.4 cycles, while HMBS and HRPT1 belonged to the group with largest Ct values, with mean Ct above 27.5 cycles. Mean Ct values of the rest (B2M, RPL13A, SDHA, YWHAZ and TBP) ranged from 22.8 to 26.9 cycles. A publicly available software tool, geNormPLUS module in qBaseþ 3.0 (Biogazelle, Gent, Belgium) was used to compare gene expression stability (Fig. 2). This program estimates an expression stability value (M) for each gene as the average pairwise variation for a particular gene with all the other tested reference genes included in the analysis: expression of the gene with lower M value is more stable.13 The statistical analysis showed that the most stable housekeeping genes were HMBS, YWHAZ and GAPDH. Analysis of PCR efficiencies revealed that amplification efficiency (E) of primer sets in GAPDH was higher than those in HMBS and YWHAZ (Table 2). Furthermore, GAPDH has been reported to exhibit

328

M. Kodama et al. / Journal of Pharmacological Sciences 140 (2019) 325e330

35

GAPDH HMBS B2M

Ct value

HRPT1 RPL13A

25

YWHAZ ACTB SDHA TBP

15

Cor.4U, 4.70 ± 1.99 for Cellartis and 1.90 ± 0.55 for ReproCardio). Expression levels of SCN5A, CACNAC1C, KCNQ1, SLC8A1, and ATP2A2 in iCell-CM were similar to human heart (ratios to human heart: 0.98 ± 0.36, 1.22 ± 0.55, 1.04 ± 0.32, 1.07 ± 0.43, and 0.78 ± 0.06, respectively), but KCNH2 and ATP1A1 expression was twice as high in iCell-CM than human heart (ratios to human heart: 2.00 ± 0.88 and 1.67 ± 0.52). To visualize of metrics of membrane channels and transporters expression levels across hiPSC-CM cell lines, mean gene expression except for HCN4 were also plotted on a radar chart. The result shows that relatively high expression of KCNH2 and ATP1A1 in iCell-CMs (Supplemental Fig. 2). 3.3. Expression of other cardiac proteins

Fig. 1. Distribution of cycle threshold values for the analyzed housekeeping genes. Cycle threshold (Ct) values obtained from three samples that vary in manufacturing lot numbers were analyzed for each hiPSC-CM cell line (except for ReproCardio cells, for which only two samples were used). The qPCR reactions were performed in triplicate. The mean Ct and SD are shown for each gene.

geNorm M

1

0.7

The gene expression of other cardiac proteins were also validated. We performed real-time qPCR of MYH6, MYH7 and PLN as markers of muscle contraction. Among hiPSC-CM cell lines, expression ratio to human heart was highest in iCell-CM and extremely low in ReproCadio cells (Fig. 4) despite variations between lots (Supplemental Fig. 3). In a comparison between iCellCM and human heart, MYH6 and MYH7 expression was lower (ratios to human heart: 0.51 ± 0.23 and 0.19 ± 0.04, respectively), whereas PLN expression was more than twice as much (ratio: 2.23 ± 1.16). We next performed real-time qPCR of GJA1 encoding gap junction a-1 protein, known as connexin 43, that plays a crucial role in electrical conduction. In comparison to human heart, GJA1 expression was much higher only in ReproCardio cells (ratio: 1.74 ± 0.29), while the expression in iCell-CM was similar to human heart (ratio: 0.83 ± 0.05).

0.4

4. Discussion

Fig. 2. Gene expression stability values (M) of the housekeeping genes. M values calculated by geNorm program are shown. A lower M value indicates a more stable gene. geNorm analysis shows that the most stable genes are HMBS, YWHAZ and GAPDH.

consistent stability in human heart with or without heart failure, and regardless of the part of the heart.15e17 Thus, we decided to use GAPDH as the reference gene for normalization in this study. 3.2. Expression of cardiac membrane transporters To characterize the electrophysiological properties of hiPSC-CM cell lines, we validated the gene expression of cardiac membrane transporters related to generation of action potentials (SCN5A, CACNA1C, KCNJ2, KCNH2, KCNQ1, HCN4, SLC8A1, ATP1A1, and ATP2A2) by real-time qPCR (Fig. 3). For each hiPSC-CM cell line, three samples that varied in manufacturing lot numbers were analyzed (Supplemental Fig. 1). Expression of target gene was normalized with that of GAPDH, and the ratio versus human heart was calculated as described in the materials and methods. Among hiPSC-CM cell lines, gene expression of cardiac membrane transporters was the highest in iCell-CM, followed by Cor.4U and Cellartis. ReproCadio cells hardly expressed cardiac membrane transporters except for HCN4 and ATP1A1. In comparison to human heart, KCNJ2 expression was much lower in every hiPS-CM cell line (ratios to human heart: 0.20 ± 0.05 for iCell-CM, 0.05 ± 0.03 for Cor.4U, 0.06 ± 0.05 for Cellartis and 0.09 ± 0.02 for ReproCardio), while HCN4 expression was much higher in every hiPS-CM cell line (ratios to human heart: 10.78 ± 6.74 for iCell-CM, 4.38 ± 2.90 for

Diverse electrophysiological phenotypes have been reported for several commercially available hiPSC-CM cell lines. The present study aimed to determine the cause of electrophysiological differences by quantifying expression of genes related to generation of action potentials. In order to obtain quantitative gene expression data, a reference gene must be selected for data normalization. For our study, the housekeeping gene GAPDH was selected because of its steady and equivalent expression in hiPSC-CMs and human heart. Using the geNormPLUS module, we found that the most stable housekeeping genes among nine commonly-used control genes were HMBS, YWHAZ and GAPDH. Although GAPDH and/or ACTB are commonly used as the reference gene,13,18,19 it is well known that expression of these genes varies considerably in adult tissues and also in human embryonic stem cells.19 Thus, it is worthwhile to systematically identify the best possible reference gene for each experiment. Based on our results, we employed GAPDH as the reference to compare gene expression among four cell lines of hiPSC-CMs relative to human heart. Fig. 3 showed that expression of cardiac membrane transporters and channels varies considerably, and the overall expression level was the highest in iCell-CM, followed by Cor.4U and Cellartis. As previously reported,4 it was significant that KCNJ2 was expressed much lower than human heart in every hiPSC-CM cell line, while HCN4 was much higher. These results reflect the functional feature that hiPSC-CMs are immature, autonomously beating cells. The levels of gene expression of SCN5A, CACNAC1C, KCNQ1, SLC8A1, and ATP2A2 in iCell-CM were similar to those of human heart, but KCNH2 and ATP1A1 gene were expressed in iCell-CM twice as much as in human heart. In Cor. 4U, the similar trend was seen from the results of the radar chart analysis, except that the relative

M. Kodama et al. / Journal of Pharmacological Sciences 140 (2019) 325e330

329

Fig. 3. Expression of cardiac membrane transporters. Quantity of mRNA of each cardiac membrane transporter was normalized with GAPDH, and the ratio versus human heart was calculated as described in Materials and Methods. Mean ± SD are shown.

Fig. 4. Expression of other cardiac proteins. Quantity of mRNA of each cardiac protein was normalized with GAPDH, and the ratio versus human heart was calculated as described in Materials and Methods. Mean ± SD are shown.

expression level of CACNA1C to other channels/transporters was lower than that in iCell-CM (Supplemental Fig. 2). The difference in CACNA1C expression between iCell-CM and Cor.4U is consistent with the previously reported MEA analysis that iCell-CM has longer field potential duration than Cor4U.2,20,21 Manufacturers of commercial hiPSC-CM cell lines disclose only the percentage of cardiac muscle cells determined by a certain cardiac marker. According to the manual, the iCell CMs and Cor.4U CMs are highly purified human cardiomyocyte (purity of cardiomyocytes: iCell CMs; >95% c-TnT(þ), Cor.4U; almost 100% c-

TnT(þ), Cellartis; > 80% c-TnT(þ), ReproCardio2: >30%), which implies that the overall expression levels of cardiac ion channels and transporters reflect the purification rate of cardiomyocytes for each cell line. One would expect that these differences in gene expression reflects differences in the resistance of the cardiac membrane. The high expression of the KCNH2 gene, which encodes the HERG channel, implies that hiPSC-CMs are suitable as an experimental system for predicting the occurrence of arrhythmia induced by HERG channel suppressions.22 Although the results of testing for

330

M. Kodama et al. / Journal of Pharmacological Sciences 140 (2019) 325e330

proarrhythmic risks using the iCell CMs and the Cor. 4U CMs suggested that cell line differences do not affect drug categorization,22 these cell line differences will be important for more in-depth analysis. For example, in order to predict the sporadic arrhythmia risks induced by drugs, it is necessary to consider the balance of gene expression profiling of various cardiac ion channel and transporters. Considering the crosstalk between cardiac ion channel and transporters to generate action potentials and large variations in their expression levels between hiPSC-CMs and normal human heart, it is not possible to unequivocally predict the proarrhythmic risk of hERG blockers in mature human heart only via experimental approaches with hiPSC-CMs. To overcome this difficulty and enable us to use hiPSC-CMs to identify drugs that generate arrhythmias, we should utilize a computational modeling and simulation approach in combination with experiments to quantify their respective effects on cardiac cell electrophysiology.23 4.1. Impact of this study on safety pharmacology Recently, a protocol to evaluate proarrhythmic potentials using a monolayer hiPSC-CM sheet has been presented by Japan iPS Cardiac Safety Assessment (JiCSA) along with endpoints to predict ventricular proarrhythmic risks of proarrhythmic test compounds.24,25 The protocol employs the multi-electrode array (MEA) platform to measure electrophysiological activities of hiPSC-CM sheets for drug safety assessment. The detailed protocol with the pilot study data allows us to easily evaluate other hiPSC-CM cell lines by the comparative information in the gene expression level of APdetermining molecules. Conflict of interest None for all authors. Acknowledgements We thank Dr. M Li, Ms. E Hayashi and Ms. R Osumi for collecting total RNA, and Ms. K Yamaguchi for supporting cooperation of researchers. This research is supported by the Research on Regulatory Harmonization and Evaluation of Pharmaceuticals, Medical Devices, Regenerative and Cellular Therapy Products, Gene Therapy Products, and Cosmetics (JP15mk0104030, JP18mk0104117, JP19mk0104117) from Japan Agency for Medical Research and Development, AMED, and Grant-in-Aid for Scientific Research (MEXT/JSPS KAKENHI JP15H04684, JP19H03380) and JSPS Research Fellow (RPD) 26-40079. This work is also supported by NankenKyoten, TMDU and a support program for woman researchers from the Tokyo Medical and Dental University and promotion grants for faculty of University of Shizuoka. Appendix A. Supplementary data Supplementary data to this article can be found online at https://doi.org/10.1016/j.jphs.2019.06.006. References 1. Kanda Y, Yamazaki D, Kurokawa J, Inutsuka T, Sekino Y. Points to consider for a validation study of iPS cell-derived cardiomyocytes using a multi-electrode array system. J Pharmacol Toxicol Methods. 2016;81:196e200. https://doi.org/ 10.1016/j.vascn.2016.06.007. 2. Yamazaki D, Kitaguchi T, Ishimura M, et al. Proarrhythmia risk prediction using human induced pluripotent stem cell-derived cardiomyocytes. J Pharmacol Sci. 2018;136:249e256. https://doi.org/10.1016/j.jphs.2018.02.005.

3. Asahi Y, Hamada T, Hattori A, et al. On-chip spatiotemporal electrophysiological analysis of human stem cell derived cardiomyocytes enables quantitative assessment of proarrhythmia in drug development. Sci Rep. 2018;8:14536. https://doi.org/10.1038/s41598-018-32921-1. 4. Li M, Kanda Y, Ashihara T, et al. Overexpression of KCNJ2 in induced pluripotent stem cell-derived cardiomyocytes for the assessment of QT-prolonging drugs. J Pharmacol Sci. 2017;134:75e85. https://doi.org/10.1016/j.jphs.2017. 05.004. 5. Nunes SS, Miklas JW, Liu J, et al. Biowire: a platform for maturation of human pluripotent stem cell-derived cardiomyocytes. Nat Methods. 2013;10:781e787. https://doi.org/10.1038/nmeth.2524. 6. Tohyama S, Fujita J, Fujita C, et al. Efficient large-scale 2D culture system for human induced pluripotent stem cells and differentiated cardiomyocytes. Stem Cell Reports. 2017;9:1406e1414. https://doi.org/10.1016/j.stemcr.2017.08.025. 7. Lopez-Redondo F, Kurokawa J, Nomura F, et al. A distribution analysis of action potential parameters obtained from patch-clamped human stem cell-derived cardiomyocytes. J Pharmacol Sci. 2016;131:141e145. https://doi.org/10.1016/ j.jphs.2016.04.015. 8. Bedada FB, Chan SS, Metzger SK, et al. Acquisition of a quantitative, stoichiometrically conserved ratiometric marker of maturation status in stem cellderived cardiac myocytes. Stem Cell Rep. 2014;3:594e605. https://doi.org/10. 1016/j.stemcr.2014.07.012. 9. Bedada FB, Wheelwright M, Metzger JM. Maturation status of sarcomere structure and function in human iPSC-derived cardiac myocytes. Biochim Biophys Acta. 2016;1863:1829e1838. https://doi.org/10.1016/j.bbamcr.2015.11. 005. 10. Bock C, Kiskinis E, Verstappen G, et al. Reference maps of human ES and iPS cell variation enable high-throughput characterization of pluripotent cell lines. Cell. 2011;144:439e452. https://doi.org/10.1016/j.cell.2010.12.032. 11. Narsinh KH, Sun N, Sanchez-Freire V, et al. Single cell transcriptional profiling reveals heterogeneity of human induced pluripotent stem cells. J Clin Investig. 2011;121:1217e1221. https://doi.org/10.1172/JCI44635. 12. Zwi L, Caspi O, Arbel G, et al. Cardiomyocyte differentiation of human induced pluripotent stem cells. Circulation. 2009;120:1513e1523. https://doi.org/10. 1161/circulationaha.109.868885. 13. Vandesompele J, De Preter K, Pattyn F, et al. Accurate normalization of realtime quantitative RT-PCR data by geometric averaging of multiple internal control genes. Genome Biol. 2002;3. research0034.0031 https://doi.org/10. 1186/gb-2002-3-7-research0034. 14. Pfaffl MW. A new mathematical model for relative quantification in real-time RT-PCR. Nucleic Acids Res. 2001;29:e45. 15. Runov AL, Kurchakova EV, Khaschevskaya DA, Moiseeva OM, Vonsky MS. Selection of reference genes for transcription analysis for myocarditis studies. Tsitologiia. 2015;57:212e217. https://doi.org/10.1134/S1990519X15040100. 16. Li M, Rao M, Chen K, Zhou J, Song J. Selection of reference genes for gene expression studies in heart failure for left and right ventricles. Gene. 2017;620: 30e35. https://doi.org/10.1016/j.gene.2017.04.006. 17. Moniotte S, Vaerman JL, Kockx MM, et al. Real-time RT-PCR for the detection of beta-adrenoceptor messenger RNAs in small human endomyocardial biopsies. J Mol Cell Cardiol. 2001;33:2121e2133. https://doi.org/10.1006/jmcc.2001. 1475. 18. Rana P, Anson B, Engle S, Will Y. Characterization of human-induced pluripotent stem cellederived cardiomyocytes: bioenergetics and utilization in safety screening. Toxicol Sci. 2012;130:117e131. https://doi.org/10.1093/toxsci/ kfs233. 19. Synnergren J, Giesler TL, Adak S, et al. Differentiating human embryonic stem cells express a unique housekeeping gene signature. Stem Cells. 2007;25: 473e480. https://doi.org/10.1634/stemcells.2006-0247. 20. Huo J, Kamalakar A, Yang X, et al. Evaluation of batch variations in induced pluripotent stem cell-derived human cardiomyocytes from 2 major suppliers. Toxicol Sci. 2017;156:25e38. https://doi.org/10.1093/toxsci/kfw235. 21. Blinova K, Stohlman J, Vicente J, et al. Comprehensive translational assessment of human-induced pluripotent stem cell derived cardiomyocytes for evaluating drug-induced arrhythmias. Toxicol Sci. 2017;155:234e247. https://doi.org/10. 1093/toxsci/kfw200. 22. Blinova K, Dang Q, Millard D, et al. International multisite study of humaninduced pluripotent stem cell-derived cardiomyocytes for drug proarrhythmic potential assessment. Cell Rep. 2018;24:3582e3592. https://doi.org/ 10.1016/j.celrep.2018.08.079. 23. Kawatou M, Masumoto H, Fukushima H, et al. Modelling Torsade de Pointes arrhythmias in vitro in 3D human iPS cell-engineered heart tissue. Nat Commun. 2017;8:1078. https://doi.org/10.1038/s41467-017-01125-y. 24. Asakura K, Hayashi S, Ojima A, et al. Improvement of acquisition and analysis methods in multi-electrode array experiments with iPS cell-derived cardiomyocytes. J Pharmacol Toxicol Methods. 2015;75:17e26. https://doi.org/10. 1016/j.vascn.2015.04.002. 25. Kanda Y, Yamazaki D, Osada T, Yoshinaga T, Sawada K. Development of torsadogenic risk assessment using human induced pluripotent stem cell-derived cardiomyocytes: Japan iPS Cardiac Safety Assessment (JiCSA) update. J Pharmacol Sci. 2018;138:233e239. https://doi.org/10.1016/j.jphs.2018.10. 010.