| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
OPEN ACCESS ARTICLE
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
EMBRYONIC STEM CELLS |
aCentro Interuniversitario di Medicina Molecolare e Biofisica Applicata, University of Firenze, Firenze, Italy;
bBiology of Aging Laboratory, Department of Rehabilitation and Geriatrics, University Hospitals of Geneva, Geneva, Switzerland
Key Words. Embryonic stem cells • Cardiac differentiation • Ion current • Maturation • Ion channel subunits • Multicellular recordings
Correspondence: Marisa Jaconi, Ph.D., Department of Pathology and Immunology, Geneva Faculty of Medicine, Centre Médical Universitaire, 1 rue Michel-Servet, 1211 Geneva 4, Switzerland. Telephone: +41-22-379-5257; Fax: +41-22-379-5479; e-mail: marisa.jaconi{at}medecine.unige.ch
Received on July 25, 2006;
accepted for publication on January 11, 2007.
First published online in STEM CELLS EXPRESS January 25, 2007.
 |
ABSTRACT |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionDisclosure of Potential...AcknowledgmentsReferences |
|---|
Disclosure of potential conflicts of interest is found at the end of this article.
|
INTRODUCTION |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosure of Potential...AcknowledgmentsReferences |
|---|
Cardiomyocytes (CMs) derived from hESCs are considered a promising source for cell-based therapies of the heart after damage resulting from myocardial infarction or cardiomyopathies. The appearance of spontaneously beating CMs upon differentiation of hESCs into EBs has been reported in several studies. In particular, detection of cardiac-specific molecular markers and analysis of cell structure confirmed the presence of differentiated CM within EBs [4, 5]. Initial functional studies showed that the cells displayed electrophysiological properties resembling those of CMs, such as propagation of electric signal [6, 7] and presence of nodal-, embryonic atrial-, and ventricular-like action potentials (APs), indicating the presence of CM subtypes [6, 8]; hESC-derived CMs are also sensitive to chronotropic agents [5, 9] and display calcium transients [2, 8]. However, very little is known about the functional maturation of human cardiac cells over time.
The shape of APs results from the ordered integration of several functional ionic currents. During cardiac development, expression and function of diverse relevant channel types occur over time. Indeed, studies on animal models show that channels undergo fetal and postnatal developmental changes, a complex process leading to the acquisition and maintenance of a mature cardiac electrophysiological phenotype [10–12]. In hESC-derived CMs, sodium current (INa), which contributes to membrane excitability, was already present at early stages of differentiation (i.e., 20–35 days), whereas the inward rectifier potassium current (IK1), which plays a role in setting the diastolic potential in mature CMs, was absent [13].
No information is available about stability and maturation of the electrophysiological cardiac phenotype for a long time. Thus, we thought it was important to assess whether hESC-derived CMs undergo functional changes during in vitro development and how the electrophysiological phenotype is maintained over time. To this end, we applied electrophysiological and molecular techniques to characterize developmental changes of ion channels in spontaneously beating CMs for up to 3 months. Electrophysiological data, obtained with intracellular and single-cell patch-clamp recordings, and mRNA expression of ionic channels show that currents controlling AP duration and diastolic phase undergo developmentally regulated changes. The relative contribution of these channels to the generation of spontaneous AP during development was assessed with selective pharmacological tools.
|
MATERIALS AND METHODS |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosure of Potential...AcknowledgmentsReferences |
|---|
To induce EB formation, colonies were incubated for 15 minutes with collagenase IV, rinsed with phosphate-buffered saline, and gently scraped in differentiation medium (KnockOut-DMEM +20% Defined Fetal Bovine Serum [FBS; Hyclone Laboratories, Logan, UT, www.hyclone.com], 1% penicillin-streptomycin, 1% NEAA, L-glutamine, and β-mercaptoethanol), as reported elsewhere [5]. EBs were cultured for 4 days in Costar ultra-low attachment six-well plates (Corning, Schiphol-Rijk, The Netherlands, www.corning.com), with medium changed every 2 days. EBs were then plated on gelatin-coated 6-cm dishes, and medium was changed every 2–3 days.
Cell Isolation
Undifferentiated single hESCs were isolated from confluent colonies by treatment with trypsin-EDTA for 3–5 minutes. Dissociated cells were collected, resuspended in normal Tyrode's solution (see Solutions), and kept at room temperature.
Beating clumps were dissected using a microscalpel and directly placed into the solution used to perform intracellular recordings. Alternatively, they were digested with collagenase (StemCell Technologies SARL, Grenoble, France, www.stemcell.com) for 20–30 minutes at 37°C with pipetting every 5–10 minutes. Cells were then plated on gelatin- and fibronectin-coated 3.5-cm dishes in differentiation medium.
Immunofluorescence Studies
Contracting areas were dissected and replated on coverslips coated with gelatin and laminin. After 3–4 days, cells were fixed with 3% paraformaldehyde and permeabilized with 0.5% Triton X-100 (Sigma-Aldrich). Primary antibodies used in this study were mouse anti-myosin heavy chain (MHC; clone MF20, dilution 1:10; Developmental Studies Hybridoma Bank, University of Iowa, Iowa City, IA, http://dshb.biology.uiowa.edu), mouse anti-
-actinin (A7811, dilution 1:200; Sigma-Aldrich). Secondary antibodies were, respectively, goat anti-mouse-tetramethyl rhodamine isothiocyanate (Southern Biotechnology, Birmingham, AL, www.southernbiotech.com) or goat anti-mouse-rhodamine (Sigma-Aldrich). Nuclei were stained with TOTO-3 (Molecular Probes, Eugene, OR, probes.invitrogen.com). Phase contrast images and movies were taken using an inverted microscope (Eclipse TE300, Nikon, Zurich, Switzerland, www.europe-nikon.com) equipped with a camera (Coolpix 995; Nikon). Confocal pictures were taken with a laser scan microscope (LSM 510, Carl Zeiss, Oberkochen, Germany, www.zeiss.com). Stacks of images were reconstructed in three dimensions with Imaris software (Bitplane AG, Zurich, Switzerland, www.bitplane.com).
RNA Isolation and Reverse Transcriptase-Polymerase Chain Reaction
Because spontaneously contracting CMs constitute only a small fraction of EBs, parts of EBs containing beating areas were dissected under a binocular microscope. Only dissected specimens were use to performed reverse transcriptase-polymerase chain reaction (RT-PCR; except for undifferentiated cells). Total RNAs from hESCs or from up to six dissected beating clusters were isolated using Trizol Reagent (Invitrogen). Reverse transcription was performed on 0.5–1 µg of total RNA with oligo-dT primers (Promega GmbH, Mannheim, Germany, www.promega.com), random hexamers (Promega), deoxynucleoside-5'-triphosphates (Promega), and Superscript II Reverse Transcriptase (Invitrogen). PCR was performed using Taq polymerase (Qiagen AG, Hombrechtikon, Switzerland, http://www1.quiagen.com). National Center for Biotechnology Information mRNA accession numbers, primer sequences, and PCR conditions are listed in supplemental online Table 1.
Primers were designed on different exons, and the absence of cross-reactivity with mouse embryonic fibroblast cDNA was verified. After amplification, PCR fragments were resolved on 1.5% agarose gels containing ethidium bromide. PCRs were repeated on three independent experiments per time point, and pictures displayed (in Figs. 2
–5) are representative.
Quantitative RT-PCR
Expression levels of Kir2.1 and HCN1, -2, and -4 genes were further investigated using real-time quantitative RT-PCR and TaqMan (Applied Biosystems, Foster City, CA, www.appliedbiosystems.com) probe-based chemistry. Primer-Express Software (Applied Biosystems) was used to design primer and TaqMan-MGB probe sets for HCN genes. Primers and probes for Kir2.1 and the endogenous control genes GAPDH and Eukaryotic 18S rRNA were obtained from Applied Biosystems' TaqMan Gene Expression Assay catalog (Kir2.1: Hs00265315_m1; GAPDH: Hs99999905_m1; 18S rRNA: 433,3760T). These assays come in a 20x reaction mix, span an exon-exon junction, and are optimized to give approximately 100% efficiency.
The real-time RT-PCR reactions were performed using TaqMan Universal PCR Master Mix (Applied Biosystems) in a 20-µl reaction volume containing 20 ng of cDNA. All reactions were performed in triplicate and included a negative control. PCR reactions were carried out using an ABI Prism 7500 Sequence Detection System (Applied Biosystems). Cycling conditions were 2 minutes at 50°C, 10 minutes at 95°C, and 40 cycles of 15 seconds at 95°C and 1 minute at 60°C. Relative quantification of mRNA levels was determined by the 7500 system software, which uses only the comparative method (
CT).
To adequately assess the stability of GAPDH gene expression, we performed real-time PCR experiments in which GAPDH was considered the target gene and eukaryotic 18S ribosomal RNA the endogenous control. Eukaryotic 18S rRNA was chosen because its expression was stable over different samples (p = Results show that GAPDH expression does not vary significantly during differentiation and among developmental stages (p = .39; supplemental online Fig. 1). Messenger RNA levels for ion channel genes could not be measured appropriately using 18S rRNA as endogenous control gene because, in our conditions, its amplification assay interfered with the detection of less abundant targets. This led to poor reproducibility and inaccurate quantification.
Patch-Clamp Recordings
The experimental set-up for patch-clamp (whole-cell) recording and data acquisition was similar to that described previously [14, 15]. The patch-clamped cell was superfused by means of a temperature-controlled (37°C) micro-superfuser, allowing rapid changes of the solution bathing the cell. Patch-clamp pipettes, prepared from glass capillary tubes (Harvard Apparatus Ltd, Kent, U.K., www.harvardapparatus.com) by means of a two-stage horizontal puller (model P-87; Sutter Instrument, Novato, CA, www.sutter.com), had a resistance of 2–3 M
when filled with the internal solution. Cell membrane capacitance (Cm) was measured by applying a ±10 mV pulse starting from a holding potential of –70 mV, as previously reported [14, 15]. Series resistance (Rs) and membrane capacitance were compensated to minimize the capacitive transient and routinely checked during the experiment. Only cells showing a stable Cm and Rs were included in the analysis. Properly modified Tyrode's or pipette solutions were used to measure the different ionic currents (see Solutions).
The rapid delayed rectifier outward current (IKr) was evoked by steps in the range of –40 to +50 mV (holding potential [HP] –40 mV) to inactivate the sodium current (INa) and T-type calcium currents (ICaT).
Transient outward potassium current (Ito1) was evoked by steps to –40/+70 mV (HP –70 mV); a pre-step to –40 mV was used to inactivate INa current. Ito1 was measured as the difference between peak outward current at the beginning of the depolarizing step and the steady-state current at the end of the step, and normalized with respect to Cm.
The funny current (If) was evoked by hyperpolarizing steps to –50/–130 mV (HP –40 mV). To evaluate steady-state values of If current, data were fitted to a monoexponential decay. The fitting allowed calculating the time constant of current activation [
(ms)], which was plotted as a function of the step potentials. If amplitudes were measured as the difference between the extrapolated value at the steady state and that at the beginning of the test pulse, and normalized with respect to Cm, as reported elsewhere [16]. IK1 was evoked by hyperpolarizing steps to –120/25 mV (HP –70 mV), measured at –90 mV as barium-sensitive current, and normalized with respect to Cm.
From a holding potential of –50 mV, ICa,L was elicited by depolarizing steps to –45/+55 mV. Peak ICa,L was measured as the difference between the peak inward current at the beginning of the depolarizing step and the steady-state current at the end of the step.
Intracellular Recordings
Spontaneously beating EBs were placed on the Sylgard bottom (Sylgard 184 Silicone Elastomer Kit; Dow Corning, Midland, MI, http://www.dowcorning.com) of a perfusion chamber and fixed with metal pins. The chamber was thermostatically controlled at 33°C–35°C and superfusing solution (see Solutions) maintained at a constant-flow. Cellular electrical activity was recorded using standard electrophysiological techniques, as described in detail previously [17]. Briefly, the recording electrode consisted of a short Ag/AgCl pin that was partly inserted into a floating glass microelectrode containing 3 M KCl and connected to the headstage of the amplifier; an Ag/AgCl pellet served as reference electrode in the perfusion chamber. The tip resistance of the microelectrode ranged between 30 and 40 M. The recording microelectrode and the reference electrode were connected through a high-input impedance amplifier (Biomedica Mangoni, Pisa, Italy, www.biomedicamangoni.it) interfaced with a computer. The microelectrode was slowly moved into the chamber under microscopic inspection with the use of a micromanipulator. The electrode potential was compensated to 0 in the bathing solution. Spontaneous APs were digitized by an A/D converter and analyzed off-line with Iox software (emka TECHNOLOGIES, Falls Church, VA, www.emkatech.com).
Solutions
Normal Tyrode's solution (in mM): NaCl, 140; KCl, 5.4; CaCl2, 1.8; MgCl2, 1.2; D-glucose, 10; and HEPES, 5 (pH 7.35 with NaOH). Modified Tyrode's solution for If current (in mM): NaCl, 140; KCl, 25; CaCl2, 1.5; MgCl2, 1.2; BaCl2, 2; MnCl2, 2; 4-aminopyridine, 0.5; glucose, 10; and HEPES-NaOH, 5 (pH 7.35); this solution allowed the reduction of interference from other currents (i.e., ICa,L, ICa,T, IK1, and Ito1). Modified Tyrode's solution for Ito1 current: normal Tyrode's solution plus 0.5 mM CdCl2. Modified Tyrode's solution for ICa,L current (in mM): TEA-Cl, 140; CsCl, 5.4; MgCl2, 1.2; CaCl2, 1.8; HEPES, 5; and glucose, 10 (adjusted to pH 7.30 with CsOH).
Pipette solutions for APs, Ito1, If, and IKr (in mM): K-aspartate, 130; Na2-ATP, 5; MgCl2, 2; CaCl2, 5; EGTA, 11; and HEPES-KOH, 10 (pH 7.2). Pipette solutions for ICa;L (in mM): Mg-ATP, 5; EGTA, 15; TEA-Cl, 20; HEPES, 10; and CsCl 125 (pH 7.20 with CsOH).
External solution for intracellular recordings (in mM): NaCl, 125; KCl, 4; NaHCO3, 25; NaH2PO4, 0.5; MgSO4, 1.2; CaCl2, 2.7; glucose, 1 (pH 7.2 when gassed with 5% CO2/95% O2).
Statistics
Data are expressed as mean ± SEM. Statistical analysis was performed using Student's t test for grouped data (in case of two groups) or one-way analysis of variance (in case of multiple groups). A p value of less than .05 was considered significant.
|
RESULTS |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosure of Potential...AcknowledgmentsReferences |
|---|
-actinin (Fig. 1Af). To compare the maturation state of contractile CMs over time, we decided to place CMs in two groups: early CMs, which were analyzed between days 15 and 40, and late CMs, which were analyzed between days 50 and 110 of differentiation, as represented in Figure 1B.
|
-subunits that assemble into the channel responsible for Ito1 have been described [19]. Therefore, we assessed by RT-PCR the expression over time of two of these isoforms, Kv1.4 and Kv4.3, which have two splice variants. Similarly to adult human heart, the shorter Kv4.3 splice variant was expressed only in late CMs (Fig. 2A), whereas the longer was homogeneously expressed in hESCs and CMs. The Kv1.4 subunit was expressed from day 25 and maintained thereafter. Kv1.4 expression was also detected in human adult heart RNA.
|
On the other hand, a different potassium current was functionally measurable both in undifferentiated hESCs and in hESC-derived CMs. Upon depolarization (HP –40 mV), hESCs and CMs displayed an outward current (Fig. 2Da, 2Db) that was sensitive to E4031 (data not shown), a selective blocker of IKr, the rapid delayed rectifier potassium current present in native CMs. Human ether-a-gogo related gene (HERG) and HERG1b, a shorter splice variant expressed in the heart and in tumor cells, constitute the channel responsible for IKr [20, 21]. RT-PCR analysis of these genes indicated the presence of HERG mRNA in all samples tested, whereas the shorter isoform HERG1b was selectively expressed in hESC-derived CMs and in adult human heart (Fig. 2C).
Temporal and Functional Changes in Diastolic Currents: Hyperpolarization-Activated and Inward Rectifier Currents
If is a diastolic current present in pacemaker, atrial, and ventricular CMs [16, 22]. Four isoforms code for f-channels: hyperpolarization-activated cyclic nucleotide-gated potassium channel (HCN)-1, -2, -3, and -4 (for review, see [23]). The isoforms have different speed of activation and coassemble to form tetrameric HCN channel pores with distinct functional properties. For that reason, we investigated the expression of HCN1, -2, and -4 subunits over time. Levels of HCN1, usually expressed during embryogenesis and typical of mammal pacemaker cells (for review, see [24]), and HCN4 were roughly conserved from hESCs to early CMs, as observed in Figure 3A. Quantitatively, their expression was significantly lower in late CMs (Fig. 3B). Interestingly, HCN1 isoform is weakly expressed in adult heart (Fig. 3A) and significantly reduced in late EBs (Fig. 3B). HCN2 mRNA was rather constant in all samples tested (Fig. 3A, 3B).
|
|
1C subunit, which mediates the voltage-dependent ICa,L in several tissues. Corresponding mRNA could be detected in undifferentiated hESCs (Fig. 5A) and in CMs. Consistent with these molecular findings, electrophysiological recordings showed the occurrence of ICa,L both in undifferentiated hESCs and in CMs. Current could be measured erratically in undifferentiated cells and consistently from small beating aggregates of CMs (Fig. 5Ba, 5Ca). Moreover, an increase of maximal current density was detected in CMs compared to hESCs; the current measured at 5 mV was –0.9 ± 0.13 pA/pF in hESCs (Fig. 5Bb; n = 3) and at –5.8 ± 1.9 mV in CMs (Fig. 5Cb; n = 4). The voltage of half maximal activation was –16.7 ± 1.7 and –25.9 ± 2.5 mV in undifferentiated and differentiated cells, respectively.
|
Figure 6A shows AP recordings from spontaneously beating clusters at an early stage of maturation in control conditions (black lines) and after exposure to 10 µM E4031 for 30 and 60 seconds (gray and red lines, respectively). E4031 produced a striking prolongation of APD after 30 seconds, which was further enhanced at 60 seconds. Afterward, clusters of CM became unexcitable because spontaneous beating stopped and could rarely be reversed, even after prolonged wash-out (data not shown). At late stages of differentiation, E4031 depolarized membrane diastolic potential and increased the frequency of spontaneous AP (data not shown). On the whole, these data suggest a key role for IKr in the regulation of membrane potential and excitability threshold of hESC-derived CMs.
|
β-Adrenoceptor activation affects excitation-contraction coupling in the human heart [25]. In our experimental conditions, isoprenaline produced a positive chronotropic effect at 20 days of differentiation (supplemental online Fig. 2A), suggesting that the β-adrenergic system was already functionally mature in early CMs. A clear-cut acceleration of spontaneous discharge was also observed in CM clusters at a late stage of differentiation (supplemental online Fig. 2B).
Finally, Figure 6D shows a typical effect of lacidipine (a calcium channel blocker of the dihydropyridine family) on the AP profile recorded from a 40-day-old beating cluster: ICa,L blockade by lacidipine caused a clear-cut reduction in the plateau duration and height, thus suggesting a major role for this current in controlling APD.
|
DISCUSSION |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosure of Potential...AcknowledgmentsReferences |
|---|
Figure 7 schematically summarizes ion currents recorded in hESCs, and early and late CMs, giving at the same time an idea of the qualitative and quantitative changes during maturation. In the sections that follow, we will discuss their relative contribution to the maturation-dependent changes of the AP profile.
Ion Currents Controlling AP Repolarization
IKr was functionally present in hESC-derived CMs and clearly controlled cellular electrical properties, as extrapolated from intracellular recordings in beating clusters challenged with the selective IKr blocker E4031. Using intracellular recordings, we demonstrated that E4031, at a concentration able to completely block IKr, [27] caused a prolongation of APD much more pronounced than previously described [6], associated with depolarization of diastolic potential. This latter effect likely results from a low functional expression of IK1 and closely resembles that described for sino-atrial node cells, which also lack IK1 [28]. Overall, these results point to a prominent role of IKr in controlling the repolarization phase of hESC-derived CMs, as previously described [6]. Both of these actions—depolarization and APD prolongation—are consistent with the function of this channel in adult ventricular APs, in which it mainly contributes to the late phase of repolarization. Interestingly, at late stages of differentiation, the major effect of HERG blockade consisted of a marked depolarization, which further caused an acceleration of spontaneous rhythm, whereas APD remained unaffected. Similarly, blockade of HERG in neonatal dog ventricular tissue results in exaggerated AP prolongation compared with adult dog [11]. Thus, in immature/early hESC-derived CMs, IKr played an essential role in the regulation of APD and, therefore, in the repolarization phase.
|
Interestingly, undifferentiated cells expressed a number of mRNA transcripts for different channel proteins. However, only part of them appeared to be translated into functional proteins. In particular, an IKr-like current, sensitive to E4031 blockade, could be clearly detected in undifferentiated hESC, together with mRNA transcripts for HERG1. The functional role of IKr in these cells is unknown. Analogously to tumor cells [20], we can infer a cell-cycle-related functional expression.
Other outward currents are likely to be involved in the control of APD during electrophysiological maturation. In particular, the density of Ito1, functionally expressed only in CMs, increased during maturation. This is not surprising, because data from animal models show that Ito1 channels undergo fetal and postnatal developmental changes, a process responsible for the acquisition and maintenance of a mature cardiac electrophysiological phenotype [10, 12, 30]. Similar results have been obtained previously in mouse ESC-derived CMs [31].
From a molecular point of view, human Ito1 channels were encoded mainly by Kv4.3, which was already present in undifferentiated hESCs. Kv1.4 mRNA, which is considered typical of endocardial ventricular cells [32], was expressed only in hESC-derived CMs. These two isoforms were differently regulated. The simultaneous appearance of Kv1.4 transcript and of the functional current suggests that this isoform contributes to Ito1 measured in hESC-derived CMs. At present, however, we did not attempt to discriminate between the different channel isoforms.
Another original finding of our work relates to the demonstration of calcium currents in hESCs and hESC-derived CMs. The functional presence of L-type calcium channels has been previously inferred in clusters of hESC-derived CMs from sensitivity to verapamil [8]. This current has been described previously in mouse ESC-derived CMs [26], but was not detected in undifferentiated hESCs [33]. Our data demonstrate that mRNA transcripts and ICaL-like currents occur in both hESC and CM; in the latter, ICaL seems to control the plateau phase, which is shortened by the selective L-type calcium antagonist lacidipine. Interestingly, molecular and functional expression of L-type calcium current seems to be enhanced upon cardiac differentiation; a larger contribution of ICaL during the plateau of action potential (characterized by high membrane resistance) may counterbalance the concomitant increase in outward potassium currents and favor the prolongation of action potential duration. However, a precise electrophysiological comparison was precluded by the low occurrence of measurable calcium currents.
Ion Currents Controlling Diastolic Potential
Similarly to Ito1, IK1 seems to be a marker of cardiac differentiation and maturation, being expressed at a significantly higher density in late stage CM. These results can explain the lack of IK1 documented in other studies focusing on hESC-derived CM at early stages of differentiation [13]. Upon differentiation toward cardiac cells, we observed a clear-cut and quantitatively significant increase of Kir2.1 expression, which was in good correlation with functional expression of IK1 (supplemental online Fig. 4A). At least in canine and rabbit heart, Kir2.1 seems to be a specific ventricular isoform [34, 35], thus reinforcing the observation that differentiation toward ventricular-like myocytes occurs in our experimental conditions. The most striking consequence of IK1 blockade in beating clusters is an acceleration of the diastolic depolarization phase. This is not surprising, because in the human heart, IK1 controls the late phase of cardiac repolarization and stabilizes the diastolic potential [36].
The inward flow of sodium ions through f-channels plays a crucial physiological role in setting spontaneous rhythm of pacemaker cells and helping them to sense their autonomic control [24]. In our study, If was clearly measured in undifferentiated hESCs and CMs. Previous studies showed similar results in mouse ESCs, but failed to demonstrate If occurrence in hESCs [33]. Moreover, our results demonstrate, for the first time, that this current undergoes developmental changes during in vitro maturation of hESC-derived CMs. In fact, the kinetics of If activation were markedly slowed in late-stage CMs compared to in early ones. Accordingly, molecular data showed that the HCN1 isoform, which is reported to have the fastest activation kinetics [37], was largely expressed in hESCs and significantly reduced during cardiac differentiation and maturation. In contrast, the slower kinetic isoform HCN2 was expressed to a similar extent throughout the differentiation process, thus increasing its contribution to current properties in late CMs. A similar pattern of expression of HCN mRNA characterizes human adult heart tissue. Thus, it is conceivable that electrophysiological properties of If reflect changes in the molecular composition of f-channels, consequent to the relative contribution of HCN isoforms; indeed, a clear-cut correlation between activation kinetics and the ratio between HCN2 versus HCN1 and HCN4 was observed (supplemental online Fig. 4B). Of note, the relative distribution of HCN isoforms changed, as expected for cardiomyocytes acquiring a ventricular phenotype, because the quantitative amount of HCN1 and HCN4 (i.e., the sino-atrial node isoforms in the adult heart) decreased significantly. Interestingly enough, the rate of diastolic depolarization, which is controlled by If activation kinetics, is also significantly slowed in late-stage CMs, and the spontaneous rhythm is correspondingly decreased.
The functional role of If is further supported by two observations obtained in beating CM clusters. Zatebradine, the prototype of f-channel blockers, decreased both the DDR and the spontaneous rhythm, whereas isoprenaline, which positively modulates f-channels via intracellular cAMP [38], increased both of them. Consistently with this result, preliminary observations show that If is positively modulated by isoprenaline in late CMs (supplemental online Fig. 3Aa, 3Ab), but not in early CMs (supplemental online Fig. 3Ba, Bb), in agreement with molecular data, suggesting a maturation-related dominance of HCN2. This isoform is indeed reported to be more sensitive to intracellular cAMP compared to HCN1. Our results, however, do not exclude the possibility that other mechanisms controlling pacemaker activity, such as inositol triphosphate-dependent calcium signaling [39], may contribute to isoprenaline-mediated effect in clusters of beating CMs.
Relevance and Limitations
Changes occurring in hESC-derived CMs during differentiation in culture may help to assess the potential of in vitro-generated CMs for heart repair. We have identified Ito1 and IK1 (absent in hESCs) as markers of phenotypical cardiac differentiation. Early CMs have quite homogeneous APs and present a low density of Ito1 and IK1, a situation compatible with an atrial or a pacemaker phenotype. Indeed, pacemaker cells are known to have low or no Ik1 and to express HCN1 [40], at variance with atrial or ventricular human cells [41]. This is confirmed by our molecular and electrophysiological data. In contrast, late-stage CMs express Ito1 and IK1 at higher densities and present a larger range of AP duration. This rather reflects differentiation into heterogeneous cardiac phenotypes, namely atrial or ventricular type. IK1 is essential for a stable and polarized diastolic membrane potential, and its absence may be arrhythmogenic in vivo [42]. Indeed, the acquisition of a suitable pattern of ion currents will be necessary for proper CM function, especially after in vivo engraftment.
Under our experimental conditions, hESC-derived cardiomyocytes reach more mature phenotypes over a period of 3 months of in vitro culturing; however, they did not reach the phenotype typical of adult ventricular cardiomyocytes. Whether this is the consequence of limitations, intrinsic to cell culture conditions or to the insufficient time of observation, will require further investigations. To our knowledge, however, the functional properties of human fetal cardiomyocytes undergoing in situ development for 2–3 months are still unknown.
Which CM population (i.e., early or late CMs) would be more suitable for heart repair remains an open question. This results from the lack of information concerning the possibility that hESC-derived CMs (or even earlier cardiac progenitors) (a) will maintain the phenotype acquired in vitro, or may undergo senescence processes as demonstrated for fetal cardiomyocytes in long-term cultures [43]; (b) will intrinsically pursue their maturation in vivo; or (c) will eventually modify their characteristics according to the surrounding tissue (via paracrine or electrical influence). Nevertheless, a precise characterization of the in vitro maturation of hESCs is a mandatory starting point for the understanding of their in vivo maturation and for their possible use in regenerative medicine.
|
DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosure of Potential...AcknowledgmentsReferences |
|---|
|
ACKNOWLEDGMENTS |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosure of Potential...AcknowledgmentsReferences |
|---|
|
REFERENCES |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosure of Potential...AcknowledgmentsReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  J. Liu, D. K. Lieu, C. W. Siu, J.-D. Fu, H.-F. Tse, and R. A. Li Facilitated maturation of Ca2+ handling properties of human embryonic stem cell-derived cardiomyocytes by calsequestrin expression Am J Physiol Cell Physiol, July 1, 2009; 297(1): C152 - C159. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Sekiguchi-Tonosaki, M. Obata, A. Haruki, T. Himi, and J. Kosaka Acetylcholine induces Ca2+ signaling in chicken retinal pigmented epithelial cells during dedifferentiation Am J Physiol Cell Physiol, May 1, 2009; 296(5): C1195 - C1206. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. Macia and P. A. Boyden Stem Cell Therapy Is Proarrhythmic Circulation, April 7, 2009; 119(13): 1814 - 1823. [Full Text] [PDF]
|
|
|
|
 |
|
 H. Reinecke, E. Minami, W.-Z. Zhu, and M. A. Laflamme Cardiogenic Differentiation and Transdifferentiation of Progenitor Cells Circ. Res., November 7, 2008; 103(10): 1058 - 1071. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Satin, I. Itzhaki, S. Rapoport, E. A. Schroder, L. Izu, G. Arbel, R. Beyar, C. W. Balke, J. Schiller, and L. Gepstein Calcium Handling in Human Embryonic Stem Cell-Derived Cardiomyocytes Stem Cells, August 1, 2008; 26(8): 1961 - 1972. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Synnergren, K. Akesson, K. Dahlenborg, H. Vidarsson, C. Ameen, D. Steel, A. Lindahl, B. Olsson, and P. Sartipy Molecular Signature of Cardiomyocyte Clusters Derived from Human Embryonic Stem Cells Stem Cells, July 1, 2008; 26(7): 1831 - 1840. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. M. Pedrotty, R. Y. Klinger, N. Badie, S. Hinds, A. Kardashian, and N. Bursac Structural coupling of cardiomyocytes and noncardiomyocytes: quantitative comparisons using a novel micropatterned cell pair assay Am J Physiol Heart Circ Physiol, July 1, 2008; 295(1): H390 - H400. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Qu, G. M. Whitaker, L. Hove-Madsen, G. F. Tibbits, and E. A. Accili Hyperpolarization-activated cyclic nucleotide-modulated 'HCN' channels confer regular and faster rhythmicity to beating mouse embryonic stem cells J. Physiol., February 1, 2008; 586(3): 701 - 716. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Liu, J. D. Fu, C. W. Siu, and R. A. Li Functional Sarcoplasmic Reticulum for Calcium Handling of Human Embryonic Stem Cell-Derived Cardiomyocytes: Insights for Driven Maturation Stem Cells, December 1, 2007; 25(12): 3038 - 3044. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 O. Caspi, I. Huber, I. Kehat, M. Habib, G. Arbel, A. Gepstein, L. Yankelson, D. Aronson, R. Beyar, and L. Gepstein Transplantation of Human Embryonic Stem Cell-Derived Cardiomyocytes Improves Myocardial Performance in Infarcted Rat Hearts J. Am. Coll. Cardiol., November 6, 2007; 50(19): 1884 - 1893. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Yanagi, M. Takano, G. Narazaki, H. Uosaki, T. Hoshino, T. Ishii, T. Misaki, and J. K. Yamashita Hyperpolarization-Activated Cyclic Nucleotide-Gated Channels and T-Type Calcium Channels Confer Automaticity of Embryonic Stem Cell-Derived Cardiomyocytes Stem Cells, November 1, 2007; 25(11): 2712 - 2719. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| STEM CELLS | THE ONCOLOGIST | CME | ALPHAMED PRESS JOURNALS |
, p < .05 late versus early CMs. Abbreviations: CM, cardiomyocyte; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; hESC, human embryonic stem cell; pA, picoampere; pF, picofarad.
