Download PDF

Angeline Lim, PhD | Applications Scientist | Molecular Devices, LLC

Paula Gedraitis, PhD | Technical Support Manager | Molecular Devices, LLC

Yu-Fen Chang, PhD | Chief Executive Officer | LumiSTAR Biotechnology

Introduction

Induced pluripotent stem cell-derived cardiomyocytes (iPSC-CMs) have been used for disease modeling and drug discovery as they recapitulate essential aspects of human biology and align with the principles of the 3Rs (Replacement, Reduction, and Refinement) for animal studies. Fluorescent calcium readouts are often used as a surrogate marker for cardiac functions, and a great variety of commercially available organic calcium indicators has made them drug discovery workhorses. However, chemical dye-based calcium-sensing probes directly inhibit the Na, K-ATPase and disrupt cellular function. Thus, tracking a drug’s chronic effect on cardiac function over hours or days has been technically challenging in the presence of chemical calcium probes.

One solution to this issue is applying fluorescent genetically encoded calcium indicators (GECIs) to study iPSC-CMs. They can be genetically engineered to traffic to specific cellular compartments and are generally less toxic than chemical calcium dyes. In addition, an increasing number of variants covering a broad excitation/emission and calcium affinity spectrum makes them ideal tools for multi-parametric measurement of cardiac functions over extended periods of time.

Here we demonstrate an in vitro model using iPSC-CMs constantly expressing a genetically encoded calcium indicator for monitoring cell responses to various compounds for an extended period of time (up to 60 days). In addition, an optogenetic actuator was used to pace the cardiomyocytes in a noninvasive and reversible manner, mimicking a physiological steady rhythm. This contactless control and recording of cardiomyocytes cannot be easily achieved through conventional electrical stimulation.

Methods

iPSC-derived cardiomyocytes

iPSC-CMs (LumiCardio iCM003, LumiSTAR Biotechnology) were used to conduct the experiments. Cells were plated in fibronectin/gelatin-coated 96-well microplate (Greiner 655090) at a density of 100,000–150,000 cells per cm2. After 24 hours, media was replaced with pre-warmed maintenance media. For long-term evaluation of cardiotoxicity or therapeutic effect, 50% of media was replaced with mock- or drug-containing fresh media every two days.

Actuator and calcium sensor transduction

72 hours after cell plating, 50% of media was replaced with 2x viral solution in maintenance media. Media was then refreshed after 16 hours of incubation at 37°C and culture was maintained in a humidified incubator. iPSC-CMs were infected with the cyto.Calcium Green2 (AL008, LumiSTAR Biotechnology) or lentivirus carrying Channelrhodopsin-2 and cyto.Calcium Red (AL007, LumiSTAR Biotechnology) at day 25 post-differentiation. Fluorescence from cyto.Calcium was visible under a microscope after 24 hr, and the expression level reached a plateau after 48–72 hr.

Calcium indicator staining

Cardiomyocytes were stained with the Fluo-4 calcium indicator according to the vendor’s instructions (Thermo Fisher F14201). Fluo-4 stock solution in DMSO was diluted to 2x concentration (10 µM) in pre-warmed Tyrode’s buffer (Sigma-Aldrich T2145) and added to the culture for a 30-minute incubation at 37°C. The cells were then washed with pre-warmed Tyrode’s buffer and immediately loaded into the ImageXpress® Micro Confocal HighContent Imaging System with environment control for image acquisition.

Calcium oscillation imaging and analysis

The spontaneous beating of cardiomyocytes was recorded on the ImageXpress Micro Confocal system at 25 frames per second for 30 seconds using a custom journal from the MetaXpress® High-Content Image Acquisition and Analysis Software. The FITC channel was used to image cyto.Calcium Green2 or Fluo-4, and the Texas Red channel was used to image cyto.Calcium Red. For optogenetics stimulation of cardiomyocytes, the MetaXpress software was configured to trigger a low-energy blue light (Cyan LED) emitting a 20ms pulse every 25 frames (1 Hz) parallel to cyto.Calcium Red imaging.

Recorded image stacks were analyzed using software specifically designed for processing calcium oscillation (LumiSTAR LUCS01a). The initial 10 seconds of decaying signal due to photobleaching were skipped, and only the remaining 20 seconds of steady oscillation were analyzed for amplitude, frequency (BPM), rise time, decay time, and calcium transient duration (CTD).

Results

Comparing the toxicity of GECI and chemical calcium indicators on iPSC-CMs beating

A typical phenotype of functional iPSC-CMs is spontaneous contraction. iPSC-CMs expressing genetically encoded calcium indicators, such as cyto.Calcium Green2, can reveal the underlying rhythmic calcium oscillation when recorded at near video rate using an imaging instrument (Figure 1A). This fluorescent readout of calcium oscillation is comparable to those observed in iPSC-CMs stained with chemical calcium indicators, such as Fluo-4 AM (Figure 1B). However, the Fluo-4 staining resulted in a gradual decrease in the oscillation frequency and amplitude, indicating a toxic effect on the iPSC-CMs due to such chemical staining. Meanwhile, the pattern of cells expressing cyto.Calcium Green2 remains steady for hours. In fact, iPSC-CMs expressing GECI show persistent brightness and calcium oscillation even weeks after the first transduction.

To demonstrate the potential application of GECI in evaluating drug effects, we tested several well-characterized small molecule ion channel inhibitors, such as E4031, Dofetilide, and Verapamil, and the alterations of calcium transients were comparable to those observed in other published results (Figure 2).

Figure 1. (A) Time-lapse image series of contracting iPSC-derived cardiomyocytes loaded with cytoCalcium Green 2. Each image was taken at 40 millisecond intervals. Yellow/red (false color scale) indicates a high Ca2+ concentration and a state of contraction. (B) Traces of fluorescence signal over time were recorded at different time points. GECI (upper panel) shows little to no toxicity for weeks compared to chemical stain (lower panel), which impairs iPSC-CMs beating one hour after staining.

Figure 2. (A) Cardiac beating patterns in the presence of several reference small molecule drugs. (B) Definition of measured characteristics of a typical cardiac calcium transient. (C) Statistic results of the peak amplitudes, calcium transient durations, and rise and decay attributes.

Controlling iPSC-CMs beating to simulate physiological conditions

Fully differentiated iPSC-CMs showed spontaneous calcium oscillation in vitro. Although many beat at a frequency similar to a physiological condition, typically 1 to 2 Hz, the patterns of such spontaneous calcium oscillation varied from well to well (Figure 3A left panel), and the frequency may change over time. In this regard, optogenetics, a technique based on light-sensitive proteins and optical manipulation offers an unprecedented opportunity for precise control of cardiomyocyte beating. Compared to traditional MEA recordings, which primarily measure electrical activity passively, by genetically modifying iPSC-CMs to express light-sensitive proteins (e.g., channelrhodopsin), researchers can use light pulses to stimulate these cells with high temporal and spatial precision.

We introduced channelrhodopsin 2 as an actuator in conjunction with a red-shift GECI, cyto.Calcium Red, to transduce iPSC-CMs. By employing repetitive light pulses to stimulate these cells, we could effectively pace them and adjust their beating frequency, regardless of their initial rhythm. This approach yielded a uniform beating pattern across the culture plate (Figure 3B right panel). Consequently, we gained the ability to modulate the frequency and duration of light stimulation, facilitating the simulation of various physiological and pathological conditions—a valuable feature for studying heart function.

To illustrate the applicability of optogenetic modulation on iPSC-CMs to assess cardiotoxicity, we subjected the model to several known cardiotoxic compounds. After incubation in test compounds for 2 hours, cardiomyocyte beating was simultaneously paced at 1 Hz and imaged. Analysis of calcium transients shows that all compounds tested affected the beating patterns in a dose-dependent manner (Figure 4), indicating the potential of utilizing GECI and optogenetics on high-throughput cardiotoxicity screening.

Figure 3. (A) Calcium transients before and after pacing. Spontaneous beating pattern (gray) varied from well to well. After being paced with 1 Hz cyan pulses (blue), iPSC-CMs showed a consistent beating pattern. (B) Quantification of beating frequency. Paced iPSC-CMs show a more uniform beating frequency than spontaneous beating.

Figure 4. (A) Calcium transient analysis of paced cardiomyocyte responses in the presence of different concentrations of test compounds. All compounds tested altered iPSC-CMs beating in a dose-dependent manner. (B) The IC50s of various test compounds were quantified from the average amplitudes and decay slopes of paced calcium transients.

Figure 5. The workflow for long-term assessment of cardiotoxicity using GECI.

Conclusion

Our proof-of-concept result showed minimal toxicity and constitutive expression of GECI in iPSC-CMs making it a robust alternative for repeated monitoring of the sample for long periods (Figure 5). Off-the-shelf GECI simplifies the experimental procedure and reduces the number of iPSC-CMs and probes needed for long-term observation, and the optogenetic tool adds another level of control over cardiomyocytes beating in various physiological conditions. A robust imaging platform that can perform sophisticated optogenetics pacing and calcium imaging – like the ImageXpress portfolio of high-content imaging systems – is imperative to obtaining consistent results, allowing the quick and reliable assessment of cardiotoxicity as well as evaluating therapeutic effects between healthy and patient samples over a long period.

Download PDF