Human-induced pluripotent stem cell-derived cardiomyocytes (hiPSC-CMs) have emerged as a promising in vitro model for drug-induced cardiotoxicity screening and disease modeling. Here, we detail a protocol for measuring the contractility and electrophysiology of hiPSC-CMs.
Drug-induced cardiotoxicity is the leading cause of drug attrition and withdrawal from the market. Therefore, using appropriate preclinical cardiac safety assessment models is a critical step during drug development. Currently, cardiac safety assessment is still highly dependent on animal studies. However, animal models are plagued by poor translational specificity to humans due to species-specific differences, particularly in terms of cardiac electrophysiological characteristics. Thus, there is an urgent need to develop a reliable, efficient, and human-based model for preclinical cardiac safety assessment. Human-induced pluripotent stem cell-derived cardiomyocytes (hiPSC-CMs) have emerged as an invaluable in vitro model for drug-induced cardiotoxicity screening and disease modeling. hiPSC-CMs can be obtained from individuals with diverse genetic backgrounds and various diseased conditions, making them an ideal surrogate to assess drug-induced cardiotoxicity individually. Therefore, methodologies to comprehensively investigate the functional characteristics of hiPSC-CMs need to be established. In this protocol, we detail various functional assays that can be assessed on hiPSC-CMs, including the measurement of contractility, field potential, action potential, and calcium handling. Overall, the incorporation of hiPSC-CMs into preclinical cardiac safety assessment has the potential to revolutionize drug development.
Drug development is a long and expensive process. A study of new therapeutic drugs approved by the US Food and Drug Administration (FDA) between 2009 and 2018 reported that the estimated median cost of capitalized research and clinical trials was $985 million per product1. Drug-induced cardiotoxicity is the leading cause of drug attrition and withdrawal from the market2. Notably, cardiotoxicity is reported among multiple classes of therapeutic drugs3. Therefore, cardiac safety assessment is a crucial component during the drug development process. The current paradigm for cardiac safety assessment is still highly dependent on animal models. However, species differences from the use of animal models are increasingly recognized as a primary cause of inaccurate predictions for drug-induced cardiotoxicity in human patients4. For example, the morphology of cardiac action potential differs substantially between humans and mice due to the contributions from different repolarizing currents5. In addition, differential isoforms of cardiac myosin and circular RNAs that can impact cardiac physiology have been well documented among species6,7. To bridge these gaps, it is imperative to establish a reliable, efficient, and human-based model for preclinical cardiac safety assessment.
The groundbreaking invention of induced pluripotent stem cell (iPSC) technology has generated unprecedented drug screening and disease modeling platforms. Over the past decade, methods to generate human-induced pluripotent stem cell-derived cardiomyocytes (hiPSC-CMs) have become well established8,9. hiPSC-CMs have attracted great interest in their potential applications in disease modeling, drug-induced cardiotoxicity screening, and precision medicine. For instance, hiPSC-CMs have been utilized to model the pathologic phenotypes of cardiac diseases caused by genetic inheritance, such as long QT syndrome10, hypertrophic cardiomyopathy11,12, and dilated cardiomyopathy13,14,15. Consequently, key signaling pathways implicated in the pathogenesis of cardiac diseases have been identified, which can shed light on potential therapeutic strategies for effective treatment. Moreover, hiPSC-CMs have been used to screen drug-induced cardiotoxicity associated with anticancer agents, including doxorubicin, trastuzumab, and tyrosine kinase inhibitors16,17,18; strategies to mitigate the resultant cardiotoxicity are under investigation. Finally, the genetic information retained in hiPSC-CMs allows for the screening and prediction of drug-induced cardiotoxicity at both individual and population levels19,20. Collectively, hiPSC-CMs have proven to be an invaluable tool for personalized cardiac safety prediction.
The overall goal of this protocol is to establish methodologies to comprehensively and efficiently investigate the functional characteristics of hiPSC-CMs, which are of great importance in applying hiPSC-CMs toward disease modeling, drug-induced cardiotoxicity screening, and precision medicine. Here, we detail an array of functional assays to assess the functional properties of hiPSC-CMs, including the measurement of contractility, field potential, action potential, and calcium (Ca2+) handling (Figure 1).
1. Preparation of media and solutions
2. Measurement of hiPSC-CM contraction motion
3. Measurement of hiPSC-CM field potential
4. Measurement of hiPSC-CM action potential
5. Measurement of hiPSC-CM Ca2+ transient
This protocol describes how to measure the contraction motion, field potential, action potential, and Ca2+ transient of hiPSC-CMs. A schematic diagram including the enzymatic digestion, cell seeding, maintenance, and functional assay conduction is shown in Figure 1. The formation of the hiPSC-CM monolayer is necessary for the contraction motion measurement (Figure 2B). A representative trace of the contraction-relaxation motion of hiPSC-CMs is shown in Figure 2C. The analyzer software enables the analysis of contraction-relaxation motion traces in an automated fashion by detecting the contraction start, contraction peak, contraction end, relaxation peak, and relaxation end using the default setting (Figure 2D). The velocities of contraction and relaxation peaks are used to assess the contraction and relaxation capacities of hiPSC-CMs. In addition, contraction and relaxation deformation distances can also be obtained.
MEA assay is the extracellular recording of action potentials, known as field potentials. Successful measurement of field potentials requires full coverage of the hiPSC-CM monolayer on all the electrodes within each well of the MEA plates (Figure 3C). The mature waveforms recorded by each electrode must reflect the cardiac field potential with easily identifiable characteristics, as shown in Figure 3D–F. The specified quality standards are 1) the baseline activity should show a spontaneous beating within 20-90 beats per min, 2) the beating rate should be within 6 SDs calculated for the baseline beating rate on all of the wells, 3) the coefficient of variation of the beat period should be less than 5%, and 4) the depolarization amplitude should be over 0.3 mV, as previously published22. Figure 3E shows representative field potential traces of hiPSC-CMs. The navigator software detects and identifies beats and FPD automatically. After analysis, beat period, FPD, and spike amplitude can be obtained (Figure 3F). Specifically, the beat period is determined by the depolarizing peak-to-peak interval, FPD is determined by the interval between depolarizing peak to repolarization peak, and spike amplitude is measured by the distance between the depolarizing positive peak to the negative peak, as shown in Figure 3F.
Single-cell patch clamp is the gold standard for assessing the electrophysiological properties of hiPSC-CMs (Figure 4B). Whole-cell patch-clamp recordings in current-clamping mode can record the spontaneous action potentials of single hiPSC-CMs (Figure 4F). After analysis, several parameters, including the amplitude, max diastolic potential (MDP), and action potential duration (APD) at 30%, 50%, and 90% repolarization can be obtained (Figure 4G).
For Ca2+ imaging, fluorescence intensity changes of individual hiPSC-CMs over time can be recorded by the software. A representative area of Fura-2-loaded hiPSC-CMs is shown in Figure 5B. The usage of an ultra-high-speed wavelength-switching light source allows for the recording of Ca2+ transients from hundreds of hiPSC-CMs via frame mode scanning. After the analysis by a spreadsheet-based program23, the F340/F380 ratio over time for each cell can be obtained. Then, raw traces of stimulated Ca2+ transients can be plotted (Figure 5C). In addition, parameters such as amplitude, diastolic Ca2+, and Ca2+ decay tau can be obtained (Figure 5D).
Figure 1. Overall schematic diagram of the protocol. On day 0, digest and seed hiPSC-CMs on basement membrane matrix-coated 96-well plates/48-well MEA plates/35 mm dishes/cell chambers. Allow a recovery period of at least 10 days before performing the functional assays using a cell motion imaging system, MEA, patch-clamp, and Ca2+ imaging system. Abbreviations: MEA = microelectrode array. Please click here to view a larger version of this figure.
Figure 2. Measurement of contraction motion. (A) The cell motion imaging system. (B) A representative region of the hiPSC-CM monolayer. Scale bar: 20 µm. (C) Representative contraction-relaxation traces of hiPSC-CMs. (D) Schematic diagram of a contraction-relaxation trace showing the contraction start, contraction peak, contraction end, relaxation peak, and relaxation end. Please click here to view a larger version of this figure.
Figure 3. Measurement of field potential. (A) Top views of (left) a 48-well MEA plate, (middle) one well with 16 electrodes, and (right) the area that a basement membrane matrix-droplet needs to cover. Add sufficient distilled water to the water reservoir to make up for evaporation. (B) hiPSC-CM suspension and the desired density after replating. (C) A representative region of the hiPSC-CM monolayer. (D) Continuous waveform plots from the 16 electrodes in one well of the 48-well MEA plate. (E) Representative field potential traces of hiPSC-CMs. (F) Schematic diagram of a field potential trace showing how the parameters were analyzed. Abbreviation: FPD = field potential duration. Please click here to view a larger version of this figure.
Figure 4. Measurement of action potential. (A) Basement membrane matrix-coating and hiPSC-CMs suspension placement on 35 mm dishes. (B) A representative region of single hiPSC-CMs. Scale bar: 20 µm. (C–E) Patch-clamp system setup. (F) Representative action potential traces of hiPSC-CMs. (G) Schematic diagram of an action potential trace showing how the parameters were analyzed. Abbreviations: APD = action potential duration, MDP = maximum diastolic potential. Please click here to view a larger version of this figure.
Figure 5. Measurement of Ca2+ transient. (A) The Ca2+ imaging system. (B) A representative region of Fura-2-loaded single hiPSC-CMs. Scale bar: 100 µm. (C) Representative Ca2+ transient traces of hiPSC-CMs. (D) Schematic diagram of a Ca2+ transient trace showing how the parameters were analyzed. Please click here to view a larger version of this figure.
Human iPSC technology has emerged as a powerful platform for disease modeling and drug screening. Here, we describe a detailed protocol for measuring hiPSC-CM contractility, field potential, action potential, and Ca2+ transient. This protocol provides a comprehensive characterization of hiPSC-CM contractility and electrophysiology. These functional assays have been applied in multiple publications from our group12,13,18,24,25,26,27.
It is essential to use the hiPSC-CMs that are in good beating conditions, otherwise, the signal-to-noise ratio of the readouts will drop dramatically. In addition, it is desirable to ensure the high purity of hiPSC-CMs before the measurements. The existence of non-cardiomyocytes could particularly compromise data accuracy of contraction motion and field potential measurements, both of which require the formation of hiPSC-CM monolayer. Furthermore, the immaturity of hiPSC-CMs remains a major concern that hinders the full application of hiPSC-CMs in various fields28. It is known that hiPSC-CMs are functionally immature in multiple aspects, including morphology, contractility, electrophysiology, Ca2+ handling, and metabolism29. Therefore, it is recommended to keep the hiPSC-CMs in the culture at least until day 30 before performing functional assays. Alternatively, various strategies that have been established to improve the maturity of hiPSC-CMs can be adopted before conducting the functional assays30. For example, the hiPSC-CMs cultured in metabolic maturation media exhibited highly negative MDP of action potentials and significantly faster decay of Ca2+ transients when compared with the hiPSC-CMs cultured in normal RPMI media31. Finally, there are batch-to-batch variations of hiPSC-CMs32. Thus, data should be collected on hiPSC-CMs generated from at least three independent differentiations of the same iPSC lines.
The cell motion imaging system provides a high throughput and efficient platform to measure the contraction and relaxation motion of hiPSC-CMs. However, a major disadvantage is that it does not measure the direct contractile force but rather the contraction and relaxation velocity instead. In addition, the velocities vary when the number of hiPSC-CMs to form a monolayer is different. Therefore, it is crucial to seed precisely 50,000 cells in each well of the 96-well plates.
FPD measured by MEA has been used as a surrogate for QT duration and thus is widely used in assessing drug-induced QT prolongation33. Similar to QT duration, FPD is beating rate dependent. Therefore, the correction in beating rate is necessary to precisely interpret the FPD data. In this protocol, it is recommended that MEA measurement is performed at least 10 days post-plating. However, sometimes it is possible that the field potential signal is still unstable after 10 days post-plating. If this happens, prolong the hiPSC-CM culture for another 2-3 days.
The patch-clamp technique is a versatile tool for assessing the electrophysiology of hiPSC-CMs and assaying ion channels. The major disadvantage of the conventional patch-clamp is its low throughput property, which hinders its application in high throughput studies. In addition, it requires a long learning curve and extensive practice before one can become experienced in this technique. However, the conventional patch-clamp technique is still considered the gold standard to understand the electrophysiology of hiPSC-CMs. If the APD of hiPSC-CMs needs to be evaluated in a high throughput manner, it is also suggested to start with optical imaging of hiPSC-CM action potential using the accelerated sensor of action potentials 2 (ASAP2)21. After primary targets are obtained, a conventional action potential can be used in further studies for more accurate phenotyping and validation. In addition, automated patch clamp has been increasingly employed in various fields, especially channelopathy. For instance, the ion channel currents carried by multiple variants of hERG channels and voltage-gated sodium channels have been characterized by automated patch clamp34,35,36,37. Moreover, the usage of automated patch clamp in hiPSC-CMs has been established recently38,39. For the execution of automated patch clamp, readers can refer to previously published protocols40. With the advantages of being high throughput and the simplicity in machine handling, there is no doubt that automated patch clamps will play more and more important roles in drug discovery and drug development.
The usage of a ratiometric Ca2+-sensitive dye, Fura-2, allows the measurement of diastolic Ca2+ of hiPSC-CMs to be much less affected by the experimental bias factors such as dye loading time, uneven dye distribution, and photo-bleaching. This is important to recapitulate the diastolic dysfunction of particular heart diseases. The main limitation is that it requires a UV excitation light source, which may be incompatible with confocal microscopy.
In summary, despite recent progress in cardiotoxicity assessment in the preclinical phase of drug development, cardiotoxicity remains one of the leading safety concerns. The growing incorporation of hiPSC-CMs into the standard preclinical safety evaluation process has the potential to improve the accuracy of toxicity prediction of candidate compounds. In addition, patient-specific hiPSC-CMs enable personalized cardiac drug selection and adverse drug response prediction. The protocol using various functional assays described here will help expand the applications of hiPSC-CMs in drug screening and disease modeling.
The authors have nothing to disclose.
We thank Blake Wu for proofreading the manuscript. This work was supported by the National Institutes of Health (NIH) R01 HL113006, R01 HL141371, R01 HL163680, R01 HL141851, U01FD005978, and NASA NNX16A069A (JCW), and AHA Postdoctoral Fellowship 872244 (GMP).
35 mm glass bottom dish with 20 mm micro-well #1.5 cover glass | Cellvis | D35-20-1.5-N | Patch clamp |
50x B27 supplements | Life Technologies | 17504-044 | hiPSC-CM culture medium |
6-well culture plate | E & K Scientific | EK-27160 | hiPSC-CM culture |
96-well flat clear bottom black polystyrene TC-treated microplates | Corning | 3603 | Contraction motion measurement |
Accutase | Sigma-Aldrich | A6964 | Enzymatic dissociation |
Axion's Integrated Studio (AxIS) | Axion Biosystems | navigator software | |
Borosilicate glass capillaries | Harvard Apparatus | BF 100-50-10, | Patch clamp |
CaCl2 1 M in H2O | Sigma-Aldrich | 21115 | Tyrode’s solution |
Cell counting chamber slides | ThermoFisher Scientific | C10228 | Cell counting |
CytoView 48-well MEA plates | Axion Biosystems | M768-tMEA-48B | MEA |
DMEM/F12 | Gibco/Life Technologies | 12634028 | Extracellular matrix medium |
DPBS, no calcium, no magnesium | Fisher Scientific | 14-190-250 | |
EGTA | Sigma-Aldrich | E3889 | Intracellular pipette solution |
EPC 10 USB patch clamp amplifier | Warner Instruments | 89-5000 | Patch clamp |
Fura-2, AM, cell permeant | ThermoFisher Scientific | F1221 | Ca2+ transient measurement |
Glucose | Sigma-Aldrich | G8270 | Tyrode’s solution |
HEPES | Sigma-Aldrich | H3375 | Tyrode’s solution |
hiPSCs | Stanford Cardiovascular Institute iPSC Biobank | ||
KCl | Sigma-Aldrich | 529552 | Tyrode’s solution |
KnockOut Serum Replacement | ThermoFisher Scientific | 10828-028 | hiPSC-CM seeding medium |
KOH 8 M | Sigma-Aldrich | P4494 | Intracellular pipette solution |
Lambda DG 4 | Sutter Instrument Company | Ca2+ transient measurement; ultra-high-speed wavelength switching light source | |
Luna-FL automated fluorescence cell counter | WISBIOMED | LB-L20001 | Cell counting |
Maestro Pro MEA system | Axion Biosystems | MEA | |
Matrigel Growth Factor Reduced (GFR) Basement Membrane Matrix | Corning | 356231 | Extracellular matrix medium |
MgATP | Sigma-Aldrich | A9187 | Intracellular pipette solution |
MgCl2 | Sigma-Aldrich | M8266 | Tyrode’s solution |
NaCl | Sigma-Aldrich | S9888 | Tyrode’s solution |
NaOH 10 M | Sigma-Aldrich | 72068 | Tyrode’s solution |
NIS Elements AR | |||
Pluronic F-127 (20% Solution in DMSO) | ThermoFisher Scientific | P3000MP | Ca2+ transient measurement |
RPMI 1640 medium | Life Technologies | 11875-119 | hiPSC-CM culture medium |
Sony SI8000 Cell Motion Imaging System | Sony Biotechnology | Contraction motion measurement | |
Sutter Micropipette puller | Sutter Instruments | P-97 | Patch clamp |
Trypan blue stain | Life Technologies | T10282 | Cell counting |