Human stem cell-derived cardiomyocytes provide new models for studying the ion channel pharmacology of human cardiac cells for both drug discovery and safety pharmacology purposes. However, detailed pharmacological characterization of ion channels in stem cell-derived cardiomyocytes is lacking. Therefore, we used patch-clamp electrophysiology to perform a pharmacological survey of the L-type Ca2+ channel in induced pluripotent and embryonic stem cell-derived cardiomyocytes and compared the results with native guinea pig ventricular cells. Six structurally distinct antagonists [nifedipine, verapamil, diltiazem, lidoflazine, bepridil, and 2-[(cis-2-phenylcyclopentyl)imino]-azacyclotridecane hydrochloride (MDL 12330)] and two structurally distinct activators [methyl 2,6-dimethyl-5-nitro-4-[2-(trifluoromethyl)phenyl]-1,4-dihydropyridine-3-carboxylate (Bay K8644) and 2,5-dimethyl-4-[2-(phenylmethyl)benzoyl]-1H-pyrrole-3-carboxylic acid methyl ester (FPL 64176)] were used. The IC50 values for the six antagonists showed little variability between the three cell types. However, whereas Bay K8644 produced robust increases in Ca2+ channel current in guinea pig myocytes, it failed to enhance current in the two stem cell lines. Furthermore, Ca2+ channel current kinetics after addition of Bay K8644 differed in the stem cell-derived cardiomyocytes compared with native cells. FPL 64176 produced consistently large increases in Ca2+ channel current in guinea pig myocytes but had a variable effect on current amplitude in the stem cell-derived myocytes. The effects of FPL 64176 on current kinetics were similar in all three cell types. We conclude that, in the stem cell-derived myocytes tested, L-type Ca2+ channel antagonist pharmacology is preserved, but the pharmacology of activators is altered. The results highlight the need for extensive pharmacological characterization of ion channels in stem cell-derived cardiomyocytes because these complex proteins contain multiple sites of drug action.
Voltage-dependent L-type Ca2+ channels provide the main pathway for Ca2+ influx into the heart, and their functioning is crucial for controlling electrical activity and excitation-contraction coupling. L-type Ca2+ channels are complex heteromeric proteins comprised minimally of α1, α2/δ, and β subunits and are modulated by numerous intracellular processes (for reviews see Bodi et al., 2005 and Benitah et al., 2010). The pharmacology of the L-type Ca2+ channel is also complex but has been very well characterized. Numerous structurally distinct antagonists, working at allosterically coupled sites on the channel, have been discovered, and several of these (e.g., dihydropyridines, benzothiazepines, and phenylalkylamines) are widely used in the treatment of cardiovascular diseases (Fleckenstein, 1983; Triggle, 1999). Although not as numerous, activators of L-type Ca2+ channels have also been discovered and have profound effects on both channel activity and cardiac functioning (Rampe and Kane, 1994). L-type Ca2+ channels are also actively studied as antitargets during drug safety testing because off-target interactions with these channels can lead to unwanted or dangerous cardiovascular side effects, including altered cardiac contractility and conduction disturbances.
In recent years cardiomyocytes derived from human stem cells have been developed and provide a new model for the study of the physiology and pharmacology of the human heart. The cells that have been developed include induced pluripotent stem cell-derived cardiomyocytes and embryonic stem cell-derived cardiomyocytes (Thomson et al., 1998; He et al., 2003; Zhang et al., 2009). In addition to their promise in the field of drug discovery and regenerative medicine, these cells are now becoming more popular as surrogates for human cardiac tissue for use in safety and toxicity testing (Anson et al., 2011). Cardiac safety testing is often focused on the interactions of drugs with various voltage-dependent ion channels because any such interaction has the potential to produce arrhythmia. Therefore, the utility of stem cell-derived cardiac myocytes, either for drug discovery or drug safety testing, will ultimately depend on whether the detailed pharmacological profile of their ion channels faithfully recapitulate those found in native myocytes. To begin an exploration of this area we have surveyed the pharmacology of the L-type Ca2+ channel in both induced pluripotent and embryonic stem cell-derived cardiomyocyte cell lines and compared the results obtained with native myocytes derived from guinea pig heart.
Materials and Methods
Single ventricular myocytes were isolated from guinea pigs as described previously (Kang et al., 2004). Male Hartley guinea pigs were anesthetized with 5% isoflurane (Baxter Healthcare Corp., Deerfield IL) in a mixture of nitrous oxide and oxygen (1:1). A thoracotomy was performed, and the heart was removed and immediately transferred to oxygenated (100%) cold saline. The heart was perfused retrogradely at 10 ml/min through the aorta with oxygenated Ca2+-free saline at 37°C in three stages: first with standard Ca2+-free saline for 5 min, second with the same solution containing 280U/ml type II collegenase (Worthington Biochemicals, Freehold, NJ) plus 0.75 U/ml type XIV protease (Sigma-Aldrich, St. Louis, MO) for 8 min, and finally with saline containing 0.2 mM CaCl2 for an additional 7 min. The left ventricle was cut into small pieces and gently shaken at room temperature for approximately 5 min to disperse single myocytes. The isolated myocytes were then maintained at 10°C for electrophysiological recordings.
Human induced pluripotent stem cell-derived cardiomyocytes (hiPSC-CM) were purchased from Cellular Dynamics International (iCell Cardiomyocytes; Cellular Dynamics International, Madison, WI) and cultured for single-cell electrophysiology recordings as instructed by the manufacturer. In brief, frozen vials of hiPSC-CM were thawed in a 37°C water bath. The thawed cells were mixed with 10 ml of ice-cold plating medium (iCell Cardiomyocyte Plating Medium; Cellular Dynamics International). The cells were diluted to approximately 20,000 to 40,000 in 2 ml of cold plating medium, and this cell suspension was transferred into 12-well culture plates containing 0.1% gelatin-coated glass coverslips. Cells were maintained in a tissue culture incubator at 37°C in an atmosphere of 93% air and 7% CO2. After 2 days of culture, the plating medium was replaced with a cell culture medium (iCell Cardiomyocyte Maintenance Medium; Cellular Dynamics International). This medium was changed every 48 h. Cells were maintained on the coverslips for 4 to 14 days before electrophysiological recordings.
Human embryonic stem cell-derived cardiomyocytes (hESC-CM) were purchased from Geron Corporation (Menlo Park, CA) and cultured for single-cell electrophysiology recordings as instructed by the manufacturer. In brief, frozen vials of hESC-CM were thawed in a 37°C water bath. The thawed cells were mixed with 10 ml of prewarmed RPMI 1640/B27 and centrifuged at 400g for 4 min. The cells were then resuspended in RPMI 1640/B27 at a concentration of approximately 20,000 to 40,000 cells/ml placed into six-well plates containing Matrigel-coated (BD Biosciences, San Jose, CA) glass coverslips. Cells were maintained in a tissue culture incubator at 37°C in an atmosphere of 95% air and 5% CO2. After 2 days and every other day thereafter, the medium was changed with RPMI 1640/B27. Cells were maintained on the coverslips for 4 to 14 days before electrophysiological recordings.
All Ca2+ channel currents were recorded at room temperature by using the whole-cell configuration of the patch-clamp techniques (Hamill et al., 1981). Electrodes (1–3 MΩ resistance) were made from TW-150F glass capillary tubes (WPI, Sarasota, FL). Electrodes were filled with a solution containing 130 mM cesium methanesulfonate, 20 mM tetraethylammonium chloride, 1 mM MgCl2, 10 mM EGTA, 10 mM HEPES, 4 mM Tris-ATP, 0.3 mM Tris-GTP, 14 mM phosphocreatine, and 50 U/ml creatine phosphokinase, pH 7.2 with CsOH. The external solution for Ca2+ channel recordings contained 137 mM NaCl, 5.4 mM CsCl, 1.8 mM CaCl2, 1 mM MgCl2, 10 mM HEPES, and 10 mM glucose, pH 7.4 with NaOH. Ca2+ channel currents were recorded by using an Axopatch 200B amplifier (Danaher, Inc. Sunnyvale, CA). Currents were analyzed by using the pCLAMP suite of software (Danaher, Inc). IC50 values were obtained by nonlinear least-squares fit of the data (GraphPad Software, Inc., San Diego, CA).
Lidoflazine and dimethyl-4-[2-(phenylmethyl)benzoyl]-1H-pyrrole-3-carboxylic acid methyl ester (FPL 64176) were obtained from Sigma-Aldrich. Bepridil, diltiazem, and S-(−)-methyl 2,6-dimethyl-5-nitro-4-[2-(trifluoromethyl)phenyl]-1,4-dihydropyridine-3-carboxylate (Bay K8644) were obtained from Tocris Bioscience (Ellisville, MD). Nifedipine was obtained from Acros (Geel, Belgium). Verapamil was obtained from Enzo Life Sciences Inc. (Farmingdale, NY), and 2-[(cis-2-phenylcyclopentyl)imino]-azacyclotridecane hydrochloride (MDL 12330) was synthesized at sanofi pharmaceuticals (Bridgewater, NJ). All other chemicals were obtained from Sigma-Aldrich.
Figure 1 illustrates the chemical structures of the Ca2+ channel antagonists and activators that were used in this study. The effects of the dihydropyridine Ca2+ channel antagonist nifedipine on Ca2+ channel currents recorded from native guinea pig ventricular myocytes, hiPSC-CM, and hESC-CM are shown in Fig. 2. All cells were held at −40 mV and depolarized for 200 ms to a test potential of 0 mV to elicit L-type Ca2+ channel currents. Consistent with its high-affinity block of this channel, nifedipine inhibited Ca2+ channel currents in guinea pig myocytes with an IC50 value of 9 nM [8–10 nM, 95% confidence limits (CL); Fig. 2, A and D]. Likewise, nifedipine produced a similar high-affinity block of the L-type Ca2+ channels recorded in both the hiPSC-CM and hESC-CM cell lines. In the hiPSC-CM cell line nifedipine inhibited Ca2+ channel currents with an IC50 value of 3 nM (2–4 nM, 95% CL; Fig. 2, B and D), whereas in the hESC-CM cell line this value measured 6 nM (5–7 nM, 95% CL; Fig. 2, C and D).
A wide variety of structurally distinct molecules are known to act as antagonists of the L-type Ca2+ channel. Six of these molecules were tested for their ability to block L-type Ca2+ channel currents in guinea pig myocytes and in hiPSC-CM and hESC-CM cell lines. For these studies, cells were held at −40 mV and depolarized to 0 mV for 200 ms, and peak inward currents in the absence and presence of ascending concentrations of drugs were used to generate dose-response relationships and corresponding IC50 values. As shown in Table 1, the IC50 values for any particular antagonist differed little between the three different cell types tested with variations of approximately 3-fold or less.
The dose-response relationships for the L-type Ca2+ channel activator Bay K8644 in guinea pig myocytes, hiPSC-CM, and hESC-CM are shown in Fig. 3. As expected, Bay K8644 produced a dose-dependent increase in peak Ca2+ channel current in guinea pig myocytes. The increase in current was evident throughout the dose-response relationship, reaching a maximum of 267 ± 54% over the control value (set at 100%) at 100 nM, the highest concentration tested (Fig. 3, A and D; n = 8). In addition, Bay K8644 accelerated Ca2+ current inactivation during the depolarizing step pulses. Compared with the results obtained in guinea pig myocytes, the effects of Bay K8644 were dramatically different in both hiPSC-CM and hESC-CM. In 23 hiPSC-CM cells tested we could find only three cells that gave responses similar to those observed in guinea pig myocytes. In 22 hESC-CM cells examined, none produced a response similar to that seen in the guinea pig myocytes. Instead, Bay K8644 produced either no or very little stimulation of calcium channel current amplitude in these cell lines or a moderate (10–40% at 100 nM) inhibition of the current. Further additional exposure to 1 μM Bay K8644 in some of these cells also failed to produce an enhancement of current. In addition, Bay K8644 slowed Ca2+ channel current inactivation especially during the initial fast component of current decay. Typical current traces in the presence and absence of Bay K8644 for hiPSC-CM and hESC-CM are illustrated in Fig. 3, B and C, respectively, and the dose-response relationship is shown in Fig. 3D.
Bay K8644 also had effects on L-type Ca2+ channel activation kinetics, and this was easiest to see by using short depolarizing pulses. Figure 4 shows the effects of 100 nM Bay K8644 on guinea pig myocytes, hiPSC-CM, and hESC-CM held at −40 mV and depolarized to 0 mV for 30 ms. Because the large-capacity transients made measuring the time constant of channel activation in the guinea pig myocytes difficult, we instead measured time to peak current. In the absence of Bay K8644 this value measured 11.4 ± 2.4 ms, whereas after the addition of Bay K8644 this value was slightly, although not significantly, shortened and measured 10.5 ± 0.8 (Fig. 4A; n = 8). Time to peak current in the hiPSC-CM (n = 11) and hESC-CM (n = 9) measured 8.7 ± 0.7 and 9.7 ± 1.2, respectively. After addition of Bay K8644 these values were significantly prolonged, measuring 14.2 ± 0.9 and 14.9 ± 1.0 ms, respectively (Fig. 4, B and C; p < 0.05; paired t test). The one effect of Bay K8644 that was consistent between the guinea pig myocytes and all of the stem cell-derived myocytes tested was its ability to prolong Ca2+ channel tail current upon repolarization of the cells (Fig. 4, arrows).
Figure 5 illustrates the effects of Bay K8644 on the L-type Ca2+ channel current-voltage (I-V) relationship measured in guinea pig myocytes, hiPSC-CM, and hESC-CM. To generate the I-V relationships, all cells were held at −40 mV and depolarized for 30 ms to potentials ranging from −40 to +40 mV in 5-mV increments, and peak currents were recorded. All currents were normalized to that obtained during the 10-mV depolarizing step in the absence of Bay K8644. Bay K8644 produced large increases in calcium current in the guinea pig myocytes over a wide range of potentials. The peak of the I-V relationship was shifted from 15 mV in control to 5 mV after the addition of Bay K8644 (Fig. 5A). By comparison, Bay K8644 had little effect on current amplitude at the various test potentials and produced a 5-mV shift (from 10 to 5 mV) in the peak of the I-V relationships in both the hiPSC-CM (Fig. 5B) and hESC-CM (Fig. 5C).
Figure 6 examines the effects of the benzoylpyrrole Ca2+ channel activator FPL 64176 on L-type Ca2+ channel currents recorded from guinea pig myocytes, hiPSC-CM, and hESC-CM. FPL 64176 (30–1000 nM) produced a dose-dependent increase in Ca2+ current amplitude in all guinea pig myocytes tested with a maximum increase of 336 ± 97% over the control value at the top concentration of 1000 nM (Fig. 6, A and D). At least a 2-fold increase in current amplitude was seen in every cell, and this was accompanied by a dramatic slowing of channel activation, inactivation, and tail current decay (Fig. 6A). The effects of FPL 64176 were less predictable in the hiPSC-CM and hESC-CM cell lines. In hiPSC-CM FPL 64176 produced a 2-fold or larger increase in calcium current amplitude in 6/14 cells tested, an increase in current amplitude that was less than 2-fold (25–82% increase) in 5/14 cells (Fig. 6B), and a slight decrease (15–43%) in current amplitude in 3/14 cells. When pooled together, FPL 64176 produced a maximal increase in Ca2+ channel current amplitude of 105 ± 29% over the control value (Fig. 6D). In hESC-CM the effects of FPL 64176 on Ca2+ current amplitude were somewhat less, producing a 2-fold or larger increase in current in 2/16 of the cells tested, a 15 to 55% maximal increase in current in 4/16 cells (Fig. 6C) and no increase or a slight decrease (up to 40%) in current amplitude in 10/16 cells tested. When pooled together, FPL 64176 produced a 7 ± 11% increase in current amplitude over the control value in hESC-CM at the 1 μM concentration (Fig. 6D). Regardless of the effects of FPL 64176 on current amplitude, the drug always produced the characteristic slowing of current activation, inactivation, and tail current decay in all hiPSC-CM and hESC-CM tested.
Human stem cell-derived cardiomyocytes offer a new approach for studying the cellular electrophysiological characteristics of the heart. However, detailed pharmacological characterization of the main cardiac ion channels in these cells requires further study. In the present article, we explore the pharmacology of the L-type Ca2+ channel in two distinct stem cell-derived cardiomyocyte cell lines and compare it with that of native myocytes isolated from guinea pig heart. The stem cell-derived myocyte cell lines used in this study were chosen because they were commercially available, they represent two distinct lineages (induced pluripotent and embryonic), and their basic phenotypic and electrophysiological characteristics have been described previously (Peng et al., 2010; Ma et al., 2011). We chose to study the L-type Ca2+ channel because it possess a rich and well characterized pharmacology comprised of many structurally distinct antagonists and activators, and because it is a target for therapeutic intervention as well as an antitarget for use in drug safety assessment.
The L-type Ca2+ channel is known to be inhibited by a variety of compounds working at distinct allosterically coupled sites, and we chose six of these to probe the pharmacology of this channel in the hiPSC-CM and hESC-CM cell lines and compare the results with those obtained from native guinea pig myocytes. These drugs included representatives from the clinically popular dihydropyridine (nifedipine), phenylalkylamine (verapamil), and benzothiazepine (diltiazem) classes (Triggle and Janis, 1987), along with lidoflazine (Barry et al., 1985), bepridil (Yatani et al., 1986), and MDL 12330 (Rampe et al., 1987). The IC50 values obtained for all of these drugs were similar to those reported in the literature and displayed a variance of approximately 3-fold or less between the three cell types studied. In terms of L-type Ca2+ channel antagonist pharmacology, the affinity of a wide variety of structurally diverse molecules is preserved in both hiPSC-CM and hESC-CM cell lines compared with native mammalian cardiomyocytes.
It was not until we examined the effects of L-type Ca2+ channel activators that we found pharmacological divergence between stem cell-derived myocytes and native ones. This was particularly true for the dihydropyridine Bay K8644. Bay K8644 is well known as an activator of L-type Ca2+ channels in a wide variety of cells, and its stimulatory activity is present regardless of whether the cells come from primary cultures or cell lines grown under a various culture conditions (for reviews see Schramm and Towart, 1985 and Bechem and Schramm, 1987). In the current study, Bay K8644 failed to produce its characteristic increases in current amplitude in either the hiPSC-CM or hESC-CM and furthermore slowed Ca2+ channel activation and inactivation in the stem cell-derived myocytes. These activities are inconsistent with the well described properties of Bay K8644 found in guinea pig myocytes in this and other studies (Hamilton et al., 1987; Zhong et al., 1997) and most importantly in human cardiac myocytes from atrial (Le Grand et al., 1991; Skasa et al., 2001) and ventricular preparations (Chen et al., 1999, 2002, 2008) where Bay K8644 has effects similar to those found in the guinea pig and other mammalian cells. The lack of effect of Bay K8644 in the stem cell lines cannot therefore be considered a species-dependent effect, but rather some specific alteration in the stem cells themselves. The only typical pharmacological characteristic of Bay K8644 that seemed to be preserved in all of the stem cell-derived myocytes tested was a prolongation of tail current decay. Increases in Ca2+ channel current amplitude to the benzoylpyrrole activator FPL 64176 were somewhat variable in the stem cell-derived myocytes with some cells showing large increases in current amplitude, whereas others did not. Characteristically large (defined here as more than 2-fold) increases in Ca2+ channel current were observed only in some of the stem cells tested. Despite this inconsistency in current amplitude, the other classic pharmacological effects of FPL 64176 (Rampe et al., 1993; Fan et al., 2000) were evident in all cells tested and included a dramatic slowing of Ca2+ channel activation, inactivation, and tail current kinetics. Whereas Bay K8644 and FPL 64176 did not reproduce some of their well established effects in the stem cell-derived cardiomyocytes, both compounds slowed Ca2+ channel inactivation and prolonged tail current decay. These actions should effectively increase the amount of Ca2+ entering the cells during a depolarizing step. Therefore both compounds may still act as Ca2+ channel activators when stem cell-derived cardiomyocytes are used in action potential recordings, microelectrode array recordings, contractility assays, or other assays that indirectly measure Ca2+ channel activity. It is only when the channel is studied directly and in isolation that the differences in pharmacology may become apparent.
It is unclear why the response to L-type Ca2+ channel activators differed between the stem cell-derived cardiomyocytes used in this study and what is known to occur in native myocytes. The binding sites for Bay K8644 and FPL 64176 are thought to be localized in the IIIS5-S6 linker of the α1 subunit of the channel (Yamaguchi et al., 2000, 2003), and single amino acid substitutions in this area dramatically reduce the ability of these molecules to enhance Ca2+ channel current amplitude. It is possible that some genomic alterations exist in the stem cells cardiomyocytes that limit their ability to respond to Ca2+ channel agonists. However, because both cell lines are of human origin, we would expect the sequence of the L-type Ca2+ channel to accurately reflect the normal genotype. Phenotypic differences between stem cell-derived cardiomyocytes and native myocytes have been described and may underlie the pharmacological profile we have observed. Many stem cell-derived myocytes, including the ones used in this article, are considered to have an immature or embryonic-like phenotype (Peng et al., 2010; Ma et al., 2011). Electrophysiologically this presents itself as a more depolarized resting membrane potential and a slower upstroke velocity during action potential recordings. It is possible that this immature phenotype in some way alters sensitivity to Ca2+ channel activators. If this is the case, such a phenomenon may be limited to certain stem cell-derived cardiomyocytes because other embryonic-like native cardiomyocytes, including those from embryonic chick heart (Anderson et al., 1990), neonatal rat heart (Rampe et al., 1993), and fetal human heart (Chen et al., 1999), all are known to retain sensitivity to Bay K8644 and/or FPL 64176. Calcium handling has also been reported to differ between stem cell-derived cardiomyocytes and native ones. In particular, the force of contraction in response to β-adrenergic agonists is reduced in stem cell-derived myocytes (Xi et al., 2010; Pillekamp et al., 2012). Contractile responses to β-adrenergic stimulation require functional L-type Ca2+ channels and sarcoplasmic reticulum, and it is believed that the latter is altered in stem cell-derived cardiomyocytes (Xi et al., 2010; Pillekamp et al., 2012). Based on our pharmacological results it is possible that functional alterations in the L-type Ca2+ channel are also present, at least to some degree, in stem cell-derived cardiomyocytes. Finally, the effects of Bay K8644 on Ca2+ channel activity are known to be reduced in myocytes taken from failing human hearts, presumably as a result of an increase in basal channel activity caused by a heightened phosphorylation status of the channel in these cells (Chen et al., 2008). Enhanced phosphorylation status could conceivably distort the response of the L-type Ca2+ channel in stem cell-derived cardiomyocytes to further stimulation by agonists such as Bay K8644. Further studies examining post-translational processing alterations (e.g., phosphorylation status) in the L-type Ca2+ channel of stem cell-derived cardiomyocytes will be useful and could possibly explain the pharmacological results obtained in this article.
In summary, we report here the first detailed pharmacological survey of the L-type Ca2+ channel in stem cell-derived cardiomyocytes and its comparison with a well studied native cardiomyocyte system. We find that the affinity for a wide range of structurally distinct antagonists is virtually identical in the stem cell-derived myocytes compared with the native cells. Conversely, Ca2+ channel activators, especially Bay K8644, failed to faithfully reproduce their established pharmacological effects in the stem cell-derived myocytes. The data point out the importance of detailed pharmacologic characterization of all ion channels in these and other stem cell-derived cell lines because these complex proteins contain multiple sites of drug action. With respect to L-type Ca2+ channels, Bay K8644 and FPL 64176 may represent a useful starting point when characterizing this channel's pharmacology in other stem cell-derived cell lines.
Participated in research design: Rampe.
Conducted experiments: Chen, Ji, and Lei.
Performed data analysis: Kang.
Wrote or contributed to the writing of the manuscript: Rampe.
Article, publication date, and citation information can be found at http://jpet.aspetjournals.org.
- human induced pluripotent stem cell-derived cardiomyocytes
- human embryonic stem cell-derived cardiomyocytes
- confidence limits
- MDL 12330
- 2-[(cis-2-phenylcyclopentyl)imino]-azacyclotridecane hydrochloride
- Bay K8644
- methyl 2,6-dimethyl-5-nitro-4-[2-(trifluoromethyl)phenyl]-1,4-dihydropyridine-3-carboxylate
- FPL 64176
- dimethyl-4-[2-(phenylmethyl)benzoyl]-1H-pyrrole-3-carboxylic acid methyl ester
- Received January 30, 2012.
- Accepted February 17, 2012.
- Copyright © 2012 by The American Society for Pharmacology and Experimental Therapeutics