Sage Journals: Discover world-class research

Abstract

Cardiomyocytes (CMs) derived from induced pluripotent stem cells (iPSCs) provide an in vitro model of the human myocardium. Complex 3D scaffolded culture methods improve the phenotypical maturity of iPSC-CMs, although typically at the expense of throughput. We have developed a novel, scalable approach that enables the use of iPSC-CM 3D spheroid models in a label-free readout system in a standard 96-well plate-based format. Spheroids were accurately positioned onto recording electrodes using a magnetic gold–iron oxide nanoparticle approach. Remarkably, both contractility (impedance) and extracellular field potentials (EFPs) could be detected from the actively beating spheroids over long durations and after automated dosing with pharmacological agents. The effects on these parameters of factors, such as co-culture (including human primary cardiac fibroblasts), extracellular buffer composition, and electrical pacing, were investigated. Beat amplitudes were increased greater than 15-fold by co-culture with fibroblasts. Optimization of extracellular Ca²⁺ fluxes and electrical pacing promoted the proper physiological response to positive inotropic agonists of increased beat amplitude (force) rather than the increased beat rate often observed in iPSC-CM studies. Mechanistically divergent repolarizations in different spheroid models were indicated by their responses to BaCl₂ compared with E-4031. These studies demonstrate a new method that enables the pharmacological responses of 3D iPSC-CM spheroids to be determined in a label-free, standardized, 96-well plate-based system. This approach could have discovery applications across cardiovascular efficacy and safety, where parameters typically sought as readouts of iPSC-CM maturity or physiological relevance have the potential to improve assay predictivity.

Keywords

stem cell cardiomyocyte co-culture contractility screening

Introduction

Adult myocardium is comprised at the cellular level predominantly of phenotypically mature cardiomyocytes (CMs) in close proximity to cardiac fibroblasts.¹ Induced pluripotent stem cell-derived CMs (iPSC-CMs) offer significant advancement toward modeling the adult myocardium. However, the phenotype is frequently more representative of an immature stage, with spontaneous pacemaker activity, lower force generation, and underdeveloped sarcoplasmic reticulum (SR) calcium (Ca²⁺) handling and sarcolemmal ion channel expression.²

Numerous studies have demonstrated that the phenotypical maturity of iPSC-CMs in vitro is limited postdifferentiation by culture in simple 2D monolayers and can be enhanced by mimicking a variety of physiological cues, such as 3D matrices and structured scaffolds,^3,4 co-culture with nonmyocyte supporting cells,^5–7 electrical pacing of contractile behavior and mechanical conditioning upon elastomeric substrates,^7–10 and utilization of appropriate energy sources.^11,12 In parallel, a wide variety of technical approaches have been developed in order to report functional performance. For example, iPSC-CMs together with nonmyocyte supporting cells or extracellular matrix proteins may be cast into strips or loops and contractility tracked optically.^{6,9–11,13–15} Alternatively, the self-assembly of 3D structures termed “microtissues,” “embryoid bodies,” or “spheroids” promoted by the use of low-adhesion surfaces or culture in hanging droplets has been combined with videomicrography to report contractility.^7,16–19 Dye-based or optogenetic reporters, and even highly manual approaches such as microelectrodes,¹⁶ multielectrode arrays,^20,21 or force transducers,¹⁰ have also been applied to 3D iPSC-CM constructs. One limitation of many of these approaches is the requirement for complex handling procedures and apparatus, resulting in low-throughput data generation, and few offer significant throughput.^11,22

It therefore seemed important to us to develop improved approaches addressing several of these known opportunities, which would permit efficient, scalable, high-throughput iPSC-CM studies. The simplest iPSC-CM 3D structures to obtain are spheroids/microtissues formed on low-adhesion surfaces or hydrogels.^7,16 One challenge is the subsequent difficulty in handling the resulting free-floating, actively beating spheroids and applying detection technologies in a robust manner. We sought to implement a high-throughput and scalable spheroid-based approach by circumventing some of these technical challenges by assembling iPSC-CM spheroids together with nonmyocyte supporting cells and gold–iron oxide nanoparticles. This rendered the resulting spheroids magnetic and able to be temporarily directed and located by arrays of neodymium magnets to facilitate accurate positioning of actively beating spheroids. We used this approach to locate actively beating co-culture spheroids within 96-well sensor plates developed for label-free functional measurements of 2D iPSC-CM monolayers²³ to determine whether contractility (impedance) and extracellular field potentials (EFPs) could be recorded from the nanoparticle-containing 3D spheroids. Remarkably, both parameters could be detected, providing the first plate-based measurements of this form from a 3D spheroid iPSC-CM model and suggesting that the approach could be scalable in combination with ordinary liquid handling automation. Indeed, pharmacological responses were produced efficiently using a combination of a 96-well liquid handling platform and acoustically dispensed compound plates. To investigate whether a 3D iPSC-CM model could recapitulate more physiological aspects of the adult myocardium, factors such as fibroblast co-culture (both human primary fibroblasts and a mouse fibroblast line), extracellular buffer composition, and electrical pacing were investigated. The utility of our new method is validated by three key studies where we demonstrate (1) strikingly enhanced contractile amplitude; (2) measurement of actual positive inotropic responses rather than beat rate effects, which are typically observed in 2D assays; and (3) responses to BaCl₂ and E-4031 as indicators of the repolarization mechanisms potentially in operation. These studies cover areas of interest to cardiovascular therapeutics as well as general in vitro safety profiling, suggesting that our new method could have applications across a range of discovery activities and provide a means to develop more physiologically relevant assays in these fields at a scale able to generate hundreds to thousands of data points.

Materials and Methods

Materials

Cells used in this study were iPSC-CMs (iCell², Cellular Dynamics International, Madison, WI) and primary human cardiac fibroblasts (PromoCell, Heidelberg, Germany) together with suppliers’ recommended media, and mouse embryonic fibroblast (MEF) NIH-3T3 cells (ATCC, Manassas, VA). Gold–iron oxide nanoparticle reagent (NanoShuttle-PL) was obtained from n3D Biosciences (Houston, Texas). Ninety-six-well ultra-low-attachment microplates (cat. 7007) were supplied by Corning Life Sciences (Corning, New York) and 96-well sensor plates (NSP-96 CardioExcyte96 sensor plates 0.6 mm, cat. 201002) were supplied by Nanion Technologies GmbH (München, Germany). Fibronectin from bovine plasma was obtained from Sigma-Aldrich (St. Louis, Missouri). Compounds were provided by the AstraZeneca compound collection or purchased from Sigma-Aldrich or Tocris Bioscience (Bristol, UK), and stock solutions were solubilized in DMSO.

Self-Assembly of iPSC-CM Co-Culture Spheroids

iPSC-CMs were resuspended from frozen vials according to supplier protocols to generate a suspension of 2000 cells/50 µL in CDI Plating media. Fibroblasts were routinely cultured according to supplier recommendations and resuspended at the required density (20–500 cells/50 µL in order to generate the ratios displayed in Fig. 3 ), also in CDI Plating media. Spheroids were seeded by combining iPSC-CMs and fibroblasts, together with 1 µL/6 × 10E4 cells NanoShuttle-PL gold–iron oxide nanoparticle reagent,²⁴ and then plating a volume of 100 µL/well in ultra-low-attachment 96-well plates. Cultures were maintained at 37 °C, 5% CO₂ in a humidified incubator. Media was changed to CDI Maintenance media (200 µL/well) after 24 h and 50% was replaced every 2–3 days subsequently. Unless otherwise stated, co-culture data refer to MEF 1:4 ratio CM spheroids.

Transfer of Co-Culture Spheroids to Sensor Plates

After assembly, nanoparticle-impregnated spheroids were transferred to sensor plates, using a custom-made neodymium magnet array located underneath the sensor plate to magnetically guide positioning of the spheroids onto the central sensing electrode located in the well base. Before transfer, a 50% media change was performed in assembly plates. In a sterile laminar-flow hood, a sensor plate (previously coated with 10 µg/mL fibronectin) was aligned with the custom-made neodymium magnet plate ( Suppl. Fig. S1 ) and the well contents of an assembly plate were transferred into the sensor plate using an eight-channel manual pipettor equipped with wide-bore tips. Care was taken to ensure that the pipetting action directed the well contents onto the central sensor electrode area and that the dispense step was as slow as possible to assist in the capture of the iPSC-CM/nanoparticle spheroids by the neodymium magnets ( Suppl. Fig. S1 ). In some cases, the spheroid had located itself very close to the recording electrode but not actually overlapping it. We anticipated that we could scan the plate via a stereomicroscope after the initial spheroid positioning step and observe and adjust the position of the spheroid onto the recording electrode using a sterile probe before the spheroids had fully adhered to the fibronectin coating. Magnetized spheroids were prevented from free-floating in the well by the arrayed magnets. After contact with the fibronectin-coated surface, the spheroid position was stable and the magnet array removed. Sensor plates were returned to the incubator.

Detection of Spheroid Contractility and EFP

Impedance and EFP signals were typically acquired 2 days after transfer into sensor plates. Before signal acquisition, a complete media change using 37 °C media ensured volumes of 250 µL/well in all wells. After a minimum 3 h equilibration period, sensor plates were loaded onto a CardioExcyte96 platform²³ (Nanion Technologies GmbH) for signal acquisition, typically with the platform located inside a tissue culture incubator. Unless otherwise stated, all measurements were performed using CDI Maintenance media as the incubation buffer. Signals were recorded in 20 s sweeps and typically low-pass-filtered at 10 and 400 Hz (impedance and EFP, respectively) before analysis.

Whole-Plate Compound Dosing Using Automated Liquid Handling

Experiments were designed where all wells in a sensor plate would be dosed with compounds simultaneously, either as one-off additions or by multiple whole-plate additions. A compound source plate was created in a 384-well V-bottomed microplate using a 384-to-96 quadrant mapping. Compounds were dispensed into the required quadrants of the compound source plate using acoustic dispensing and Echo Cherry Pick software (Echo 555 platform, Labcyte, USA), creating a 250 nL/well dispense of 100% DMSO stock solution (1000× final concentration). A liquid handling platform (CyBio FeliX, Analytik Jena AG) equipped with a 96-pipettor head was used to perform whole-plate dosing. Briefly, the sensor plate was removed from the CardioExcyte96 platform and positioned on the stage of the FeliX. Fifty microliters per well of warm media was aspirated from the sensor plate, dispensed into the appropriate quadrant of the 384-well compound source plate, aspirated, and dispensed (off-center) back into the sensor plate. Compound dosing was achieved in less than 20 s, and the sensor plate returned to the CardioExcyte96 platform.

Electrical Pacing and Modified Tyrodes Conditions

Electrical field stimulation via the CardioExcyte96 platform (typically 1%–5% power) was adjusted to capture spheroid beating while preventing pacing artifacts. Experiments were performed at 1 Hz pacing frequency unless otherwise stated. For reduced external Ca²⁺ experiments, warmed Tyrodes solution (145 mM NaCl, 4 mM KCl, 1 mM MgCl₂, 11.1 mM glucose, 10 mM HEPES, pH 7.4) containing the required amount of CaCl₂ was substituted in place of the culture media by a 50% substitution, followed by 100% substitution, over the course of 30 min, with a subsequent 1 h equilibration period before further additions.

Optical Flow Analysis of Contractility

Video clips (20 s) were captured via bright-field illumination on an inverted microscope equipped with a 10× objective directly from the clear plastic assembly plates. Video files were converted to 8-bit grayscale Z stacks in ImageJ and cropped to 250 pixels per side (sufficient to encompass the complete spheroid). Optical flow was calculated using particle image velocimetry (ImageJ), determining the vector magnitudes per Z-stack image. Z-axis histograms were generated to display average vector magnitude against time.

Histopathology Processing

Spheroids were processed as formalin-fixed, paraffin-embedded blocks and sections stained with H&E as previously described.²⁵

Data Analysis

Impedance and EFP recordings from a single sensor plate experiment were processed using Nanion CardioExcyteControl software. Beat detection and determination of average parameters²³ in impedance mode (amplitude, beat rate, pulse width 50%, rising time, and falling time) or EFP mode (FPDmax) were calculated for each sweep. Manual quality control was performed to exclude individual wells with unstable beat detection or unstable electrical pacing. Further analysis was performed in RStudio and Microsoft Excel. Data are presented as mean ± SD, and statistical significance was calculated using the Student t test or Kruskal–Wallis test with Dunn’s post hoc test as indicated. A p value <0.05 considered significant.

Data Availability

The datasets generated and analyzed during the current study are available from the corresponding author upon reasonable request.

Results

Self-Assembly of Co-Culture Spheroid Clusters and Production of Plate-Based Arrays

Diffuse aggregates of iPSC-CMs together with primary human cardiac fibroblasts or MEFs visible after seeding in assembly plates ( Fig. 1a ) spontaneously condensed into spheroid clusters over the course of 4 days (iPSC-CMs only) or 1–2 days (fibroblast co-cultures) ( Fig. 1b ), with synchronized contractility emerging during this phase. Spheroid clusters composed of 2000 iPSC-CMs together with fibroblasts showed no evidence of tissue necrosis in the central core region ( Fig. 1d ). Visual inspection of assembly plates demonstrated that spontaneous contractile activity emerged in all co-culture spheroids.

Figure 1.

(a) Self-assembly of co-culture spheroid clusters and production of plate-based arrays. Schematic of workflow, including assembly of co-cultures with iron–gold nanoparticles in ultra-low-attachment assembly plates and transfer to sensor plates using a custom-built magnet array aligned to electrode positions to deposit spheroids onto recording electrodes. (b) Self-assembly of iPSC-CM spheroids over 4 days (days 1–4) postseeding into assembly plates, in iPSC-CM-only or fibroblast (fb) co-culture formats. (c) Neodymium magnet array with dimensions identical to 12 × 8 electrode array position in sensor plates. (d) H&E-stained paraffin section showing incorporation of gold–iron oxide nanoparticles into spheroids (white arrows) and the absence of any central necrosis. (e) Neutral Red-stained spheroid after 4 days adhered to a coated sensor plate electrode, showing the maintained 3D structure and lack of 2D monolayer appearance. Scale bars = 100 µm.

Incorporation of gold–iron oxide nanoparticles into the iPSC-CM spheroids during assembly ( Fig. 1d ) enabled positional manipulation via magnetic fields. A custom-made neodymium magnet array ( Fig. 1c ) that precisely corresponded to the position and spacing of the 12 × 8 arrays of circular 0.6 mm gold recording electrodes in the base of 96-well sensor plates was used to generate plate-based spheroid arrays, with spheroids located on top of the central sensing electrode ( Fig. 1e ). This approach led to successful detection of contractility in approximately 60% of wells by impedance sensing, which increased to >95% with small manual adjustments of position, while spheroids were prevented from free-floating in the well by the magnet array. Importantly, the 3D shape of spheroids was maintained after 4 days of adherence to fibronectin-coated sensor plate electrodes, with no evidence that spheroids had assumed a 2D monolayer-type structure ( Fig. 1e ).

Proof-of-Concept Detection of Contractility Parameters by Impedance Sensing in iPSC-CM 3D Spheroids

Spontaneous contractile activity of iPSC-CM spheroids could be detected by impedance sensing ( Fig. 2a ). Beat amplitudes increased after the transfer of spheroids into sensing plates ( Suppl. Fig. S2 ), increasing beat detection success rates. After 2 days in sensor plates, during which time beat rates were stable ( Suppl. Fig. S2 ), iPSC-CM spheroids displayed typical beat rates of 0.8 ± 0.04 Hz or 49 ± 2.4 bpm and amplitudes <5 ohm. Amplitudes were variable due to influences of electrode alignment accuracy, although all parameters, once established, were stable for at least 12 h of continuous recording ( Fig. 2b ). Comparison of iPSC-CM spheroids generated with and without the incorporation of gold–iron oxide nanoparticles indicated that beat characteristics were unaltered by the incorporation of nanoparticles, except rising time, which was slightly increased in the absence of nanoparticles ( Table 1 ) and may reflect the fact that CM-only spheroids formed in a less quick and compact manner ( Fig. 1 ).

Table 1.

Comparison of Beat Characteristics in Impedance Recording of iPSC-CM Spheroids Assembled with or without Gold–Iron Oxide Nanoparticles and Fibroblasts.

	Amplitude (Ohm)	Rate (min^–1)	Beat Width (s)	Rising Time (s)	Falling Time (s)
CM	1.2 ± 0.8	48.5 ± 2.3	0.31 ± 0.07	0.090 ± 0.015	0.264 ± 0.101
CM_Nano	2.2 ± 0.9	48.7 ± 2.4	0.32 ± 0.04	0.080 ± 0.011*	0.262 ± 0.098
CM_Fb_Nano	45.0 ± 18.1*†	34.9 ± 3.5*†	0.34 ± 0.05†	0.077 ± 0.018*	0.202 ± 0.056

Spheroids composed of iPSC-CMs (CM) were assembled with the incorporation of nanoparticles (CM_Nano) and MEFs (CM_Fb_Nano), and beat characteristics were quantified in impedance mode. Data are presented as mean ± SD. *p < 0.05 and †p < 0.05 compared with CM and CM_Nano, respectively (Kruskal–Wallis test).

Figure 2.

Detection of contractility parameters by impedance sensing in iPSC-CM spheroids. (a) Representative 10 s impedance trace showing downward deflections during individual contractions and regular beat activity. (b) Beat characteristics were stable over extended time periods. (c) Traces showing effects of 400 nM isoproterenol (blue) or 1 µM verapamil (black) compared with control (gray). (d) Mean data for panel c (n = 5). (e) Concentration–response data were generated using automated liquid handling for the cumulative addition of verapamil (n = 9 replicates).

An efficient approach to compound addition was developed based on whole-plate experiments using an automated 96-well liquid handling platform and compound source plates prepared by acoustic dispensing of DMSO stock solutions. Warm media was aspirated from the recording plate, rapidly mixed with the 250 nL DMSO stock solution volumes within the source plate, and then dispensed back into the recording plate. This approach avoided the preparation of additional warmed compound dilutions for dosing, minimized experimental times, and maintained final DMSO concentrations at 0.1% (or lower if required) to prevent vehicle artifacts. Rapid compound dosing (less than 20 s for automated media transfer), coupled with large well volumes (250 µL), maintained sensor plate temperature, which was stable to within 0.5 °C during subsequent recordings 3 min after compound addition. Effects consistent with known pharmacology were observed following the addition of 1 µM verapamil (L-type Ca²⁺ channel inhibitor) or 400 nM isoproterenol (β-adrenoceptor agonist), resulting in cessation of beating activity or an increase in beat rate, respectively ( Fig. 2c,d ). Concentration–response data could also be generated using a cumulative-dosing strategy, for example, a 6-point, 0.5-log unit dilution series of verapamil ( Fig. 2e ).

Fibroblast Co-Culture Dramatically Modulates Intrinsic Contractility

Self-assembly of iPSC-CM spheroids together with fibroblasts of primary human cardiac or MEF origin resulted in co-culture spheroids with significantly altered functional characteristics. Most apparent was a dramatic increase in contractile amplitude, with fibroblast co-culture resulting in an approximately 15-fold (mouse) or 6-fold (primary human cardiac) increase in beat amplitude (representative traces shown in Fig. 3a ). These differences were also clear upon visual inspection of spheroids in assembly plates (Suppl. Fig. S3 Movies), and optical flow analysis provided endpoint-independent confirmation. With this approach, vector magnitudes representing beat deflections were increased more than 10-fold by MEF co-culture ( Fig. 3b ).

Figure 3.

Fibroblast co-culture dramatically modulates intrinsic contractile behavior. (a) Representative impedance recordings showing beat amplitude of iPSC-CM-only spheroids (CM only) compared with co-culture spheroids containing either MEFs (CM-Fb) or primary human cardiac fibroblasts (CM-hCFb). (b) Optical detection of beat characteristics in assembly plates (optical flow analysis of videomicrographs) also demonstrated a dramatic influence of fibroblast co-culture on beat amplitude (i, iPSC-CM only; ii, MEF co-culture). Effect of increasing proportions of fibroblasts in 2000 cell iPSC-CM spheroids, with assembly of spheroids using human (i or open square) or MEF (ii or closed square) fibroblasts, on impedance sensing beat (c) amplitude and (d) rate. Mean ± SD and n values are shown; *p < 0.05 compared with iPSC-CMs (Kruskal–Wallis test).

Primary human cardiac fibroblasts required an initial seeding ratio of at least 1:4 to significantly increase beat amplitude, although MEF seeding ratios as high as 1:100 induced a dramatic effect on analysis on day 10 (Fig. 3c). It is important to note that the variability in the amplitude parameter apparent in Figure 3c is only present in the absolute parameter and not in the normalized parameter (e.g., Fig. 4b–d ), and likely represents the fact that the position of the spheroids is not identical on all recording electrodes. Beat rates were also influenced by fibroblast co-culture. Increasing proportions of primary human cardiac fibroblasts increased the rate, while the rate decreased with MEFs (Fig. 3d, Table 1). However, robust contractions were apparent for a variety of seeding ratios over a variety of time points, suggesting that contractility was more sensitive to the presence of fibroblasts per se rather than the precise fibroblast ratio.

Figure 4.

Inotropic responses are modulated by extracellular Ca²⁺ concentrations and electrical pacing. (a) Isoproterenol (100 nM) increased the beat amplitude (inotropic response) under conditions of 0.6 mM Ca²⁺ Tyrodes solution and 1 Hz electrical pacing (before [black line] and after [dashed line] isoproterenol). The normalized beat trace showed a reduced beat width under paced conditions. (b) Extracellular Ca²⁺ levels less than 1.2 mM were required for inotropic responses to isoproterenol stimulation under paced conditions. (c) In 0.9 mM Ca²⁺ Tyrodes solution, pacing for 20 s without isoproterenol stimulation elicited an increase in beat amplitude between the first (solid line) and last (dashed line) beats, and this effect was absent in control media or 1.5 mM Ca²⁺ Tyrodes solution. (d) The positive inotrope digoxin (1 µM), but not the phosphodiesterase inhibitor (and chronotropic agent presumably via HCN pacemaker current) sildenafil (10 µM), evoked an increase in paced amplitude similar to isoproterenol. Mean ± SD and n values are shown; p < 0.05 versus control or as indicated (Kruskal–Wallis test).

Inotropic Responses Revealed by Electrical Pacing with Reduced Sarcolemmal Ca²⁺ Influx

Positive inotropes such as isoproterenol typically evoke chronotropic rather than inotropic responses in iPSC-CMs (e.g., Fig. 2a ), which may reflect the relative underdevelopment of SR compared with sarcolemmal Ca²⁺ handling.² Reduction of sarcolemmal Ca²⁺ influx, by substituting modified (decreased Ca²⁺) Tyrodes solution, was combined with electrical pacing to control rate. Isoproterenol stimulation (100 nM) of co-culture spheroids under these conditions evoked approximately twofold increases in beat amplitude, demonstrating a bona fide increase in force generation in response to β-adrenergic stimulation ( Fig. 4a ). Beat width was reduced concomitantly with the isoproterenol-evoked increase in paced amplitude (Fig. 4a, Suppl. Table S1). Titration of extracellular Ca²⁺ concentrations from 1.5 mM indicated that this effect was dependent upon concentrations of 0.9 mM or less, and was more apparent at lower concentrations of 0.6 mM ( Fig. 4b ). Long-term effects of pacing were not examined in this study; however, pacing for 20 s without isoproterenol stimulation was sufficient to evoke a 1.24- ± 0.8-fold increase in beat amplitude between the first and last beats under 0.9 mM Ca²⁺ conditions. This effect was absent in control media or 1.5 mM Tyrodes conditions ( Fig. 4c ), confirming that the mechanism is related to reduction in sarcolemmal Ca²⁺ influx. Although the stress–strain relationship of co-culture spheroids has not been quantified, force generation (stress) within a given spheroid is likely represented by relative changes in beat amplitude (strain), given that the spheroid elastic modulus (stiffness) remains constant during acute experiments.

Similar to isoproterenol, neither digoxin (Na⁺/K⁺ ATPase inhibitor) nor sildenafil (phosphodiesterase type 5 inhibitor) elicited increased beat amplitude responses under basal conditions (data not shown). However, under paced/reduced Ca²⁺ conditions, digoxin evoked a significant inotropic response (1.9- ± 0.5-fold increase in beat amplitude), while sildenafil was without a positive inotropic effect (0.8- ± 0.1-fold). Hence, these conditions confer a degree of mechanistic translation to the native adult CM, given that digoxin is a positive inotrope in adult ventricular myocytes by a non-β-adrenergic mechanism,²⁶ and that sildenafil is only chronotropic via HCN (hyperpolarization-activated and cyclic nucleotide-gated channels) pacemaker channel activation in isolated myocytes.²⁷

Detection of EFP in Actively Beating 3D Spheroids

EFP could be recorded from actively beating co-culture spheroids ( Fig. 5a ). Pharmacological modulation of the iPSC-CM action potential using verapamil (100 nM) and the L-type channel activator Bay K-8644 (1 µM) caused marked EFP shortening and lengthening, respectively ( Fig. 5b,c ), consistent with their known effects. Interestingly, delayed after-depolarizations (DADs) were clearly apparent following Bay K-8644 ( Fig. 5c ) and represent overload of SR Ca²⁺ content, leading to spontaneous release events and transient activation of inward currents.²⁸ The detection of EFP signals from actively beating spheroids was less robust than that of impedance signals (approximately 50% of the impedance success rate). The impact of active contractility and spheroid movement during EFP recording is not known; however, pharmacological modulation of FPD (field potential duration) parameters was clearly established. No overt effect on EFP was apparent upon inclusion of gold–iron oxide nanoparticles. Under basal conditions, EFP and impedance beat detection were quantified in 1037 co-culture spheroids between days 7 and 20 postseeding from a total of four independent batches across 12 sensor plates ( Fig. 5d ), a study requiring only several weeks to complete. There was a trend apparent toward reduced beat rates with culture time and also increased FPD data scatter after 12 days. The mean variability (%CV) values across all plates were 8% and 10% for beat rate and FPD, respectively, and variability in one parameter was not associated with variability in the other ( Suppl. Fig. S4 ).

Figure 5.

Detection of EFP in actively beating iPSC-CM co-culture spheroids. Representative traces showing EFP signals detected in (a) control, (b) 100 nM verapamil-treated, and (c) 1 µM Bay-K8644-treated co-culture spheroids. FPD (b) shortening and (c) lengthening were apparent, together with (c) DADs in Bay-K8644-treated spheroids. Data from 1037 fibroblast co-culture spheroids, between days 7 and 20 postseeding in 12 separate recording plates, detected in both EFP and impedance modes, demonstrating baseline FPD and beat rate apparent during that period. Mean ± SD, n, and %CV values of these parameters are presented in Supplementary Figure S4 .

Repolarization Mechanisms Differ in iPSC-CM and Co-Culture Spheroids

Co-culture using MEFs significantly affected beat rate and FPD, with reduced beat rate and marginally increased FPD compared with iPSC-CM-only spheroids (BR: 41.3 ± 4.7 bpm vs 52.9 ± 6.4 bpm; FPD: 0.53 ± 0.06 s vs 0.50 ± 0.07 s). These differences were further examined using short (3–10 min) applications of either vehicle, 30 nM E-4031 (hERG channel inhibitor), or 100 µM BaCl₂ (inhibitor of several K⁺ channels, including Kir2.1). Responses to E-4031 were time dependent and arrhythmogenic ( Suppl. Fig. S5 ). Prior to arrhythmia, acute E-4031 effects (3 min) were similar in both iPSC-CM and co-culture spheroids, with an increase in FPD ( Fig. 6 ) in line with other studies.²⁰ In iPSC-CM spheroids, the linear FPD versus beat rate relationship was maintained, suggesting that a deviation from rate-corrected FPD coincided with onset of arrhythmic activity. In co-culture spheroids, FPD clearly deviated from the baseline relationship before loss of rhythm. This was in marked contrast to BaCl₂ (3 min) responses, which diverged markedly in MEF co-culture spheroids by increasing both BR and FPD dimensions ( Fig. 6 ). BaCl₂ in co-culture spheroids was the only condition to increase beat rates. Furthermore, these effects of BaCl₂ were also apparent in primary human cardiac fibroblast co-culture spheroids ( Suppl. Fig. S6 ).

Figure 6.

Repolarization mechanisms in iPSC-CM-only and co-culture spheroids. FPD versus beat rate relationships of iPSC-CM-only spheroids (iPSC-CM, closed symbols) and MEF co-culture spheroids (CM+Fb, open symbols) following dosing with DMSO vehicle (blue symbols), 30 nM E-4031 (green symbols), or 100 µM BaCl₂ (red symbols) to examine repolarization mechanisms. (a) Mean data are shown, together with scatterplots for (b) E-4031 and (c) BaCl₂ responses. (a,c) A marked increase in beat rate following BaCl₂ treatment was apparent in co-culture spheroids only, suggesting ion channel expression may differ between spheroid types. Mean ± SD values are shown; *p < 0.05 and #p < 0.05 (Kruskal–Wallis test) versus DMSO for beat rate and FPD, respectively.

Discussion

In this study, we have profiled more than 1000 iPSC-CM co-culture spheroids in a simple, scalable approach that returned label-free measures of contractility and EFP. An efficient process for dosing with pharmacological agents was demonstrated using a combination of an acoustically dispensed compound source plate and an automated liquid handling platform. Furthermore, these methods are amenable to profiling the 3D co-culture phenotype of iPSC-CMs and reveal that iPSC-CM phenotypes under these conditions can share multiple maturity-associated features usually found in more complex and low-throughput approaches. To validate our approach in relevant areas of CM biology, we demonstrate that spheroid co-culture of iPSC-CMs with fibroblasts of diverse origin increased beat amplitudes, by up to an order of magnitude in some conditions. Inotropic responses usually absent from iPSC-CMs could be revealed by reducing sarcolemmal Ca²⁺ flux and electrically pacing the spheroids to control beat rate, and the same effect was seen with inotropes of diverse mechanisms. Finally, action potential repolarization apparently involved divergent ion channel ensembles in iPSC-CMs compared with co-culture spheroids, including a BaCl₂-sensitive component distinct from the E-4031 response. These results clearly demonstrate that the iPSC-CMs respond phenotypically to the 3D co-culture spheroid environment, and the system provides the throughput to profile further optimizations, such as cell type, long-term optogenetic or electrical pacing, and extracellular matrices. This novel and simple approach recapitulates multiple maturation-associated features of iPSC-CM physiology usually attributed to more technically complex and low-throughput approaches and facilitates further development of spheroid co-culture models for applications as diverse as disease modeling, iPSC differentiation, the cardiovascular therapy area, or drug safety.

Numerous studies seek to enhance the contractility of iPSC-CM constructs into the range of native human myocardium. Engineered tissue approaches frequently incorporate nonmyocyte cells derived from the progenitor cell differentiation protocol as support cells that aid the formation of the tissue constructs, enhance the ventricular phenotype, and modulate contractility both positively and negatively according to their relative proportions.^5,7,14,15 Inclusion of additional fibroblasts into these constructs can be associated with increased action potential duration, reduced propagation (conduction velocity), and altered ion channel repertoires.^3,6,16,29 In the current study, contractility was profoundly enhanced by primary human fibroblast co-culture (Fig. 3, Suppl. Fig. S3 Movies), which may reflect the relative iPSC-CM purity and absence of nonmyocyte support cells in the starting material. However, the much greater degree of contractility apparent in MEF co-cultures suggests the specific potentiation of pathways linked to the formation and activation of the contractility machinery. For instance, fibroblasts secrete extracellular matrix proteins shown to be key regulators of CM differentiation in vitro,³⁰ and may also stimulate CM maturation via increased mechanical stiffness of the co-culture construct^4,8 or engagement of integrin receptor proteins that form part of the protein scaffold integrating force transduction.³¹ MEFs excel in progenitor cell support applications^32–34 and highly express the CM maturation marker connexin 43^35,36 required for the synchronization and propagation of the cardiac action potential and also propagation of microRNA-directed CM signaling.³⁷ In fact, a very recent study specifically links intercellular communication via connexin 43 expression in co-culture fibroblasts to the maturation of the iPSC-CM phenotype.³⁸ The current study demonstrates an approach that can be applied to iPSC-CM spheroids of any composition, and our data indicate that the functional properties may be altered significantly in divergent models. Hence, inclusion of human cardiac fibroblasts or nonhuman fibroblasts (i.e., MEFs), or indeed other cell types not investigated in the current study, may provide the most appropriate functional phenotypes for further profiling using these methods.

A key aspect of CM contractility is the precise organization and function of SR Ca²⁺ handling apparatus. The (lack of) inotropic responses of iPSC-CMs reflect the relative immaturity of the intracellular SR/T-tubule system and a dominant effect of sarcolemmal ion fluxes.² However, a degree of functional integration between sarcolemmal and SR Ca²⁺ pathways was confirmed in co-culture spheroids by the appearance of DADs following Bay K-8644 ( Fig. 5c ), which mechanistically reflect an SR overloaded with sarcolemmal Ca²⁺.²⁸ Furthermore, reduction of sarcolemmal Ca²⁺ flux under paced conditions revealed β-adrenergic-evoked increases in force generation in co-culture spheroids (Fig. 4a, Suppl. Table S1), an effect previously noted in iPSC-CM cast hydrogel constructs.^5,14 Also revealed under reduced Ca²⁺ influx was an acute effect where 20 s of paced beating caused a 24% increase in force generation ( Fig. 4c ), similar to effects noted in longer-term studies.^7,9,10 A component of functional SR underlies these effects, as pacing under normal conditions (1.8 mM Ca²⁺ media or 1.5 mM Tyrodes solution) confirmed an absence of myofilament Ca²⁺ sensitization or sarcolemmal Ca²⁺ flux effects. The action of inotropes in paced force generation confirms further mechanistic aspects (i.e., digoxin that relies on the expression of sodium–calcium exchanger activity for inotropic effect²⁶) and indicates a repertoire of iPSC-CM maturity-associated functional Ca²⁺ handling by the nascent SR that is usually masked by dominant sarcolemmal fluxes.

The ensemble of expressed ion channels mediating action potential characteristics is a significant aspect of the mature iPSC-CM phenotype. Repolarization of cardiac action potential involves a number of ionic fluxes, notably I_Kr (hERG channel) and the contribution of I_K1 (Kir2.1 channel) to resting membrane potential. Comparing the effects 30 nM E-4031 (selective hERG blocker) with 100 µM BaCl₂ (which also blocks hERG channels at high concentration but is relatively selective for I_K1 at concentrations of 50–100 µM³⁹) demonstrated unique effects of BaCl₂ in the fibroblast co-culture spheroids, specifically an increase in the spontaneous beat rate consistent with depolarization of the resting membrane potential. This effect was apparent in both primary human cardiac fibroblast co-cultures ( Suppl. Fig. S6 ) and MEF co-cultures ( Fig. 6 ), and widely divergent from selective hERG blockade with E-4031, which reduced beat rate and increased FPD. It should be noted, however, that the basal beat rates in co-culture spheroids were not consistent with a reduction in basal membrane potential ( Fig. 3 ), and molecular characterization to confirm the presence of Kir2.1 would complement functional studies achieved using the current approach. Reduced Kir2.1 expression and function are frequently cited as a significant deficiency in the use of iPSC-CMs in translational studies;^39–42 hence, the present results using the I_K1 blocker BaCl₂ are intriguing. Herron et al.⁴³ studied 2D elastomeric iPSC-CM constructs in combination with extracellular matrix (Matrigel) coatings and demonstrated increased expression of I_K1 current and Kir2.1 protein, via an integrin β1 signaling mechanism not activated by a combination of fibronectin on glass. Fibroblast co-culture spheroids, which deform during each beat (Suppl. Fig. S3 Movies) and likely contain fibroblast matrix secretions, may replicate these elastomeric and extracellular matrix protein features. Nunes et al.⁶ also demonstrated increased I_K1 current and more negative resting membrane potentials in fibroblast-containing tissue constructs compared with iPSC-CM-only embryoid bodies.

Use of iPSC-CMs in a spheroid format of approximately 2000 cells is an efficient approach compared with engineered strips or loops requiring 1 × 10E6 cells per construct, with additional Matrigel and collagen gels.^6,14 Spheroids formed by self-assembly of iPSC-CMs, fibroblasts, and gold–iron oxide nanoparticles are mechanically robust and reliably initiate spontaneous beating within 2 days of seeding, which is not always the case for other HT approaches, such as 384-well 2D monolayers. Using automated liquid handling platforms and acoustically dispensed compound source plates facilitated robust experimental design and maintained low DMSO concentrations, a process amenable to generating IC₅₀ data via cumulative dosing. Gold–iron oxide nanoparticles did not interfere with sensor readout or influence beat characteristics. Understanding how the field potential of a beating (moving) spheroid relates to effects of intercellular conduction and individual ionic conductances remains to be determined, although empirical data in this study and others,^20,21 including simulations,⁴⁴ confirm the validity. Stable EFP measurements were detected at approximately half the rate of impedance measurements, which may relate to these factors. Further developments of the current approach could involve use of automated liquid handling platforms to perform the transfer of assembled spheroids into sensor plates, which may improve the reproducibility of spheroid EFP measurements. Additionally, long-term pacing of the spheroids to further enhance maturity,^7,9,10 perhaps via an optogenetic approach implemented in fibroblasts rather than iPSC-CMs,⁴⁵ could be explored.

Overall, these novel methods offer a high-throughput, scalable approach to understand the iPSC-CM 3D spheroid functional phenotype and develop diverse applications in tissue engineering, drug discovery, and disease modeling.

Supplemental Material

sj-pdf-1-jbx-10.1177_2472555220975332 – Supplemental material for A Scalable Approach Reveals Functional Responses of iPSC Cardiomyocyte 3D Spheroids

Supplemental material, sj-pdf-1-jbx-10.1177_2472555220975332 for A Scalable Approach Reveals Functional Responses of iPSC Cardiomyocyte 3D Spheroids by Matthew P. Burnham, Rachel Harvey, Rebecca Sargeant, Niels Fertig and Malcolm Haddrick in SLAS Discovery

Footnotes

Acknowledgements

We thank Sonja Stölzle-Feix, Leo Dörr, and Ulrich Thomas for their assistance with the CardioExcyte96 platform.

Declaration of Conflicting Interests

The authors declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: M.P.B., R.S. and M.H. are/were employees of AstraZeneca and potentially own stock and/or stock options in the company. N.F. is founder and CEO of Nanion Technologies GmbH.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Matthew P. Burnham

Supplemental material is available online with this article.

References

Camelliti

Borg

T. K.

Kohl

Structural and Functional Characterisation of Cardiac Fibroblasts. Cardiovasc. Res. 2005, 65, 40–51.

Yang

Pabon

Murry

C. E.

Engineering Adolescence: Maturation of Human Pluripotent Stem Cell-Derived Cardiomyocytes. Circ. Res. 2014, 114, 511–523.

Liau

Christoforou

Leong

K. W.

, et al Pluripotent Stem Cell-Derived Cardiac Tissue Patch with Advanced Structure and Function. Biomaterials 2011, 32, 9180–9187.

Ribeiro

A. J.

Ang

Y. S.

J. D.

, et al Contractility of Single Cardiomyocytes Differentiated from Pluripotent Stem Cells Depends on Physiological Shape and Substrate Stiffness. Proc. Natl. Acad. Sci. U.S.A. 2015, 112, 12705–12710.

Zhang

Shadrin

I. Y.

Lam

, et al Tissue-Engineered Cardiac Patch for Advanced Functional Maturation of Human ESC-Derived Cardiomyocytes. Biomaterials 2013, 34, 5813–5820.

Nunes

S. S.

Miklas

J. W.

Liu

, et al Biowire: A Platform for Maturation of Human Pluripotent Stem Cell-Derived Cardiomyocytes. Nat. Methods 2013, 10, 781–787.

Eng

Lee

B. W.

Protas

, et al Autonomous Beating Rate Adaptation in Human Stem Cell-Derived Cardiomyocytes. Nat. Commun. 2016, 7, 10312.

Hazeltine

L. B.

Simmons

C. S.

Salick

M. R.

, et al Effects of Substrate Mechanics on Contractility of Cardiomyocytes Generated from Human Pluripotent Stem Cells. Int. J. Cell Biol. 2012, 2012, 508294.

Hirt

M. N.

Boeddinghaus

Mitchell

, et al Functional Improvement and Maturation of Rat and Human Engineered Heart Tissue by Chronic Electrical Stimulation. J. Mol. Cell. Cardiol. 2014, 74, 151–161.

10.

Ruan

J. L.

Tulloch

N. L.

Razumova

M. V.

, et al Mechanical Stress Conditioning and Electrical Stimulation Promote Contractility and Force Maturation of Induced Pluripotent Stem Cell-Derived Human Cardiac Tissue. Circulation 2016, 134, 1557–1567.

11.

Mills

R. J.

Titmarsh

D. M.

Koenig

, et al Functional Screening in Human Cardiac Organoids Reveals a Metabolic Mechanism for Cardiomyocyte Cell Cycle Arrest. Proc. Natl. Acad. Sci. U.S.A. 2017, 114, E8372–E8381.

12.

Ulmer

B. M.

Stoehr

Schulze

M. L.

, et al Contractile Work Contributes to Maturation of Energy Metabolism in hiPSC-Derived Cardiomyocytes. Stem Cell Rep. 2018, 10, 834–847.

13.

Sidorov

V. Y.

Samson

P. C.

Sidorova

T. N.

, et al I-Wire Heart-on-a-Chip I: Three-Dimensional Cardiac Tissue Constructs for Physiology and Pharmacology. Acta Biomater. 2017, 48, 68–78.

14.

Mannhardt

Breckwoldt

Letuffe-Brenière

, et al Human Engineered Heart Tissue: Analysis of Contractile Force. Stem Cell Rep. 2016, 7, 29–42.

15.

Tiburcy

Hudson

J. E.

Balfanz

, et al Defined Engineered Human Myocardium with Advanced Maturation for Applications in Heart Failure Modeling and Repair. Circulation 2017, 135, 1832–1847.

16.

Desroches

B. R.

Zhang

Choi

B. R.

, et al Functional Scaffold-Free 3-D Cardiac Microtissues: A Novel Model for the Investigation of Heart Cells. Am. J. Physiol. Heart Circ. Physiol. 2012, 302, H2031–H2042.

17.

Beauchamp

Moritz

Kelm

J. M.

, et al Development and Characterization of a Scaffold-Free 3D Spheroid Model of Induced Pluripotent Stem Cell-Derived Human Cardiomyocytes. Tissue Eng. Part C Methods 2015, 21, 852–861.

18.

Pointon

Pilling

Dorval

, et al High-Throughput Imaging of Cardiac Microtissues for the Assessment of Cardiac Contraction during Drug Discovery. Toxicol. Sci. 2017, 155, 444–457.

19.

Polonchuk

Chabria

Badi

, et al Cardiac Spheroids as Promising In Vitro Models to Study the Human Heart Microenvironment. Sci. Rep. 2017, 7, 7005.

20.

Jahnke

H. G.

Steel

Fleischer

, et al A Novel 3D Label-Free Monitoring System of hES-Derived Cardiomyocyte Clusters: A Step Forward to In Vitro Cardiotoxicity Testing. PLoS One 2013, 8, e68971.

21.

Rast

Weber

Disch

, et al An Integrated Platform for Simultaneous Multi-Well Field Potential Recording and Fura-2-Based Calcium Transient Ratiometry in Human Induced Pluripotent Stem Cell (hiPSC)-Derived Cardiomyocytes. J. Pharmacol. Toxicol. Methods 2015, 75, 91–100.

22.

Klimas

Ambrosi

C. M.

, et al OptoDyCE as an Automated System for High-Throughput All-Optical Dynamic Cardiac Electrophysiology. Nat. Commun. 2016, 7, 11542.

23.

Doerr

Thomas

Guinot

D. R.

, et al New Easy-to-Use Hybrid System for Extracellular Potential and Impedance Recordings. J. Lab. Autom. 2015, 20, 175–188.

24.

Souza

G. R.

Molina

J. R.

Raphael

R. M.

, et al Three-Dimensional Tissue Culture Based on Magnetic Cell Levitation. Nat. Nanotechnol. 2010, 5, 291–296.

25.

Archer

C. R.

Sargeant

Basak

, et al Characterization and Validation of a Human 3D Cardiac Microtissue for the Assessment of Changes in Cardiac Pathology. Sci Rep. 2018, 8, 10160.

26.

Smith

T. W.

The Basic Mechanism of Inotropic Action of Digitalis Glycosides. J. Pharmacol. 1984, 15 (Suppl. 1), 35–51.

27.

Isidori

A. M.

Cornacchione

Barbagallo

, et al Inhibition of Type 5 Phosphodiesterase Counteracts β2-Adrenergic Signalling in Beating Cardiomyocytes. Cardiovasc. Res. 2015, 106, 408–420.

28.

Volders

P. G.

Vos

M. A.

Szabo

, et al Progress in the Understanding of Cardiac Early Afterdepolarizations and Torsades de Pointes: Time to Revise Current Concepts. Cardiovasc. Res. 2000, 46, 376–392.

29.

Miragoli

Gaudesius

Rohr

Electrotonic Modulation of Cardiac Impulse Conduction by Myofibroblasts. Circ. Res. 2006, 98, 801–810.

30.

Wong

McCloskey

K. E.

Combinatorial Fibronectin and Laminin Signaling Promote Highly Efficient Cardiac Differentiation of Human Embryonic Stem Cells. Biores. Open Access 2014, 3, 150–161.

31.

Traister

Aafaqi

, et al Integrin-Linked Kinase Mediates Force Transduction in Cardiomyocytes by Modulating SERCA2a/PLN Function. Nat. Commun. 2014, 5, 4533.

32.

Schrader

Tuft

S. J.

Beaconsfield

, et al Evaluation of Human MRC-5 Cells as a Feeder Layer in a Xenobiotic-Free Culture System for Conjunctival Epithelial Progenitor Cells. Curr. Eye Res. 2012, 37, 1067–1074.

33.

Bian

Lin

, et al Identification of Human Fibroblast Cell Lines as a Feeder Layer for Human Corneal Epithelial Regeneration. PLoS One 2012, 7, e38825.

34.

Otsuka

Sasaki

Okimura

, et al Micropatterned Co-Culture of Hepatocyte Spheroids Layered on Non-Parenchymal Cells to Understand Heterotypic Cellular Interactions. Sci. Technol. Adv. Mater. 2013, 14, 065003.

35.

Sullivan

C. W.

Expression of a Connexin 43/Beta-Galactosidase Fusion Protein Inhibits Gap Junctional Communication in NIH3T3 Cells. J. Cell Biol. 1995, 130, 419–429.

36.

Saito

Schlegel

Andresson

, et al Induction of Cell Transformation by Mutated 16K Vacuolar H+-Atpase (ductin) Is Accompanied by Down-Regulation of Gap Junctional Intercellular Communication and Translocation of Connexin 43 in NIH3T3 Cells. Oncogene 1998, 17, 1673–1680.

37.

Hosoda

Zheng

Cabral-da-Silva

, et al Human Cardiac Stem Cell Differentiation Is Regulated by a Mircrine Mechanism. Circulation 2011, 123, 1287–1296.

38.

Giacomelli

Meraviglia

Campostrini

, et al Human-iPSC-Derived Cardiac Stromal Cells Enhance Maturation in 3D Cardiac Microtissues and Reveal Non-Cardiomyocyte Contributions to Heart Disease. Cell Stem Cell 2020, 26, 862–879.

39.

Doss

M. X.

Di Diego

J. M.

Goodrow

R. J.

, et al Maximum Diastolic Potential of Human Induced Pluripotent Stem Cell-Derived Cardiomyocytes Depends Critically on I(Kr). PLoS One 2012, 7, e40288.

40.

Bett

G. C.

Kaplan

A. D.

Lis

, et al Electronic “Expression” of the Inward Rectifier in Cardiocytes Derived from Human-Induced Pluripotent Stem Cells. Heart Rhythm 2013, 10, 1903–1910.

41.

Jonsson

M. K.

Vos

M. A.

Mirams

G. R.

, et al Application of Human Stem Cell-Derived Cardiomyocytes in Safety Pharmacology Requires Caution Beyond hERG. J. Mol. Cell. Cardiol. 2012, 52, 998–1008.

42.

Lieu

D. K.

J. D.

Chiamvimonvat

, et al Mechanism-Based Facilitated Maturation of Human Pluripotent Stem Cell-Derived Cardiomyocytes. Circ. Arrhythm. Electrophysiol. 2013, 6, 191–201.

43.

Herron

T. J.

Rocha

A. M.

Campbell

K. F.

, et al Extracellular Matrix-Mediated Maturation of Human Pluripotent Stem Cell-Derived Cardiac Monolayer Structure and Electrophysiological Function. Circ. Arrhythm. Electrophysiol. 2016, 9, e003638.

44.

Tertoolen

L. G. J.

Braam

S. R.

van Meer

B. J.

, et al Interpretation of Field Potentials Measured on a Multi Electrode Array in Pharmacological Toxicity Screening on Primary and Human Pluripotent Stem Cell-Derived Cardiomyocytes. Biochem. Biophys. Res. Commun. 2018, 497, 1135–1141.

45.

Jia

Valiunas

, et al Stimulating Cardiac Muscle by Light: Cardiac Optogenetics by Cell Delivery. Circ. Arrhythm. Electrophysiol. 2011, 4, 753–760.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.48 MB

A Scalable Approach Reveals Functional Responses of iPSC Cardiomyocyte 3D Spheroids