1,4,5Inositol TrisphosphateOperated Intracellular Ca2+ Stores and ...

EMBRYONIC STEM CELLS/INDUCED PLURIPOTENT STEM CELLS 1,4,5-Inositol Trisphosphate-Operated Intracellular Ca2ⴙ Stores and Angiotensin-II/Endothelin-1 Signaling Pathway Are Functional in Human Embryonic Stem Cell-Derived Cardiomyocytes OSHRA SEDAN,a KATYA DOLNIKOV,a NAAMA ZEEVI-LEVIN,a,b NOA LEIBOVICH,a MICHAL AMIT,a,b JOSEPH ITSKOVITZ-ELDOR,a,b OFER BINAHa a

Ruth and Bruce Rappaport Faculty of Medicine, Technion-Israel Institute of Technology, Haifa, Israel; bRambam Medical Center, Haifa, Israel Key Words. Human embryonic stem cell-derived cardiomyocytes • 1,4,5-Inositol trisphosphate • Angiotensin-II • Endothelin-1 • Intracellular Ca2⫹ transient • Contraction

ABSTRACT On the basis of previous findings suggesting that in human embryonic stem cell-derived cardiomyocytes (hESC-CM) the sarcoplasmic reticulum Ca2ⴙ-induced release of calcium machinery is either absent or immature, in the present study we tested the hypothesis that hESC-CM contain fully functional 1,4,5-inositol trisphosphate (1,4,5-IP3)-operated intracellular Ca2ⴙ ([Ca2ⴙ]i) stores that can be mobilized upon appropriate physiological stimuli. To test this hypothesis we investigated the effects of angiotensin-II (AT-II) and endothelin-1 (ET-1), which activate the 1,4,5-IP3 pathway, on [Ca2ⴙ]i transients and contractions in beating clusters of hESC-CM. Our major findings were that in paced hESC-CM both AT-II and ET-1 (10ⴚ9 to 10ⴚ7 M) increased

the contraction amplitude and the maximal rates of contraction and relaxation. In addition, AT-II (10ⴚ9 to 10ⴚ7 M) increased the [Ca2ⴙ]i transient amplitude. The involvement of 1,4,5-IP3dependent intracellular Ca2ⴙ release in the inotropic effect of AT-II was supported by the findings that (a) hESC-CM express AT-II, ET-1, and 1,4,5-IP3 receptors determined by immunofluorescence staining, and (b) the effects of AT-II were blocked by 2 ␮M 2-aminoethoxyphenyl borate (a 1,4,5-IP3 receptor blocker) and U73122 (a phospholipase C blocker). In conclusion, these findings demonstrate for the first time that hESC-CM exhibit functional AT-II and ET-1 signaling pathways, as well as 1,4,5-IP3-operated releasable Ca2ⴙ stores. STEM CELLS 2008;26:3130 –3138

Disclosure of potential conflicts of interest is found at the end of this article.

INTRODUCTION Since current therapies for major heart diseases, such as myocardial ischemia and infarction, as well as congestive heart failure, are limited in their efficacy because of their inability to repair or replace damaged myocardium, myocardial regeneration by means of cardiac stem cells is a potential novel therapeutic modality [1]. In this regard, recent studies have shown that improvement of myocardial function can be achieved in experimental animal models of heart failure and infarction, by transplanting stem cell-derived cardiomyocytes into the compromised myocardium [2, 3]. Of the variety of potential stem cells sources, human embryonic stem cell-derived cardiomyocytes (hESC-CM) are plausible candidates for cell therapy, since they exhibit cardiac molecular markers and spontaneous action potential generation and contraction, as well as ␤-adrenergic responsiveness [4 – 6].

In agreement with the notion that the functional compatibility between donor hESC-CM and the host myocardium is likely to contribute to improved outcome of cardiac cell transplantation, we recently investigated basic properties of the excitation contraction coupling in contracting embryoid bodies (EBs) derived from clone H9.2 (7–55 days old) [7]. Specifically, we have shown that the functional properties of hESC-CM related to intracellular Ca2⫹ handling differ from those of the adult myocardium, as reflected by negative force-frequency relations and lack of postrest potentiation. Furthermore, hESC-CM express Sarc/Endoplasmic reticulum Ca2⫹-ATPase (SERCA2) and the Na⫹/Ca2⫹ exchanger but not calsequestrin or phospholamban, collectively suggesting immature Ca2⫹ handling. As our above-mentioned findings suggested that in hESCCM, sarcoplastic reticulum (SR) Ca2⫹ stores are not fully functional, we hypothesized that hESC-CM contain functional 1,4,5inositol trisphosphate (1,4,5-IP3)-operated intracellular Ca2⫹ stores that can be mobilized upon appropriate physiological stimuli. As reported previously, 1,4,5-inositol trisphosphate-

Author contributions: O.S.: conception and design, collection and/or assembly of data, data analysis and interpretation, manuscript writing, final approval of manuscript; K.D.: conception and design, data analysis and interpretation; N.Z.-L. and M.A.: provision of study material or patients; N.L.: data analysis and interpretation; J.I.-E.: conception and design, financial support, provision of study material or patients, final approval of manuscript; O.B.: conception and design, financial support, provision of study material or patients, data analysis and interpretation, manuscript writing, final approval of manuscript. Correspondence: Ofer Binah, Ph.D., Rappaport Institute, P.O. Box 9697, Haifa 31096, Israel. Telephone: 972-4-8295262; Fax: 972-48513919; e-mail: [email protected]; or Joseph Itskovitz-Eldor, M.D., Ph.D., Rambam Medical Center, Haifa 31096, Israel. Telephone: 972-4-8542536; e-mail: [email protected] Received August 12, 2008; accepted for publication September 11, 2008; first published online in STEM CELLS EXPRESS September 25, 2008. ©AlphaMed Press 1066-5099/2008/$30.00/0 doi: 10.1634/stemcells.20080777

STEM CELLS 2008;26:3130 –3138 www.StemCells.com

Sedan, Dolnikov, Zeevi-Levin et al. operated intracellular receptors (IP3Rs) are Ca2⫹ release channels, but unlike the ryanodine receptors (RyRs), IP3Rs are much less abundant in the mammalian myocardium, and their direct role in excitation contraction coupling is not entirely clear [8 –10]. To investigate this hypothesis, we tested the effects of two established IP3RS activators, angiotensin-II (AT-II) and endothelin-1 (ET-1) [11–13]. Both AT-II and ET-1 interact with their respective sarcolemmal receptors; this interaction results in activation of phospholipase C (PLC), in turn promoting 1,4,5IP3 synthesis from phosphatidylinositol 4,5-bisphosphate (PIP2) [13, 14]. Subsequently, 1,4,5-IP3 binds to the IP3Rs, resulting in release of Ca2⫹ from intracellular stores. In support of the 1,4,5-IP3 hypothesis, we discovered that (a) hESC-CM express IP3Rs, as determined by fluorescence immunostaining; (b) AT-II and ET-1 cause positive inotropic effects; and (c) the effect of AT-II (e.g., increased contraction amplitude) is blocked by U73122 (a PLC blocker) and by 2-aminoethoxyphenyl borate (2-APB), a selective inositol trisphosphate (IP3) receptor blocker. These findings demonstrate for the first time that hESC-CM exhibit functional AT-II and ET-1 signaling pathways, as well as 1,4,5-IP3-operated releasable Ca2⫹ stores.

3131

4% paraformaldehyde for 20 minutes, and incubated with 1% Triton for 10 minutes for permeabilization. After incubating the preparation with a blocking solution for 15 minutes, the primary rabbit anti-IP3 receptor polyclonal antibody (dilution, 1:75; Santa Cruz Biotechnology Inc., Santa Cruz, CA, http://www.scbt.com), the primary rabbit anti-AT-1 receptor polyclonal antibody (dilution, 1:30; Alomone Labs, Jerusalem, http://www.alomone.com), the primary rabbit anti-ET-1 A receptor polyclonal antibody (dilution, 1:180; Chemicon, Temecula, CA, http://www.chemicon.com) or the primary mouse anti-troponin I monoclonal antibody (dilution, 1:2,500; Chemicon) were added at 4°C overnight. After several washes with phosphate-buffered saline (PBS), goat anti-rabbit secondary antibody and goat anti-mouse secondary antibody were added in darkness at room temperature for 1 hour. After several washes with PBS, the coverslips with the stained preparations were mounted on glass, and 1–2 drops of 4,6-diamidino-2-phenylindole (DAPI) were added to stain the nucleus. The staining was visualized using a confocal inverted microscope.

Chemicals Unless otherwise indicated, chemicals were purchased from SigmaAldrich Israel (Rehovot, Israel, http://www.sigmaaldrich.com).

Statistical Analysis

MATERIALS

AND

METHODS

Human Embryonic Stem Cells: Culture and Differentiation Human embryonic stem cells (hESC) from clone H9.2 were grown on mouse embryonic fibroblast feeder, and EBs were prepared as previously described [7]. Daily microscopic observations were conducted to detect the first spontaneous contractions, determined as day 1. The contracting areas were carefully dissected out by microscalpel and transferred to gelatin-coated 30-mm-diameter glass slides suitable for fluorescence measurements (Fisher Scientific International, Hampton, NH, http://www.fisherscientific.com) or transferred to gelatin-coated 22 ⫻ 22 mm microscope coverslips for immunofluorescence staining.

Measurements of Intracellular Ca2ⴙ Transients and Contractions Intracellular Ca2⫹ ([Ca2⫹]i) transients and contractions were measured from small contracting areas of EBs by means of fura-2 fluorescence and video edge detector, respectively [7]. Briefly, spontaneously contracting areas of 2-week-old to 2-month-old EBs with a size range of 0.5–1 mm were mechanically dissected and adhered onto 30-mm-diameter gelatin- or collagen-coated glass slides. Subsequently, fura-2-stained contracting areas were transferred to a chamber mounted on the stage of an inverted microscope and perfused with Tyrode’s solution at 37°C [7]. The preparations were paced at a basic cycle length (BCL) 10% shorter than the spontaneous BCL; in the majority of experiments, a BCL of 500 – 1,000 milliseconds was sufficient to override the spontaneous contracting rate. To characterize the [Ca2⫹]i transient amplitude (RAmp), the differences between the maximal (systolic) ratio and the minimal (diastolic) ratio were calculated in 10 successive transients and averaged. The maximal rates of [Ca2⫹]i rise (d[Ca2⫹]/dtContrac) and relaxation (d[Ca2⫹]/dtRelax) following pacing or spontaneous contractions were calculated and averaged over 10 transients. To characterize the contraction amplitude, the differences between minimal and maximal video cursor positions (LAmp) were calculated in 10 successive contractions and averaged. The maximal rates of contraction (dL/dtContrac) and relaxation (dL/dtRelax) were calculated and averaged over 10 contractions.

Immunofluorescence Staining The spontaneously contracting areas of 3-week-old to 1-month-old EBs were dissected using a microscalpel, plated on gelatin-coated plates containing 22 ⫻ 22 mm microscope glass coverslips, fixed by

www.StemCells.com

Results were expressed as mean ⫾ SEM. Means of two populations were compared using Student’s t test for paired observations. A value of p ⬍ .05 was considered significantly different. To compare dose-response relations, one-way analysis of variance (ANOVA) test was performed followed by Dunnett’s multiple comparison test, using GraphPad Prism 5.00 software (GraphPad Software, Inc., San Diego, http://www.graphpad.com).

RESULTS The Effects of AT-II on the [Ca2ⴙ]i Transients and Contractions of hESC-CM First, we studied the effects of AT-II (10⫺9 to 10⫺6 M) on the [Ca2⫹]i transients and contractions in paced hESC-CM. As shown in Figure 1A, increasing AT-II concentration in the Tyrode’s solution superfusing the bath caused, within 2–3 minutes, a dosedependent increase in contraction amplitude. In addition, in 3 of the 12 paced preparations in which stable contractions were successfully recorded throughout the experiment, AT-II caused a positive chronotropic effect (Fig. 1C); namely, the intrinsic spontaneous rate exceeded the pacing rate. Occasionally, the chronotropic effect was associated with what appear as arrhythmogenic beats (Fig. 1D). Furthermore, in agreement with the effects of AT-II in the adult myocardium [14], AT-II elevated diastolic [Ca2⫹]i (Fig. 1E). This latter response was observed only in two (17–19 days old) of the seven preparations (ages 14 –32 days old) in which [Ca2⫹]i transients were successfully recorded. This increase in diastolic [Ca2⫹]i was transient (lasting 100 –150 seconds after addition of AT-II) and reversible, and it reappeared upon exposure to higher concentrations of AT-II. Next, we analyzed the effects of AT-II on the contractions and [Ca2⫹]i transients parameters. In agreement with the representative experiment shown in Figure 1A, the summary of these experiments demonstrates that AT-II increased in a dose-dependent manner (p ⬍ .05) the maximal rates of contraction and relaxation and the contraction amplitude (Figs. 2A–2C). Collectively, these results demonstrate that in hESC-CM at the age range of 14 –33 days (with one EB at 45 days), the AT-II signaling pathway is functional. Respecting [Ca2⫹]i transients analysis, because of a relatively small number of successful experiments, we could not generate complete dose-response relations, and the results were analyzed and displayed as follows. In different experiments, for each parameter, only the maximal response was included in the analysis, regardless of the concentration at which this re-

3132

A Functional 1,4,5-IP3 Pathway in hESC-CM

Figure 1. The effects of AT-II and Iso on the contraction of hESC-CM. (A): A representative experiment illustrating the increase in contraction amplitude in response to increasing concentrations of AT-II (the preparation was paced at 1 Hz). (B): A representative experiment illustrating an increase in contraction amplitude in response to increasing concentrations of Iso (the preparation was paced at 0.5 Hz). All six preparations studied responded similarly to Iso. (C): A representative experiment illustrating the positive chronotropic effect of AT-II (10⫺9 and 10⫺8 M) on intracellular Ca2⫹ ([Ca2⫹]i) transients and contractions recorded simultaneously from a 21-day-old EB paced at 0.5 Hz. AT-II caused a positive chronotropic effect in 3 of 12 preparations in which stable contractions were recorded during the entire experiment. (D): An experiment illustrating the initiation of arrhythmogenic beats by AT-II (10⫺8 and 10⫺7 M) in an EB paced at 0.5 Hz. (E): A representative experiment illustrating an increase in the diastolic [Ca2⫹]i in response to AT-II (10⫺8 and 10⫺7 M) in a preparation paced at 0.5 Hz. AT-II caused a similar effect in two of seven preparations in which transients were successfully recorded. Abbreviations: AT-II, angiotensin-II; F, fluorescence; hESC-CM, human embryonic stem cell-derived cardiomyocytes; Iso, isoproterenol; R, ratio; S, seconds.

sponse was recorded. In summary (Fig. 2D), AT-II increased (p ⬍ .05) the maximal rates of [Ca2⫹]i rise and relaxation and the [Ca2⫹]i transient amplitude. To determine whether the positive inotropic response to AT-II is comparable to that of ␤-adrenergic stimulation (which is known to generate a substantial positive inotropic effect), we determined in hESC-CM the effect of the ␤-adrenergic agonist isoproterenol, operating via a signaling pathway different from that of AT-II. Importantly, this is the first study in which a dose-response relation of isoproterenol with respect to its pos-

itive inotropic effect is generated in hESC-CM. As seen by a representative experiment (Fig. 1B) and Figures 3A–3C, isoproterenol caused a dose-dependent (10⫺9, 5 ⫻ 10⫺9, 10⫺8, 5 ⫻ 10⫺8, and 10⫺7 M) increase in the contraction amplitude and in the maximal rate of contraction (p ⬍ .05). Despite the seemingly large effect of isoproterenol on the maximal rate of relaxation, the one-way ANOVA yielded a nonsignificant difference (p ⬎ .05). To evaluate the respective inotropic capacity of the ␤-adrenergic and AT-II signaling pathways, the maximal inotropic responses caused by isoproterenol and AT-II were compared. As

Sedan, Dolnikov, Zeevi-Levin et al.

3133

seen in Figure 3D, the maximal effects of AT-II and isoproterenol on the maximal rates of contraction and relaxation and contraction amplitude were similar (p ⬎ .05).

The Inotropic Response of hESC-CM to AT-II Is Mediated by the 1,4,5-IP3 Signaling Pathway

Figure 2. The effects of angiotensin-II (AT-II) on the contraction and intracellular Ca2⫹ ([Ca2⫹]i) transient parameters in hESC-CM. (A–C): Dose-response relations for the effects of AT-II on the contraction parameters in hESC-CM. n ⫽ 11–12 EBs. (D): The effects of AT-II on the [Ca2⫹]i transient parameters in hESC-CM (n ⫽ 4 –7 EBs). ⴱ, p ⬍ .05 (A–D). In all panels, AT-II effects are presented as percent change of their respective controls. Abbreviations: d[Ca2⫹]i/dtContrac, maximal rate of change in intracellular Ca2⫹ concentrations during contraction; d[Ca2⫹]i/dtRelax, maximal rate of change in intracellular Ca2⫹ concentrations during relaxation; dL/ dtContrac, maximal rate of contraction; dL/dtRelax, maximal rate of relaxation; hESC-CM, human embryonic stem cell-derived cardiomyocytes; RAmp, the [Ca2⫹]i transient amplitude, which is the difference between the maximal (systolic) ratio and the minimal (diastolic) ratio.

To test the hypothesis that hESC-CM contain fully functional 1,4,5-IP3-operated intracellular Ca2⫹ stores that can be mobilized upon the appropriate physiological stimuli, we measured the responses of hESC-CM to AT-II in the presence of 2-APB, which blocks the 1,4,5-IP3 receptor [11, 15], and U73122, which blocks 1,4,5-IP3 production from PIP2 by PLC [11, 16, 17]. Importantly, as indicated by Peppiatt et al. [15], “at concentrations ⬍2 ␮M, 2-APB did not have any apparent effect on electrically evoked Ca2⫹ transients,” except for inhibiting 1,4,5-IP3-mediated release of intracellular Ca2⫹ in rat atrial myocytes. We performed the 2-APB and the U73122 experiments as follows: first, the preparation was exposed to AT-II for 5–7 minutes, and contraction was recorded. Thereafter, the preparation was superfused with drug-free Tyrode’s solution for 10 minutes, followed by exposure for 5 minutes to 2-APB (2 ␮M) or U73122 (2 ␮M), followed by 2-APB ⫹ AT-II or U73122 ⫹ AT-II. In support of the 1,4,5-IP3 hypothesis, while AT-II (10 ⫺7 or 10 ⫺8 M) increased contraction amplitude by ⬃30%, AT-II was ineffective in the presence of 2-APB (Figs. 4A, 4B). These findings are compatible with those of Peppiatt et al. [15], who nicely showed in rat atrial myocytes that in addition to abrogating hormone-evoked Ca2⫹ responses, 2-APB (2 ␮M) antagonized the Ca2⫹ signals evoked by a membrane permeant 1,4,5-IP3 ester. Next, we tested whether pretreatment with U733122 abolishes the inotropic response to AT-II. Indeed, as seen in a representative experiment in a 45-day-old EB (Fig. 4C), whereas AT-II alone caused a pronounced positive inotropic effect, the effect was prevented by the presence of 2 ␮M U73122. In summary (Figs. 4D– 4F), U73122 prevented an AT-II-induced increase in contraction amplitude and in the maximal rates of contraction and relaxation, demonstrating,

Figure 3. The effects of Iso on the contraction parameters in human embryonic stem cell-derived cardiomyocytes (hESCCM). (A–C): Dose-response relations for the effects of AT-II on the contraction parameters. n ⫽ 6 EBs. (D): Comparison of the effect of AT-II (n ⫽ 11–12 EBs) and Iso (n ⫽ 6 EBs) on the contraction parameters. In all panels, Iso effects are presented as percent change of their respective controls. ⴱ, p ⬍ .05 (A–D). Abbreviations: AT-II, angiotensin-II; dL/ dtContrac, maximal rate of contraction; dL/ dtRelax, maximal rate of relaxation; Iso, isoproterenol; LAmp, contraction amplitude.

www.StemCells.com

3134


Figure 4. The involvement of 1,4,5-inositol trisphosphate signaling pathway in the inotropic response to AT-II: 2-APB and U73122 block the positive inotropic effect of AT-II. Experimental details are given in the text. (A): A representative experiment demonstrating that 2-APB blocks the AT-II-positive inotropic effect. The 33-day-old EB was paced at 0.5 Hz. (B): A summary of four experiments in which the ability of 2-APB to block the positive inotropic effect of AT-II was investigated. ⴱ, p ⬍ .05. (C): A representative experiment demonstrating that U73122 blocks AT-II-positive inotropic effect. The 45-day-old EB was paced at 0.8 Hz. (D–F): A summary of five experiments in which the ability of U73122 to block the inotropic effect of AT-II was investigated. (D): Contraction amplitude. (E): dL/dtContrac. (F): dL/dtRelax. ⴱ, p ⬍ .05. Abbreviations: 2-APB, 2-aminoethoxyphenyl borate; AT-II, angiotensin-II; dL/dtContrac, maximal rate of contraction; dL/dtRelax, maximal rate of relaxation; hESC-CM, human embryonic stem cellderived cardiomyocytes.

along with the 2-APB results, that the effects of AT-II in hESC-CM are mediated by the 1,4,5-IP3 signaling pathway.

The Expression of the AT-II and 1,4,5-IP3 Receptors in hESC-CM To further establish the expression of the AT-II-1,4,5-IP3 signaling pathway in hESC-CM, we determined, by means of confocal immunofluorescence staining analysis, whether the AT-II receptor and the IP3Rs are expressed in contracting areas from 21–30-day-old EBs. As seen in Figures 5 (IP3 receptor analysis) and 6 (AT-II receptor analysis), each picture depicts the three individual confocal layers of the immunofluorescence staining: the upper left layer is the DAPI staining of the nuclei (blue staining), the upper right layer is the staining of the IP3Rs or AT-II receptor (red staining), the bottom left layer is the staining of troponin I (a cardiac-specific protein; green staining), and the bottom right layer is the merge of all three layers. In summary, in support of the 1,4,5-IP3 hypothesis, the contracting areas of two different EBs express IP3Rs, as indicated by the red staining in the cytoplasm of cardiomyocyte (positively stained for troponin I). Altogether, positive staining for IP3Rs was repeated in three different EBs. In further support of the 1,4,5-IP3 hypothesis, the contracting

areas of three different EBs express AT-II receptor, expressed as red staining mainly at the cell perimeter. Altogether, positive staining for the AT-II receptor was repeated in four different EBs.

hESC-CM Respond to Endothelin-1 and Express the Respective Receptor To strengthen the 1,4,5-IP3 hypothesis, contracting EBs were exposed to ET-1 (10⫺9, 10⫺8, and 10⫺7 M), which augments myocardial contraction and causes arrhythmias by releasing Ca2⫹ from 1,4,5-IP3-dependent intracellular stores [11, 12]. As shown by the representative experiment (Fig. 7A) and by the summary of 4 –5 experiments (Fig. 7B), ET-1 increased the maximal rates of contraction and relaxation and the contraction amplitude. Occasionally, ET-1 caused arrhythmias (data not shown), an effect compatible with the effect of ET-1 in rat ventricular myocytes [11]. Furthermore, as depicted in Figure 7B, the magnitudes of the inotropic responses of hESC-CM to ET-1 and AT-II were practically identical. Finally, as shown in Figure 7C, in agreement with the functional findings, the contracting area of a 31-day-old EB express the ET-1 receptor, as indicated by the intense red staining in the troponin I-positive cardiomyocytes. Similar staining for the ET-1 receptor was


3135

17– 45 days, hESC-CM contain functional 1,4,5-IP3-operated Ca2⫹ stores, as evidenced by the inotropic, lusitropic, and chronotropic responses to AT-II and ET-1. (b) In support of the functionality of the 1,4,5-IP3 pathway, we determined that the positive inotropic response to AT-II was blocked by the 1,4,5IP3 receptor blocker 2-APB and by the PLC blocker U73122, and that hESC-CM express 1,4,5-IP3, AT-II, and ET-1 receptors. These findings suggest that hESC-CM express the functional 1,4,5-IP3 signaling pathway responsible for the effects of AT-II and ET-1.

The Responses to AT-II and ET-1 Are Dependent on an Intact 1,4,5-IP3 Pathway In the first series of experiments in hESC-CM we found that at the concentration range of 10⫺9 to 10⫺6 M, AT-II increased the amplitude and the maximal rates of rise and decline of the [Ca2⫹]i transient. Accordingly, AT-II increased the maximal rates of contraction and relaxation and the contraction amplitude. Furthermore, AT-II also caused positive chronotropic and arrhythmogenic effects (Fig. 1), which are compatible with the findings of Proven et al. in rat ventricular myocytes [11]. Similarly, ET-1 caused inotropic and lusitropic effects in hESC-CM (Fig. 7). Since both AT-II and ET-1 are known activators of the 1,4,5-IP3 pathway, the positive responses to these agonists support the hypothesis that hESC-CM contain functional 1,4,5-IP3 pathway.

The Responses of hESC-CM to AT-II and Isoproterenol In addition to the above-mentioned findings, this study demonstrates for the first time in hESC-CM dose-response relations of the ␤-adrenergic agonist isoproterenol. To determine whether the inotropic responses of hESC-CM to activation of the ␤-adrenergic and 1,4,5-IP3 pathways are similar, we compared the effects of isoproterenol and AT-II. Since the effects of AT-II and isoproterenol on the contraction parameters (Fig. 3D) were alike, we concluded that these responses probably represent the maximal inotropic capacity of hESC-CM. In summary, our findings show that hESC-CM express functional 1,4,5-IP3 pathway, responsive to physiological agonists such as AT-II and ET-1.

The Functionality of the 1,4,5-IP3 Signaling Pathway in hESC-CM

Figure 5. The expression of 1,4,5-IP3 receptors in two different EBs. Spontaneously contracting EBs (n ⫽ 3 EBs) were fixed and processed as described in Materials and Methods. (A, B): Upper left panel, 4,6-diamidino-2-phenylindole staining (blue) of nuclei; upper right panel, red staining for 1,4,5-IP3 receptor; lower left panel, green staining for troponin I; lower right panel, merge of all three layers. (A): A 25-day-old EB; magnification, 2 ⫻ 40. (B): A 30-day-old EB; magnification, 4 ⫻ 40. Abbreviation: 1,4,5-IP3, 1,4,5-inositol trisphosphate.

evidenced in all three EBs (30, 30, and 31 days old) analyzed for the ET-1 receptor expression.

DISCUSSION In the present work we tested the hypothesis that hESC-CM feature 1,4,5-IP3-operated intracellular Ca2⫹ stores that can be mobilized upon physiological stimuli, for example by AT-II or ET-1. Our major findings are as follows. (a) At the age range of www.StemCells.com

Although several studies have suggested that in hESC-CM the excitation-contraction coupling machinery is not fully compatible with the adult myocardium, this study shows that hESC-CM can mobilize intracellular Ca2⫹ by activating intracellular IP3Rs. Importantly, our previous findings [7] suggesting that SR Ca2⫹ release is not the major contributor to contraction are not in conflict with the present findings, which demonstrate that hESC-CM contain 1,4,5-IP3-dependent Ca2⫹ stores, since these two Ca2⫹ release stores are dissimilar. Numerous reports have shown that IP3Rs are ubiquitous intracellular Ca2⫹ release channels that are present in atrial and ventricular muscle, albeit at lower levels than those of the RyRs [8, 9, 18 –20]. As indicated by several authors [18 –20], the role of IP3Rs in excitation contraction coupling and cardiac function in adult mammalian heart is still not entirely clear. According to Bers [10], although IP3Rs have at most a minor effect on cardiac excitation contraction coupling, they have other distinct roles such as control of cardiac excitation-transcription coupling. Furthermore, Marks [9] suggested that IP3Rs play a role in activating [Ca2⫹]i-dependent enzymes in the heart and in regulating membrane per-

3136


Figure 6. The expression of AT-II receptors in human embryonic stem cell-derived cardiomyocytes. Spontaneously contracting EBs were fixed and processed as described in Materials and Methods. (A–C): Upper left panel, 4,6-diamidino-2-phenylindole staining (blue) of nuclei; upper right panel, red staining for AT-II receptor; lower left panel, green staining for troponin I; lower right panel, merge of all three layers. (A): A 25day-old EB; magnification, 2 ⫻ 40. (B): A 30-day-old EB; magnification, 4 ⫻ 40. (C): A 29 old EB; magnification, 4 ⫻ 40. Abbreviation: AT-II, angiotensin-II.

meability, such as the nuclear membrane, since IP3Rs are present there as well. In a more recent report, Zima and Blatter [13] suggested that IP3Rs have a modulatory effect on cat atrial excitation-contraction coupling, particularly on calcium induced calcium release (CICR), by facilitating [Ca2⫹]i release via RyRs. These authors hypothesized that [Ca2⫹]i release through IP3Rs alters [Ca2⫹]i in the microenvironment of the RyRs, which in turn facilitates [Ca2⫹]i release from SR through RyRs (thus enhancing CICR). In contrast to the debated role of IP3Rs in the mature myocardium, several studies have proposed an important function of these receptors during development and embryogenesis. For example, IP3Rs mRNA is detected at high levels in the developing heart tube, whereas at the same developing stage (8.5 days after egg fertilization), RyRs mRNA is only beginning to be detected. Shortly after birth, when the mitotic potential of cardiomyocytes is lost, IP3Rs expression declines and RyRs expression is on the rise [21]. These findings suggest that IP3Rs are required at least initially for the proliferation and differentiation of developing cardiomyocytes. On the basis of these and other findings, it was assumed that the first event of intracellular

Ca2⫹ cycling within the myocytes is via the IP3R signaling pathway rather than by RyRs. Importantly, this alternative Ca2⫹ signaling pathway probably underlies the initiation of cardiac pacemaking [22–24] and thus constitutes a critical step in cardiogenesis. In agreement with our current findings, Satin et al. [25] provided experimental evidence supporting the functionality of the 1,4,5-IP3 pathway in hESC-CM with respect to “action-potential-induced Ca transients,” but not of the contraction. In brief, hESC-CM were exposed to a low concentration of 2-APB (2 ␮M), which completely eliminated (in a reversible manner) all Ca2⫹ transients in 4 of 22 cells tested and decreased Ca2⫹ transient amplitude by 30% ⫾ 20% in 18 of 22 cells tested.

Study Limitations and Heterogeneity of Excitation-Contraction Coupling Maturation in hESC-CM As discussed in our recent publication [7], the contracting areas of the EBs are heterogeneous and are composed, among noncardiomyocytes, of ventricular, atrial, and nodal-like myocytes. Consequently, the responses ([Ca2⫹]i transients


3137

Figure 7. The effects of ET-1 on the contraction parameters in hESC-CM and the expression of the ET-1 receptor. (A): A representative experiment illustrating the positive inotropic effect of ET-1 (10⫺9 and 10⫺8 M) on the contraction recorded from a 56day-old EB paced at 0.3 Hz. (B): A comparison of the effects of angiotensin-II (n ⫽ 11–12 EBs) and ET-1 (n ⫽ 4 –5 EBs) on the contraction parameters. ⴱ, p ⬍ .05. (C): A 31-day-old spontaneously contracting EB was fixed and stained for ET-1 as described in Materials and Methods. Magnification, 2 ⫻ 40. Upper left panel, 4,6-diamidino-2phenylindole staining (blue) of nuclei; upper right panel, red staining for ET-1 receptor; lower left panel, green staining for troponin I; lower right panel, merge of all three layers. Abbreviations: dL/dtContrac, maximal rate of contraction; dL/dtRelax, maximal rate of relaxation; ET-1, endothelin-1; hESCCM, human embryonic stem cell-derived cardiomyocytes; LAmp, contraction amplitude; S, seconds.

and contractions) recorded from the contracting areas isolated from EBs represent the weighted responses of all cell types. However, since the majority of myocytes within the contracting areas are ventricular-like [5, 6], it is reasonable to conclude that these responses represent the ventricular phenotype. Under these experimental conditions, on the basis of functional, pharmacological, and molecular analyses, we proposed that in contracting EBs, the excitation-contraction coupling appears immature. In general agreement with our previous findings, Liu et al. [26] have shown, in small-cell clusters (as opposed to larger EBs used in our study) containing 10 –15 cells from different (we used clone H9.2) hESC clones (H1 and HES2), that ⬃2⁄3 of the clusters were unresponsive to caffeine, ryanodine, and thapsigargin, indicating a nonfunctional SR. The residual ⬃1⁄3 of the clusters exhibited functional (albeit immature) SR, as indicated by a response to caffeine, ryanodine, and thapsigargin. Contrasted with our study and that of Liu et al. [26], Satin et al. [25] have shown in dispersed cells functional SR, as indicated by “mature-like” pharmacological responses of [Ca2⫹]i transients to ryanodine and caffeine. However, since Satin et al. [25] studied dispersed cell (the phenotype of which was not indicated) and contractions were not measured from these cells, it remains to be determined whether the contraction per se was dependent on SR Ca2⫹ stores. www.StemCells.com

SUMMARY In the present study, we tested the hypothesis that hESC-CM contain 1,4,5-IP3-operated Ca2⫹ stores that can be mobilized upon the appropriate physiological stimuli. Furthermore, these responses were found to be compatible with the responses induced by ␤-adrenergic stimulation. In conclusion, these findings demonstrate for the first time that hESC-CM exhibit fully functional AT-II and ET-1 signaling pathways, operating via the 1,4,5-IP3-operated releasable Ca2⫹ stores.

ACKNOWLEDGMENTS This work was supported by the Israel Science Foundation, the Israeli Ministry of Science and Technology, and the Rappaport Family Institute for Research in the Medical Sciences.

DISCLOSURE

OF

POTENTIAL CONFLICTS

OF INTEREST The authors indicate no potential conflicts of interest.


3138

REFERENCES 1 2 3 4 5 6 7

8 9 10 11 12 13

Mummery C. Cardiomyocytes from human embryonic stem cells: More than heart repair alone. Bioessays 2007;29:572–579. Klug MG, Soonpaa MH, Koh GY et al. Genetically selected cardiomyocytes from differentiating embryonic stem cells form stable intracardiac grafts. J Clin Invest 1996;98:216 –224. van Laake LW, Hassink R, Doevendans PA et al. Heart repair and stem cells. J Physiol 2006;577:467– 478. Kehat I, Kenyagin-Karsenti D, Snir M et al. Human embryonic stem cells can differentiate into myocytes with structural and functional properties of cardiomyocytes. J Clin Invest 2001;108:407– 414. He JQ, Ma Y, Lee Y et al. Human embryonic stem cells develop into multiple types of cardiac myocytes. Action potential characterization. Circ Res 2003;93:32–39. Mummery C, Ward-van Oostwaard D, Doevendans P et al. Differentiation of human embryonic stem cells to cardiomyocytes: Role of coculture with visceral endoderm-like cells. Circulation 2003;107:2733–2740. Dolnikov K, Shilkrut M, Zeevi-Levin N et al. Functional properties of human embryonic stem cell-derived cardiomyocytes: Intracellular Ca2⫹ handling and the role of sarcoplasmic reticulum in the contraction. STEM CELLS 2006;24:236 –245. Perez PJ, Ramos-Franco J, Fill M et al. Identification and functional reconstitution of the type 2 inositol 1,4,5-trisphosphate receptor from ventricular cardiac myocytes. J Biol Chem 1997;272:23961–23969. Marks AR. Cardiac intracellular calcium release channels: Role in heart failure. Circ Res 2000;87:8 –11. Bers DM. Cardiac excitation-contraction coupling. Nature 2002;415: 198 –205. Proven A, Roderick HL, Conway SJ et al. Inositol 1,4,5-trisphosphate supports the arrhythmogenic action of endothelin-1 on ventricular cardiac myocytes. J Cell Sci 2006;119:3363–3375. Li X, Zima AV, Sheikh F et al. Endothelin-1-induced arrhythmogenic Ca2⫹ signaling is abolished in atrial myocytes of inositol-1,4,5-trisphosphate (IP3)-receptor type 2-deficient mice. Circ Res 2005;96:1274 –1281. Zima AV, Blatter LA. Inositol-1,4,5-trisphosphate-dependent Ca2⫹ signaling in cat atrial excitation-contraction coupling and arrhythmias. J Physiol 2004;555:607– 615.

14 Petroff MG, Aiello EA, Palomeque J et al. Subcellular mechanisms of the positive inotropic effect of angiotensin II in cat myocardium. J Physiol 2000;529:189 –203. 15 Peppiatt CM, Collins TJ, Mackenzie L et al. 2-aminoethoxydiphenyl borate (2-APB) antagonises inositol 1,4,5-trisphosphate-induced calcium release, inhibits calcium pumps and has a use-dependent and slowly reversible action on store-operated calcium entry channels. Cell Calcium 2003;34:97–108. 16 Vassilopoulos D, Smallridge RC, Tsokos GC. Effects of an aminosteroid inhibitor of phospholipase C-dependent processes on the TCR-mediated signal transduction pathway in human T cells. Clin Immunol Immunopathol 1995;77:59 – 68. 17 Barac DY, Zeevi-Levin N, Yaniv G et al. The 1,4,5-inositol trisphosphate pathway is a key component in Fas-mediated hypertrophy in neonatal rat ventricular myocytes. Cardiovasc Res 2005;68:75– 86. 18 Mackenzie L, Roderick HL, Proven A et al. Inositol 1,4,5-trisphosphate receptors in the heart. Biol Res 2004;37:553–557. 19 Woodcock EA, Matkovitch SJ. Ins(1,4,5)P3 receptors and inositol phosphates in the heart: Evolutionary artefacts or active signal transducers? Pharmacol Ther 2005;107:240 –251. 20 Wu X, Zhang T, Bossuyt J et al. Local InsP3-dependent perinuclear Ca2⫹ signaling in cardiac myocyte excitation-transcription coupling. J Clin Invest 2006;116:675– 682. 21 Rosemblit N, Moschella MC, Ondriasa´ E et al. Cardiac intracellular calcium release channels: Role in heart failure. Dev Biol 1999;206: 163–177. 22 Kapur N, Banach K. Inositol-1,4,5-trisphosphate-mediated spontaneous activity in mouse embryonic stem cell-mediated cardiomyocytes. J Physiol 2007;581:1113–1127. 23 Me´ry A, Aimond F, Me´nard C et al. Initiation of embryonic cardiac pacemaker activity by inositol 1,4,5-trisphosphate-dependent calcium signaling. Mol Biol Cell 2005;16:2414 –2423. 24 Roderick HL, Bootman MD. Pacemaking, arrhythmias, inotropy and hypertrophy: The many possible facets of IP3 signaling in cardiac myocytes. J Physiol 2007;581:883– 884. 25 Satin J, Itzhaki I, Rapoport S et al. Calcium handling in human embryonic stem cell-derived cardiomyocytes. STEM CELLS 2008;26:1961–1972. 26 Liu J, Fu JD, Siu CW et al. Functional sarcoplasmic reticulum for calcium handling of human embryonic stem cell-derived cardiomyocytes: Insights for driven maturation. STEM CELLS 2007;25:3038–3044.