electrophysiological properties of human induced ...electrophysiological properties of human induced...

Electrophysiological properties of human induced pluripotent stem cells

Peng Jiang,1,2 Stephanie N. Rushing,1,2 Chi-wing Kong,1,2,3,4 Jidong Fu,1,2 Deborah Kuo-Ti Lieu,1,2

Camie W. Chan,1,2 Wenbin Deng,1,2 and Ronald A. Li1,2,3,4

1Institute of Pediatric Regenerative Medicine, Shriners Hospital for Children of North America, Sacramento; and 2HumanEmbryonic Stem Cell Consortium, University of California, Davis, California; and 3Department of Medicine, Stem Celland Regenerative Medicine Programme, and 4Heart, Brain, Hormone, and Healthy Aging Research Center, Universityof Hong Kong, Hong Kong

Submitted 10 June 2009; accepted in final form 23 November 2009

Jiang P, Rushing SN, Kong CW, Fu J, Lieu DK, Chan CW, DengW, Li RA. Electrophysiological properties of human induced pluripotentstem cells. Am J Physiol Cell Physiol 298: C486–C495, 2010. Firstpublished December 2, 2009; doi:10.1152/ajpcell.00251.2009.—Humanembryonic stem cells (hESCs) can self-renew while maintaining theirpluripotency. Direct reprogramming of adult somatic cells to inducedpluripotent stem cells (iPSCs) has been reported. Although hESCs andhuman iPSCs have been shown to share a number of similarities, suchbasic properties as the electrophysiology of iPSCs have not beenexplored. Previously, we reported that several specialized ion chan-nels are functionally expressed in hESCs. Using transcriptomic anal-yses as a guide, we observed tetraethylammonium (TEA)-sensitive(IC50 � 3.3 � 2.7 mM) delayed rectifier K� currents (IKDR) in 105 of110 single iPSCs (15.4 � 0.9 pF). IKDR in iPSCs displayed a currentdensity of 7.6 � 3.8 pA/pF at �40 mV. The voltage for 50%activation (V1/2) was �7.9 � 2.0 mV, slope factor k � 9.1 � 1.5.However, Ca2�-activated K� current (IKCa), hyperpolarization-acti-vated pacemaker current (If), and voltage-gated sodium channel (NaV)and voltage-gated calcium channel (CaV) currents could not be mea-sured. TEA inhibited iPSC proliferation (EC50 � 7.8 � 1.2 mM) andviability (EC50 � 5.5 � 1.0 mM). By contrast, 4-aminopyridine(4-AP) inhibited viability (EC50 � 4.5 � 0.5 mM) but had less effecton proliferation (EC50 � 0.9 � 0.5 mM). Cell cycle analysis furtherrevealed that K� channel blockers inhibited proliferation primarily byarresting the mitotic phase. TEA and 4-AP had no effect on iPSCdifferentiation as gauged by ability to form embryoid bodies andexpression of germ layer markers after induction of differentiation.Neither iberiotoxin nor apamin had any function effects, consistentwith the lack of IKCa in iPSCs. Our results reveal further differencesand similarities between human iPSCs and hESCs. A better under-standing of the basic biology of iPSCs may facilitate their ultimateclinical application.

ion channels; potassium currents; patch-clamp recording

HUMAN EMBRYONIC STEM CELLS (hESCs), isolated from the innercell mass of blastocysts, can self-renew while maintaining theirpluripotency to differentiate into all cell types (22). AlthoughhESCs may provide an unlimited ex vivo source for cell-basedtherapies, numerous hurdles must be overcome before theirclinical application. For instance, generation of patient-specificcells for autologous transplantation has been pursued to avoidimmune rejection of the transplanted grafts. Direct reprogram-ming of adult somatic cells to become pluripotent hESC-likecells (aka induced pluripotent stem cells or iPSCs) has beenrecently reported, eliminating potential ethical concerns: forcedexpression of four pluripotency genes, namely, Oct3/4, Sox2,

c-Myc, and Klf4 (12, 18) or Oct3/4, Sox2, Nanog, and Lin28(26), suffices to reprogram mouse and human fibroblasts intoiPSCs. Human iPSCs are similar to hESCs in their morphol-ogy, proliferation, feeder dependence, surface markers, geneexpression, epigenetic status, formation of embryoid bodies invitro, promoter activities, telomerase activities, and in vivoteratoma formation (12, 18, 26). Transcriptomic analysis ofhuman iPSCs and hESCs showed that their global gene ex-pression patterns are also remarkably similar. Technically,iPSCs are cultured under conditions virtually identical to thosefor hESCs and have the capability of differentiating into allthree germ layers and their derivatives.

Although hESCs and human iPSCs have been shown to besimilar in many aspects, such basic properties as the electro-physiology of iPSCs have not been explored. Ion channels aremembrane-bound signaling proteins that play crucial biologi-cal roles in excitable as well as inexcitable cells. For instance,the complex interplays of ionic channels in neuronal, muscle,and pancreatic cells shape their action potential profiles and,subsequently, physiological functions from cognition to heartpumping and insulin secretion. As for inexcitable cells, severalK� channels have been implicated in the proliferation, cellcycle transition, and apoptosis of mesenchymal stem cells(MSCs) and tumor cells (3, 5, 6, 9, 10, 19, 20). Previously, wereported (24) that several specialized ion channels are func-tionally expressed in hESCs. When ion channels are blocked,proliferation of hESCs is significantly inhibited. Given that theconcern of tumorigenicity primarily arises from pluripotentcells (14), the results suggest that targeted inhibition of specificK� channel activity may lead to novel strategies for arrestingundesirable cell division in tumorigenic cells. Here we reportthe presence of functional ion channels in human iPSCs. Ourresults reveal further differences and similarities between hu-man iPSCs and hESCs. A better understanding of the basicbiology of iPSCs may facilitate their ultimate clinical applica-tion.

MATERIALS AND METHODS

Culturing and differentiation of iPSCs. Human iPSCs (foreskin,clone 3) (26), a kind gift from Dr. James Thomson (University ofWisconsin-Madison, Madison, WI), were maintained on irradiatedmouse embryonic fibroblasts (MEFs) in Dulbecco’s modified Eagle’smedium (DMEM) supplemented with 20% knockout serum replace-ment, 1 mM L-glutamine, 0.1 mM �-mercaptoethanol, 1% nonessen-tial amino acid, and 20 ng/ml human basic fibroblast growth factor(bFGF) (all from GIBCO-BRL, Gaithersburg, MD). The medium waschanged every day. To isolate single iPSCs for experiments, onlyiPSC colonies with morphology typical of undifferentiated cells weremanually dissected out with glass needles followed by enzymaticdissociation with 0.25% trypsin-EDTA (GIBCO-BRL). Before ionic

Address for reprint requests and other correspondence: R. A. Li, Center ofCardiovascular Research Mount Sinai School of Medicine, New York, NY10029.

Am J Physiol Cell Physiol 298: C486–C495, 2010.First published December 2, 2009; doi:10.1152/ajpcell.00251.2009.

0363-6143/10 $8.00 Copyright © 2010 the American Physiological Society http://www.ajpcell.orgC486

on March 8, 2010

ajpcell.physiology.orgD

ownloaded from

http://ajpcell.physiology.org

current recordings, single cells were allowed to attach to poly-D-lysine(Sigma-Aldrich, St. Louis, MO)-coated glass coverslips for 30 min.

To induce the formation of embryoid bodies (EBs), iPSCs weredetached with 1 mg/ml type IV collagenase (GIBCO-BRL) andtransferred to Costar ultra-low-attachment six-well plates (Corning,Schiphol-Rijk, The Netherlands) in DMEM supplemented with 20%fetal bovine serum defined (Hyclone, Logan, UT), 2 mM L-glutamine,and 1% nonessential amino acid stock in the absence of human bFGF.The aggregates were cultured in suspension for 7 days, and themedium with or without K� channel blockers was changed every day.

Immunostaining. Human iPSC colonies were fixed in 4% parafor-maldehyde for 15 min at room temperature (21–22°C), washed withPBS, and permeabilized with 0.1% Triton X-100-PBS. The colonieswere then blocked with 4% goat serum in PBS for 2 h at roomtemperature. Fixed colonies were incubated with primary antibodies ata dilution of 1:25 (for SSEA-4, Chemicon) or 1:100 (for Oct4, SantaCruz Biotechnology) overnight at 4°C, followed by incubation withfluorescence-labeled secondary antibodies for 1 h at room temperatureand visualization by laser-scanning confocal microscopy.

Cell proliferation assay. To examine the role of K� channels in cellproliferation, human iPSCs were treated with specified concentrationsof tetraethylammonium (TEA), 4-aminopyridine (4-AP), iberiotoxin(IBTX), or apamin for 24, 48, or 72 h as indicated. Cell proliferationwas determined in 96-well plates with a nonradioactive chemilumi-nescent bromodeoxyuridine (BrdU) kit (Roche Diagnostics, Basel,Switzerland) according to the manufacturer’s protocols. Briefly, BrdUlabeling solution was added to give a final concentration of 10 �MBrdU. Medium was then removed after 2 h, and cells were fixed withethanol p.a. (70%) in HCl (final concentration 0.5 M) for 30 min at�20°C. After the fixative was removed, 100 �l of freshly diluted1:100 anti-BrdU peroxidase solution was added to each well for 30min, followed by washing three times. Finally, 100 �l of substratesolution was added, and luminescence was read at 405 nm by amultiwell scanning spectrophotometer automatic luminometer.

Cell viability was determined in 96-well plates with a colorimetric3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT)kit (Roche Diagnostics). Briefly, specified concentrations of TEA,4-AP, IBTX, or apamin were added for 24, 48, or 72 h as indicated.MTT (10 �l) labeling reagent (5 mg/ml in PBS) was then added toeach well, followed by incubation at 37°C for 4 h. Solubilizationsolution (100 �l) was added to dissolve the formazan crystals formed.

Absorbance at 540 nm was read by a spectrophotometer. Untreatedcells were used as control.

DNA cytometric analysis. After incubation of iPSCs in control versusK� channel blocker-containing medium for 24 h, iPSC colonies weremanually isolated from the underlying MEFs and trypsinized to obtain asingle-cell suspension. Univariate analysis of the cellular DNA contentwas performed on propidium iodide-stained nuclei from ice-cold 70%ethanol-fixed iPSCs with a Cytomics FC500 (Beckman Coulter). Appro-priate doublet discrimination was achieved through proper gating. DNAcontents were measured in �10,000 cells/sample for generating singleparameter histograms. Cell cycle distribution was deconvoluted by fittingwith the mathematical models in ModFit LT (version 3.1).

Electrophysiology. Whole cell voltage-clamp recordings were per-formed at room temperature. Pipette electrodes (TW120F-6; WorldPrecision Instruments, Sarasota, FL) were fabricated with a SutterP-97 horizontal puller and had final tip resistances of 4–6 M�.Membrane currents were measured with a patch-clamp amplifier(Axon 200B, Axon Instruments, Foster City, CA), sampled, andanalyzed with a Digidata 1320A interface and a personal computerwith Clampex and Clampfit software (version 9.0.1, Axon Instru-ments). In most experiments, 70–90% series resistance was compen-sated. All recordings were performed in Tyrode solution consisting of(in mM) 140 NaCl, 5 KCl, 1 CaCl2, 1 MgCl2, 10 HEPES, and 10glucose, pH adjusted to 7.4 with NaOH. The internal solution con-tained (in mM) 4 NaCl, 130 KCl, 0.5 MgCl2, 10 HEPES, 10 EGTA,and 5 MgATP, pH adjusted to 7.3 with KOH. For recording Ca2�-activated K� current (IKCa), the EDTA in the internal solution wasdecreased to 1 mM. For recording Na� or Ca2� currents, K� in theinternal solution was replaced by equimolar Cs�. Blockers werediluted to final concentrations in the bath solution as indicated andadministrated via superfusion (at least 10 ml) with a fast-exchangeperfusion system. The current amplitude at �50 mV was monitoredevery 30 s after 5 min of incubation until steady-state current blockagewas achieved.

Clampfit software was used for data analysis. Half-blocking con-centrations (IC50) were determined from the following equation: E �Emax[1 � (IC50/C)nH], where E is the inhibition of K� current inpercentage at concentration C, Emax is the maximum inhibition, andnH is the Hill coefficient. The curves presented in Figs. 2, 3, and 5were fitted to averaged data points pooled from all experiments.

Table 1. Primers used in RT-PCR analysis (annealing temperature 55°C)

Gene Accession No. Forward Sequence Reverse Sequence Length, bp

CACNA1H (Cav3.2) NM_001005407 ACTAATGCTCTGGAGATCAGCAAC GATGCTGAAGATGAAAATGAAGAGC 330CACNA2D1 (Cav, 2/-subunit1) NM_000722 AGCAGTCCATATTCCTACTGACATC GACAGATGTTCGGATCAGTTTAAGT 337CACNA2D2 (Cav, 2/-subunit2) NM_001005505 AGAACAAGGTCAACTATTCATACGC ATTCACATAGTCATCATCAGACAGC 392CACNG4 (Cav, �4-subunit) NM_014405 CACTTCCCAGAGGACAATGACTAC GTAAAAAGACCAGCCGTAGTTGTAA 306CACNG7 (Cav, �7-subunit) NM_031896 ACTGACTACTGGCTGTACATGGAA CGTAGCGATAATGAAAATACTGCTC 439HCN4 NM_005477 AGGAGATCATCAACTTTAACTGTCG TCTCATTCTCCTGGTAGTTGAAGA 486KCNC4 (Kv3.4) NM_153763 CTCTTCGAGGATCCCTACTCCT ATGATGTTGAGCAGGTTCTTGAC 308KCNK1 (K2P1.1) NM_002245 GTCACTGTGTCCTGCTTCTTCTT ACAGTTGGTCATGCTCTATGATGT 446KCNK12 (K2P12.1) NM_022055 GCTCATCGGCCTCTACCTG AAGAGGTTGAAGAACAGGATGGT 360KCNK5 (K2P5.1) NM_003740 GAGGCCAAGAAAAACTACTACACAC AGTAGTAGAGGCCCTCGATGTAGTT 493KCNK6 (K2P6.1) NM_004823 ACTTCTGCTTTATCTCTCTGTCCAC GATGGAAGCGTAGTCGGTGT 317KCNMB4 (BK, M, �4) NM_014505 GAACAACTCTGAGTCCAACTCTAGG CAGCTTCTGTACTTTGCTTTCTACA 320KCNN2 (KCa2.2) NM_170775 TGGTGACATGGTACCTAACACATAC TCTTCACTCCTTTCGTTTAAGTCAG 423KCNQ2 (Kv7.2) NM_172109 CTACATCCTGGAAATCGTGACTATC CTCTGCCAAGTACACCAGGAAC 385KCNS3 (Kv9.3) NM_002252 GTACAACCAGAGAACAGGATTCTTC CACATTCAGGTTGACAAGTTCCT 281SCN8A (Nav1.6) NM_014191 GATTCAGTCTGTGAAGAAACTGTCA AGAACTGTTCCCACAGAGTAAAGGT 253NEFH NM_021076 AGGTGAACACAGACGCTATGC GCCTCTTTCTCCTCTTCTTCAGTC 520-Actinin NM_001103 GATCCAGAACAAGATGGAGGAG AGCCGTCTCTCTAGTCATGAAGTC 514C-actin NM_005159 GACTCTGGGGATGGTGTAACTC CTGGAAGGTAGATGGAGAGAGAAG 698Albumin NM_000477 GACTATCTATCCGTGGTCCTGAAC CTCATGGTAGGCTGAGATGCTT 443Caspase-3 NM_004346 AGAGGGGATCGTTGTAGAAG GTTGCCACCTTTCGGTTAAC 304GAPDH NM_002046 TGGAAGGACTCATGACCACA AGGGGTCTACATGGCAACTG 628

C487ELECTROPHYSIOLOGICAL PROPERTIES OF HUMAN IPSCS

AJP-Cell Physiol • VOL 298 • MARCH 2010 • www.ajpcell.org

on March 8, 2010


ownloaded from


Reverse transcription-polymerase chain reaction. Total RNA wasprepared from iPSCs with the RNeasy Mini Kit (Qiagen, Valencia,CA). Single-stranded cDNA was synthesized from �1 �g of totalRNA with random hexamers and SuperScript reverse transcription(RT) (Invitrogen, Carlsbad, CA) according to the manufacturer’sprotocols, followed by polymerase chain reaction (PCR) amplificationwith gene-specific primers. Primers, annealing temperature, and prod-uct sizes are given in Table 1. cDNA was replaced by sterile nuclease-free water for negative control in each pair of primers, and nosignificant band was observed in the negative control. The reactionwas conducted using the following protocol: initial denaturing of thetemplate for 5 min at 94°C followed by 32 repeating cycles ofdenaturing for 1 min at 94°C, annealing for 1 min, extension for 1 minat 72°C, and a final elongation at 72°C for 7 min. PCR products weresize-fractionated by 1% agarose gel electrophoresis and visualized by

ethidium bromide staining. For semiquantitative analysis, GAPDHwas employed as a reference gene since its expression has been shownnot to vary significantly during hESC differentiation and amongdevelopmental stages (15). PCR products were quantified by scanningdensitometry.

Bioinformatics. Normalized microarray data were analyzed withGene Cluster (Stanford University) and TreeView (Eisen Software;Stanford University). The microarray data sets of human iPSC(GSM230054, GSM230233, GSM230235, GSM230236, GSM230237,GSM230238, GSM230239, GSM230240, GSM230241, GSM230243,GSM230245, and GSM230246), ESC (H1 GSM230264, H7 GSM230265,H9 GSM230266, H13B GSM230267, and H14A GSM230279), and MSC(GSM230260, GSM230261, GSM230262, and GSM230263) lines werefrom NCBI GEO GSE9071 (//www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc�GSE9071).

Fig. 1. Transcript expression of voltage-dependent ionchannels in human induced pluripotent stem cells (iPSCs).A: normalized transcript levels of the substantially ex-pressed ion channel genes in iPSCs. Each bar represents theaverage value of all iPSC cell lines analyzed from the dataset GSE9071. B: expression of ion channel transcripts iniPSCs probed by semiquantitative RT-PCR. C, left: 21channel genes were differentially expressed among humaniPSC, embryonic stem cell (ESC), and mesenchymal stemcell (MSC) cell lines. Right: average expression levels ofthese genes were at similar levels in iPSC and ESC lines,while they were significantly different from those of MSCcell lines.*P 0.05 compared with iPSC and ESC celllines.

C488 ELECTROPHYSIOLOGICAL PROPERTIES OF HUMAN IPSCS


on March 8, 2010


ownloaded from


Statistics. All data reported are means � SE. Paired and unpairedStudent’s t-tests were performed to evaluate the statistical significanceof differences between two groups. Analysis of variance (ANOVA)was used for multiple comparisons. Values of P 0.05 were consid-ered statistically significant.

RESULTS

Transcriptomic analysis of ion channel genes in humaniPSCs. As a first step, we analyzed the transcriptomes of iPSCs toobtain insights into the expression profiles of ion channel tran-scripts. For voltage-dependent ion channels, a total of 32, 16, 75,13, 18, and 4 transcripts were identified to encode voltage-gatedcalcium channel (CaV), voltage-gated sodium channel (NaV), volt-age-gated potassium channel (KV), calcium-activated potassiumchannel (KCa), two-pore domain potassium channel (K2P), and

hyperpolarization-activated cyclic nucleotide-modulated nonse-lective ion channel (HCN) genes, respectively, on the microarraychips employed for the transcriptomic analyses of human iPSCs,ESCs, and MSCs. After normalization, the data indicated thatamong the total of 158 voltage-dependent ion channels men-tioned, the transcripts of 5 Cav, 1 Nav, 3 Kv, 2 KCa, 4 K2P, and 1HCN channel were significantly expressed in iPSCs (Fig. 1A).Specifically, the corresponding gene products were CACNA1H(Cav3.2), CACNA2D1 (Cav 2/-subunit1), CACNA2D2 (Cav 2/-subunit2), CACNG4 (Cav �4-subunit), CACNG7 (Cav �7-subunit), SCN8A (Nav1.6), KCNC4 (Kv3.4), KCNK1 (K2P1.1),KCNK12 (K2P12.1), KCNK5 (K2P5.1), KCNK6 (K2P6.1), KCNQ2(Kv7.2), KCNS3 (Kv9.3), KCNMB4 (BK, subfamily M, �4),KCNN2 (KCa2.2), and HCN4. Figure 1B shows that RT-PCRconfirmed the array results for 15 of the 16 transcripts. Of note,

Fig. 2. Ionic currents recorded in human iPSCs.The iPSCs cultured under our conditions dis-played a normal karyotype (A) and were posi-tive for pluripotent markers Oct4 and SSEA-4(B). C: representative whole cell current trac-ings recorded from iPSCs. Inset, electrophysi-ological protocol: 300 ms voltage depolarizingsteps between �70 and �70 mV with 10-mVincrements were given from a holding potentialof �80 mV. Frequency was 0.2 Hz. The currentwas slowly activated on depolarization with asignificant tail current at �40 mV. D: represen-tative current tracing recorded from a typicalexperiment with 3-s depolarization and show-ing slight inactivation. E: current-voltage (I-V)relationship of whole cell currents determinedas an average of 8 iPSCs. Vm, membrane po-tential. F: I-V relationship of tail currents mea-sured at �40 mV. G: steady-state activation ofthe current (normalized conductance, g/gmax)determined with tail current, and g/gmax fit toBoltzmann functions: y � 1/{1 � exp[(Vm �V1/2)/k]}, where V1/2 is the estimated midpointand k is the slope factor. Mean V1/2 for activa-tion of the current was �7.9 � 2.0 mV, and kwas 9.1 � 1.5 (n � 8).



on March 8, 2010


ownloaded from


with differential expression analysis among the human iPSC,ESC, and MSC cell lines, the ion channel transcriptome of iPSCscould not be distinguished from that of hESC lines (P � 0.05) butwas distinct from human MSCs (P 0.05) as shown by theaveraged signal intensity plots (Fig. 1C).

Ionic currents in human iPSCs. Human iPSCs maintained anormal karyotype in culture (Fig. 2A), and their colonies werepositive for the pluripotency markers Oct4 and SSEA-4 (Fig.2B), the same as hESCs (22). To functionally probe theirelectrophysiological properties, patch-clamp recordings wereperformed on single isolated iPSCs. The averaged membranecapacitance was 15.4 � 0.9 pF (n � 110), comparable to hESCs(24). When iPSCs were held at �80 mV and stepped to a familyof 300-ms voltages from �70 mV to �70 mV, depolarization-activated, time-dependent outwardly rectifying currents that in-creased progressively with positive voltages and resembled thedelayed rectifier K� currents (IKDR) could be readily recorded(Fig. 2C) in 95.5% of iPSCs (105 of 110 cells). Compared withIKDR measured in hESCs, the K� current density in iPSCs wasapproximately sixfold smaller (7.6 � 3.8 pA/pF at �40 mV iniPSCs vs. 47.5 � 7.9 pA/pF at �40 mV in hESCs; Ref. 24).Inactivation of K� currents in iPSCs became apparent whendepolarization was prolonged to 3 s (Fig. 2D), with a linearcurrent-voltage (I-V) relationship and threshold potential of �20mV (Fig. 2E). As shown in Fig. 2G, normalized conductance

(g/gmax) was determined from the I-V relationship of the tailcurrent (Fig. 2F) for each cell and fitted to the Boltzmann equationto obtain the voltage for 50% activation (V1/2 � �7.9 � 2.0 mV)and slope factor (k � 9.1 � 1.5; n � 8). These gating propertieswere comparable to those of KCNQ potassium channels ex-pressed in mammalian cell lines (16) and IKDR in hESCs (24).

Figure 3, A and B, further shows that outwardly rectifyingcurrents in iPSCs could be dose-dependently inhibited by theknown K� channel blocker TEA with a half-blocking concen-tration (IC50) of 3.3 � 2.7 mM (n � 10). In the presence of 5mM 4-AP, a known blocker of several K� channel subtypesbut not KCNQ (13), a modest but detectable transient outwardcurrent (Ito)-like outward inactivating current was revealed(Fig. 3, C and D; n � 6). Such 4-AP-sensitive currentsprogressively increased in amplitude on depolarization untilsaturation was reached at �80 mV. The large-conductanceCa2�-activated K� channel-specific blocker IBTX (100 nM)and the small-conductance Ca2�-activated K� channel-specificblocker apamin (100 nM) had no effect on the outward currents(Fig. 4, A and B), suggestive of the absence of IKCa. Thehyperpolarization-activated pacemaker current (If), which hasbeen shown to express in hESCs (15), was absent in iPSCs(Fig. 4C; n � 15). Neither NaV nor CaV currents could bedetected (Fig. 4D; n � 10), even when the holding potential

nH

Fig. 3. Dissecting the individual ionic component.A: representative tracings showing that whole cellcurrent could be inhibited by the K� channelblocker tetraethylammonium (TEA). The tracing atbottom was obtained by subtracting the tracing atmiddle from that at top to show the TEA-sensitivecomponent. B: dose-response relationship for TEA-sensitive whole cell current (n � 6–10). nH, Hillcoefficient. C: representative tracings showing thecurrent blocked by the outward transient K� current-selective blocker 4-aminopyridine (4-AP). The tracingat bottom was obtained by subtracting the tracing atmiddle from that at top. This current was muchsmaller than the TEA-sensitive current (note differ-ence in scale bars). D: I-V relationship of 4-AP-sensitive current (n � 6).



on March 8, 2010


ownloaded from


was hyperpolarized to �120 mV to enable recovery from anyinactivation that might have been present.

Effects of ion channel blockers on iPSC proliferation anddifferentiation. To explore the biological roles of the ioniccurrents identified in iPSCs, we next studied the functionalconsequences of their pharmacological blockade by assessingthe effects of extracellular application of K� channel blockerson cell proliferation as well as differentiation. For proliferation,we measured DNA synthesis as an index for replication byquantifying the incorporation of BrdU into the genomic DNAduring the S phase of the cell cycle that is proportional to therate of cell division (25). To assess the cytotoxic effect of K�

channel blockers, a colorimetric MTT assay was also employed.As shown in Fig. 5A1, TEA treatment for 24 h inhibited iPSCproliferation in a dose-dependent manner. However, such inhibi-tory effects could not be distinguished from their cytotoxic effectas reflected by MTT assay (effective concentrations at 50%inhibition or EC50 were 7.8 � 1.2 and 5.5 � 1.0 mM forinhibiting proliferation and cytotoxic effect, respectively; P �

0.05, n � 4). However, 4-AP inhibited viability but had much lesseffect on proliferation (EC50 � 4.5 � 0.5 and 0.9 � 0.5 mM,respectively). In contrast, IBTX and apamin had significant effectson neither proliferation nor viability (P � 0.05, n � 6), consistentwith the lack of IKCa in iPSCs.

Figure 5, B and C, shows that longer treatments of iPSCswith TEA, 4-AP, apamin, or IBTX for 48 and 72 h did not alterthe inhibitory effects on proliferation, or the lack thereof,observed after 24 h (P � 0.05, except for 1 mM 4-AP after 72h). However, 48- and 72-h treatments enhanced the cytotoxiceffects of both TEA and 4-AP (P 0.05), although direction-ally similar results were still observed. Consistent with theirlack of effect on viability, longer incubations with apamin orIBTX had no further effect (P � 0.05).

Role of K� channels in cell cycle of iPSCs. For mechanisticinsights, we have also characterized the effects of K� channelblockers on the cell cycle distribution of iPSCs. As shown inFig. 6, all K� channel blockers tested at the same concentra-tions used for our proliferation and viability experiments sig-

Fig. 4. Absence of Ca2�-activated K� cur-rent (IKCa), hyperpolarization-activated cur-rents, and inward currents in iPSCs. A: IKCa

was recorded in a typical experiment underconditions of lower concentration of EGTA(1 mM) in the pipette solution. However, asshown in representative current tracings, noobvious IKCa was revealed by Ca2�-activatedK� channel (KCa) blockers iberiotoxin(IBTX) and apamin. B: I-V relationship wasnot significantly changed in the presence ofIBTX or apamin (n � 5). C: stimulationprotocol and representative tracing showingthat no hyperpolarization-activated currentwas observed. D: stimulation protocol andrepresentative tracing showing no inwardcurrents (INa or ICa).



on March 8, 2010


ownloaded from


nificantly reduced the proportion of cells in the G0/G1 phase(P 0.05) but without affecting the S phase (P 0.05). Also,compared with the untreated controls (21.7 � 1.2%), anincreased proportion of G2/M cells was observed after treat-ments with TEA (28.5 � 1.3% and 30.2 � 1.4% by 1 mM and3 mM, respectively, n � 3 each; P 0.05) or XE-991, a potentKCNQ channel-selective inhibitor (23) (35.7 � 1.4% by 1 �M,n � 3; P 0.01). These results support the notion that TEA andXE-991, but not 4-AP, blunted the proliferation of iPSCs andfurther suggest that iPSCs were arrested in the mitotic phase.

Effects of K� channel blockers on differentiation. To furtherprobe the roles of 4-AP- and TEA-sensitive currents, wecultured iPSCs in suspension to form three-dimensional EBsfor inducing differentiation. As shown in Fig. 7A, iPSCs couldreadily form EBs with or without TEA or 4-AP. Of note, lowconcentrations of TEA (1 mM) and 4-AP (0.3 mM) wereemployed according to the cell viability MTT assays toeliminate any potential effect due to cytotoxicity of K�

channel blockers. RT-PCR analysis further showed thatiPSC-derived EBs formed in the presence or absence ofTEA or 4-AP similarly expressed the ectodermal (NEFH),mesodermal (-actinin and C-actin), and endodermal (albu-min) germ layer markers (Fig. 7B), indicating that pluripo-tency was unaltered. Consistently, the expression levels ofseveral germ layer markers, the cardiac-specific marker-actinin, the liver-specific marker albumin, as well as theapoptotic marker caspase-3 (21), were not changed (P �0.05, n � 3).

DISCUSSION

In the present study, we demonstrate that pluripotent humaniPSCs functionally express several specialized ion channels. Inour previous study (24), we reported that IKDR is expressed inall undifferentiated H1 hESCs. Furthermore, KCNQ2 (KV7.2)that underlies the noninactivating, slowly deactivating M cur-rent (2) or IKDR shows the highest transcript expression in bothhuman iPSCs and hESCs (24). Indeed, IKDR is the major ioniccomponent of the depolarization-activated currents recorded inall iPSCs and had kinetics similar to what have been reportedfor undifferentiated hESCs (15, 24). Another previous study(15) reported the expression of HERG channels in H1 hESCs.However, the same was not observed in iPSCs either function-ally by patch-clamp recordings or in our RT-PCR experi-ments (data not shown). The lack of significant sensitivity ofIKDR in iPSCs to 4-AP, a blocker that inhibits several K�

channel types but not KCNQ (13), lends further support tothe notion that KCNQ2 is a molecular component of theIKDR recorded in iPSCs. In addition to KCNQ2, CNC4 andKCNS3, which encode for delayed rectifier KV3.4 channelsand the silent modulatory -subunit of IKDR channels (17),respectively, were also expressed. Collectively, these ionchannel genes are prime candidates that underlie the cur-rents identified in iPSCs.

Our results have indicated that iPSCs appear to be highlyelectrophysiologically homogenous: almost 100% of cells re-corded display IKDR and 4-AP-sensitive currents but not INa,ICa, If, and IKCa. As such, this functional homogeneity of iPSCsmirrors that of hESCs but starkly contrasts those of MSCs andfibroblasts. In human MSCs (1, 7, 10) and newborn foreskinfibroblasts (4), the expression levels of a range of ion

Fig. 5. A: effect of TEA (1), 4-AP (2), IBTX (3), and apamin (4) on iPSCproliferation after 24-h incubation assessed by bromodeoxyuridine (BrdU)incorporation and cytotoxicity by MTT assay after an incubation period of 24h. B and C: effects of 48-h (B) and 72-h (C) treatments assessed by BrdU andMTT assay. *P 0.05, **P 0.01, ***P 0.001.



on March 8, 2010


ownloaded from


channels such as INa, ICa, IKDR, Ito, and IKCa, etc, have beenreported to vastly vary even within the same populationsubsets. Of note, the iPSC line investigated in the presentstudy has been reprogrammed from newborn foreskin fibro-

blast, further implicating that reprogramming to the pluri-potent state returns cells to a primitive state that is relativelyfunctionally homogeneous, at least from the electrophysio-logical viewpoint.

Fig. 6. Effects of K� channel blockers on human iPSC cell cycle distribution. A: representative histograms showing the DNA contents of propidium iodide(PI)-labeled fixed cells. Control iPSCs were highly proliferative with �20% in G0/G1 phase. Relative proportions of G0/G1, S, and G2/M phase are shown.B: DNA content analysis of iPSCs in control vs. K� channel blocker-treated conditions and % of cells in each of the cell cycle phases. *P 0.05, **P 0.01.



on March 8, 2010


ownloaded from


Unlike its effect on hESCs, TEA inhibits proliferation andaffects viability of iPSCs with comparable EC50 values thatparallel that for current blockade (IC50 � 3.3 � 2.7 mM),raising the possibility that their binding to the channel receptorleads to a cascade of events that subsequently result in thefunctional consequences observed. Our cell cycle analysisfurther suggests that K� channel blockade inhibits iPSC pro-liferation by affecting the mitotic phase. Our experiments,however, do not enable us to exclude the alternative possibilitythat K� channel antagonists exert their effects on cell cycle,proliferation, and cytotoxicity via cellular uptake (e.g., endo-cytosis) followed by interacting with intracellular targets otherthan K� channels. Indeed, the latter possibility may apply to4-AP, which displays significant cytotoxicity but exerts muchless effect on proliferation and cell cycle. This observation is inaccordance with the role of K� channels in proliferation, given therelatively small 4-AP-sensitive currents expressed in iPSCs. IBTXand apamin have significant effects on neither proliferation norcell viability, consistent with the absence of IKCa.

Ion channels have been demonstrated to be crucial in initi-ating cell differentiation (8). A recent study demonstrates thelink of KCNQ1 potassium channel functions to the control ofmigration, shape, and mitotic rate during mouse embryonicmorphogenesis (11). KCNQ1 is not shown to express in humaniPSCs and ESCs according to microarray analyses. Indeed, thechannel blockers investigated in the present study exhibit nodiscernible effect on pluripotency and differentiation as gaugedby EB morphology and expression of markers for the primitivegerm layers as well as their tissue derivatives. Considering thecytotoxicity of the K� blockers, low concentrations of K�

channel blockers, at which the ionic currents could not be

vigorously blocked, were used to probe their roles in iPSCdifferentiation. As a result, the roles of the K� channel couldbe masked. To further address the functional roles of ionchannels in iPSC differentiation, specific knockout/knockdownof K� channels with small interfering RNA during the forma-tion of iPSC EBs needs to be employed in future studies.

Taken collectively, our present results reveal electrophysio-logical similarities and differences between human iPSCs,ESCs, and MSCs. Although the ionic components in humaniPSCs largely resemble those in hESCs, specific differences intheir properties and biological roles still exist. While thesimilarities likely reflect the known effect of IKDR on cellproliferation, other differences could be attributed to poorlydefined factors such as the differential expression of ion chan-nel and other accessory proteins and subunits.

ACKNOWLEDGMENTS

We thank Dr. Daniel Feldman for helpful discussion.

GRANTS

This work was in part supported by grants from the National Institutes ofHealth (R01-NS-059043, R01-ES-015988 to W. Deng and R01-HL-72857 toR. A. Li), the California Institute for Regenerative Medicine (to R. A. Li), theNational Multiple Sclerosis Society (to W. Deng), Shriners Hospitals for Children(to W. Deng), and the CC Wong Foundation Stem Cell Fund (to R. A. Li).

DISCLOSURES

The authors are not aware of financial conflict(s) with the subject matter ormaterials discussed in this manuscript with any of the authors, or any of theauthors’ academic institutions or employers.

Fig. 7. Effects of blocking ion channels on human iPSCsdifferentiation. A: in the presence and absence of K�

channel blockers TEA (1 mM) and 4-AP (0.3 mM), iPSCscould form embryoid bodies (EBs) in suspension culture.Scale bars, 100 �m. B: RT-PCR result showed that theiPSC EBs (7 days) expressed ectodermal (NEFH), meso-dermal (-actinin and C-actin), and endodermal (albumin)germ layer markers. GAPDH was used as loading control.Caspase-3 was used to compare cell death among thegroups. C: after semiquantitative analysis using GAPDHas loading control, expression levels of the germ layermarkers and caspase-3 were not significantly altered com-pared with the control group (n � 3).



on March 8, 2010


ownloaded from


REFERENCES

1. Boheler KR. Functional markers and the “homogeneity” of human mes-enchymal stem cells. J Physiol 554: 592, 2004.

2. Cooper EC, Jan LY. M-channels: neurological diseases, neuromodula-tion, and drug development. Arch Neurol 60: 496–500, 2003.

3. Deng XL, Sun HY, Lau CP, Li GR. Properties of ion channels in rabbitmesenchymal stem cells from bone marrow. Biochem Biophys Res Com-mun 348: 301–309, 2006.

4. Estacion M. Characterization of ion channels seen in subconfluent humandermal fibroblasts. J Physiol 436: 579–601, 1991.

5. Fiske JL, Fomin VP, Brown ML, Duncan RL, Sikes RA. Voltage-sensitive ion channels and cancer. Cancer Metastasis Rev 25: 493–500,2006.

6. Fraser SP, Pardo LA. Ion channels: functional expression and therapeu-tic potential in cancer. Colloquium on Ion Channels and Cancer. EMBORep 9: 512–515, 2008.

7. Heubach JF, Graf EM, Leutheuser J, Bock M, Balana B, Zahanich I,Christ T, Boxberger S, Wettwer E, Ravens U. Electrophysiologicalproperties of human mesenchymal stem cells. J Physiol 554: 659–672,2004.

8. Hinard V, Belin D, Konig S, Bader CR, Bernheim L. Initiation ofhuman myoblast differentiation via dephosphorylation of Kir2.1 K� chan-nels at tyrosine 242. Development 135: 859–867, 2008.

9. Li GR, Deng XL, Sun H, Chung SS, Tse HF, Lau CP. Ion channels inmesenchymal stem cells from rat bone marrow. Stem Cells 24: 1519–1528, 2006.

10. Li GR, Sun H, Deng X, Lau CP. Characterization of ionic currents inhuman mesenchymal stem cells from bone marrow. Stem Cells 23:371–382, 2005.

11. Morokuma J, Blackiston D, Adams DS, Seebohm G, Trimmer B,Levin M. Modulation of potassium channel function confers a hyperpro-liferative invasive phenotype on embryonic stem cells. Proc Natl Acad SciUSA 105: 16608–16613, 2008.

12. Okita K, Ichisaka T, Yamanaka S. Generation of germline-competentinduced pluripotent stem cells. Nature 448: 313–317, 2007.

13. Robbins J. KCNQ potassium channels: physiology, pathophysiology, andpharmacology. Pharmacol Ther 90: 1–19, 2001.

14. Sales KM, Winslet MC, Seifalian AM. Stem cells and cancer: anoverview. Stem Cell Rev 3: 249–255, 2007.

15. Sartiani L, Bettiol E, Stillitano F, Mugelli A, Cerbai E, Jaconi ME.Developmental changes in cardiomyocytes differentiated from human

embryonic stem cells: a molecular and electrophysiological approach.Stem Cells 25: 1136–1144, 2007.

16. Selyanko AA, Hadley JK, Wood IC, Abogadie FC, Jentsch TJ, BrownDA. Inhibition of KCNQ1–4 potassium channels expressed in mammaliancells via M1 muscarinic acetylcholine receptors. J Physiol 522: 349–355,2000.

17. Stocker M, Kerschensteiner D. Cloning and tissue distribution of twonew potassium channel alpha-subunits from rat brain. Biochem BiophysRes Commun 248: 927–934, 1998.

18. Takahashi K, Tanabe K, Ohnuki M, Narita M, Ichisaka T, Tomoda K,Yamanaka S. Induction of pluripotent stem cells from adult humanfibroblasts by defined factors. Cell 131: 861–872, 2007.

19. Tao R, Lau CP, Tse HF, Li GR. Functional ion channels in mouse bonemarrow mesenchymal stem cells. Am J Physiol Cell Physiol 293: C1561–C1567, 2007.

20. Tao R, Lau CP, Tse HF, Li GR. Regulation of cell proliferation byintermediate-conductance Ca2�-activated potassium and volume-sensitivechloride channels in mouse mesenchymal stem cells. Am J Physiol CellPhysiol 295: C1409–C1416, 2008.

21. Tewari M, Quan LT, O’Rourke K, Desnoyers S, Zeng Z, Beidler DR,Poirier GG, Salvesen GS, Dixit VM. Yama/CPP32 beta, a mammalianhomolog of CED-3, is a CrmA-inhibitable protease that cleaves the deathsubstrate poly(ADP-ribose) polymerase. Cell 81: 801–809, 1995.

22. Thomson JA, Itskovitz-Eldor J, Shapiro SS, Waknitz MA, SwiergielJJ, Marshall VS, Jones JM. Embryonic stem cell lines derived fromhuman blastocysts. Science 282: 1145–1147, 1998.

23. Wang HS, Pan Z, Shi W, Brown BS, Wymore RS, Cohen IS, DixonJE, McKinnon D. KCNQ2 and KCNQ3 potassium channel subunits:molecular correlates of the M-channel. Science 282: 1890–1893, 1998.

24. Wang K, Xue T, Tsang SY, Van Huizen R, Wong CW, Lai KW, Ye Z,Cheng L, Au KW, Zhang J, Li GR, Lau CP, Tse HF, Li RA.Electrophysiological properties of pluripotent human and mouse embry-onic stem cells. Stem Cells 23: 1526–1534, 2005.

25. Yu CC, Woods AL, Levison DA. The assessment of cellular proliferationby immunohistochemistry: a review of currently available methods andtheir applications. Histochem J 24: 121–131, 1992.

26. Yu J, Vodyanik MA, Smuga-Otto K, Antosiewicz-Bourget J, FraneJL, Tian S, Nie J, Jonsdottir GA, Ruotti V, Stewart R, Slukvin II,Thomson JA. Induced pluripotent stem cell lines derived from humansomatic cells. Science 318: 1917–1920, 2007.



on March 8, 2010


ownloaded from


electrophysiological properties of human induced ...electrophysiological properties of human induced...

Documents