This method can be used to examine sarcomere shortening using pluripotent stem cell-derived cardiomyocytes with fluorescent-tagged sarcomere proteins.
Pluripotent stem cell-derived cardiomyocytes (PSC-CMs) can be produced from both embryonic and induced pluripotent stem (ES/iPS) cells. These cells provide promising sources for cardiac disease modeling. For cardiomyopathies, sarcomere shortening is one of the standard physiological assessments that are used with adult cardiomyocytes to examine their disease phenotypes. However, the available methods are not appropriate to assess the contractility of PSC-CMs, as these cells have underdeveloped sarcomeres that are invisible under phase-contrast microscopy. To address this issue and to perform sarcomere shortening with PSC-CMs, fluorescent-tagged sarcomere proteins and fluorescent live-imaging were used. Thin Z-lines and an M-line reside at both ends and the center of a sarcomere, respectively. Z-line proteins — α-Actinin (ACTN2), Telethonin (TCAP), and actin-associated LIM protein (PDLIM3) — and one M-line protein — Myomesin-2 (Myom2) — were tagged with fluorescent proteins. These tagged proteins can be expressed from endogenous alleles as knock-ins or from adeno-associated viruses (AAVs). Here, we introduce the methods to differentiate mouse and human pluripotent stem cells to cardiomyocytes, to produce AAVs, and to perform and analyze live-imaging. We also describe the methods for producing polydimethylsiloxane (PDMS) stamps for a patterned culture of PSC-CMs, which facilitates the analysis of sarcomere shortening with fluorescent-tagged proteins. To assess sarcomere shortening, time-lapse images of the beating cells were recorded at a high framerate (50-100 frames per second) under electrical stimulation (0.5-1 Hz). To analyze sarcomere length over the course of cell contraction, the recorded time-lapse images were subjected to SarcOptiM, a plug-in for ImageJ/Fiji. Our strategy provides a simple platform for investigating cardiac disease phenotypes in PSC-CMs.
Cardiovascular diseases are the leading cause of mortality worldwide1 and cardiomyopathy represents the third cause of cardiac-related deaths2.Cardiomyopathy is a collective group of diseases that affect cardiac muscles. The recent developments of induced pluripotent stem (iPS) cells and the directed-differentiation of iPS cells toward cardiomyocytes (PSC-CMs) have opened the door for studying cardiomyocytes with patient genome as an in vitro model of cardiomyopathy. These cells can be used to understand the pathophysiology of cardiac diseases, to elucidate their molecular mechanisms, and to test different therapeutic candidates3. There is a tremendous amount of interest, thus, patient-derived iPS cells have been generated (e.g., hypertrophic cardiomyopathy [HCM]4,5, arrhythmogenic right ventricular cardiomyopathy [ARVC]6, dilated cardiomyopathy [DCM]7, and mitochondrial-related cardiomyopathies8,9). Because one of the characteristics of cardiomyopathy is the dysfunction and disruption of sarcomeres, a valid tool that uniformly measures sarcomere function is needed.
Sarcomere shortening is the most widely used technique to assess sarcomere function and the contractility of adult cardiomyocytes derived from animal models and humans. To perform sarcomere shortening, well-developed sarcomeres that are visible under phase-contrast are required. However, PSC-CMs cultured in vitro display underdeveloped and disorganized sarcomeres and, therefore, are unable to be used to properly measure sarcomere shortening10. This difficulty to properly assess the contractility of PSC-CMs hinders their usage as a platform to assess cardiac functions in vitro. To assess PSC-CMs contractility indirectly, atomic force microscopy, micro-post arrays, traction force microscopy, and impedance measurements have been used to measure the effects of the motion exerted by these cells on their surroundings11,12,13. More sophisticated and less invasive video-microscopy recordings of actual cellular motion (e.g., SI8000 from SONY) can be used to alternatively assess their contractility, however, this method does not directly measure sarcomere motion or force generation kinetics14.
To directly measure sarcomere motion in PSC-CMs, new approaches, such as fluorescent-tagging to sarcomere protein, are emerging. For example, Lifeact is used to label filamentous actin (F-actin) to measure sarcomere motion15,16. Genetically modified iPS cells are another option for tagging sarcomere proteins (e.g., α-actinin [ACTN2] and Myomesin-2 [MYOM2]) by fluorescent protein17,18,19.
In this paper, we describe how to perform time-lapse imaging for measuring sarcomere shortening using Myom2-TagRFP (mouse embryonic stem [ES] cells) and ACTN2-mCherry (human iPS cells). We also show that a patterned culture facilitates sarcomere alignment. In addition, we describe an alternative method of sarcomere labeling, using adeno-associated viruses (AAVs), which can be widely applied to patient-derived iPS cells.
1. Differentiation of mouse pluripotent stem cells
2. Differentiation of human pluripotent stem cells
3. Fluorescent labeling of sarcomeres using adeno-associated viruses
4. [Optional] AAV-based purification of PSC-CMs
5. Preparation of PDMS stamps
6. Patterned culture of pluripotent stem cell-derived cardiomyocytes
7. Time-lapse imaging of sarcomeres under fluorescent microscope
8. Analysis of time-lapse imaging using SarcOptiM, an ImageJ/Fiji plugin
Measuring sarcomere shortening using knock-in PSC-CMs reporter lines. Sarcomere-labeled PSC-CMs were used to measure sarcomere shortening. The lines express Myom2-RFP and ACTN2-mCherry from endogenous loci. TagRFP was inserted to Myom2, coding M-proteins that localize to the M-line, while mCherry was knocked-in to ACTN2, coding α-Actinin, which localizes to the Z-line18,25. Time-lapse images were obtained and used to determine sarcomere shortening as presented in Figures 1 and 2 and Movie 1-3.
To overcome the disorganized sarcomere of PSC-CMs, specific PDMS stamps were used to culture PSC-CMs in the stripe pattern. This patterned culture promoted an elongated cell shape and a more organized sarcomere pattern compared to cells cultured in the non-pattern area (Figures 2B and 2C). With this advantage, the patterned culture promoted better contraction of the cells and provided a smooth sarcomere length profile as shown in Movies 2 and 3 and Figure 2D.
Fluorescent tagging of Z-line protein using AAV vectors. To visualize the Z-line of PSC-CMs without generating knock-in iPS cells, fluorescent-tagged Z-line proteins were expressed using AAV transduction. Two of small Z-line proteins, Telethonin (TCAP) and Actin-associated LIM protein (PDLIM3) with GFP, were tagged and packaged using the AAV6 capsid (Figure 3A). Once PSC-CMs were differentiated and purified, AAVs were transduced to PSC-CMs (Figure 3B). The transduced PSC-CMs expressed sarcomeric GFP signals along the PSC-CMs as early as three days post-transduction (Figures 3C and 3D). Typically, the transduction of AAV at an MOI of 105 vg/cell is sufficient to visualize fluorescent-tagged sarcomere proteins and a higher titer may cause non-specific localization of GFP to cytoplasm though it increases overall GFP intensity.
Purification of PSC-CMs using AAV vectors. Current methods rely on the drug selection cassette that is already on the genome of PSC-CMs, either transgenic or knock-in line. However, it is labor-intensive to produce such a line from patient-derived iPS cells. As AAV vectors have been demonstrated to drive the expression of fluorescent-labeled Z-line proteins without the need for knock-in, we sought to establish the purification method without knock-in as well (Figure 4). To this end, a new AAV vector, which encodes blasticidin-resistant gene under the control of cTNT promoter, was constructed (Figure 4A). The AAV (MOI of 105 vg/cell) was transduced to differentiating human iPS cells at day 4. Then cells were treated with 2.5-10 μg/mL of blasticidin (need to titrate for each cell line) between days 7 and 9 (Figure 4B). At day 14, the purity of PSC-CMs was more than 90% (Figure 4C).
Figure 1: Sarcomere shortening of the mouse PSC-CMs derived from the Myom2-TagRFP cell line. A. The timeline for mouse PSC-CM differentiation. B. Representative images for sarcomere shortening in different time points with measuring regions as indicated by yellow bars. Scale bar = 10 μm. C. Sarcomere length profile during contraction of the cardiomyocytes that was stimulated with electricity at 1 Hz. The framerate was 50 frames per second. The pixel size was 0.26 μm. Please click here to view a larger version of this figure.
Figure 2: Representative data showing sarcomere shortening of the human PSC-CMs derived from the ACTN2-mCherry cell line in non-patterned and patterned culture. A. The timeline for human PSC-CM differentiation. B. and C. Cardiomyocytes cultured in non-patterned cultures show disorganized sarcomere patterns (B) while patterned cultures promote a good alignment of the sarcomere (C). Measuring regions presented by yellow bars. D. Corresponding sarcomere length profiles obtained during cell contraction induced by electrical stimulation at 0.5 Hz and a 100 frames per second frame rate. Pixel size = 0.26 μm, scale bar = 10 µm. Please click here to view a larger version of this figure.
Figure 3: Mouse PSC-CMs after AAV transduction for 3 days. A. Schematic vector map of AAV for sarcomere labeling. A sarcomere protein (gene of interest, GOI) is linked to GFP with a Gly-Gly-Gly-Ser linker (L) and expressed under the control of cardiac troponin T (cTNT) promoter. B. The timeline for mouse PSC-CM differentiation and AAV transduction. C. and D. Representative images showing clear sarcomere localization and the corresponding sarcomere length profile of TCAP-GFP (C) and PDLIM3-GFP (D) after 3 days of transduction into PSC-CMs generated from the Myom2-TagRFP cell line. Scale bar = 10 μm. Please click here to view a larger version of this figure.
Figure 4: Blasticidin Purification of human PSC-CMs without knock-in.
A. Schematic vector map of AAV, in which a blasticidin-resistance gene cassette (BSR) is inserted downstream to cTNT promoter. B. The timeline of human PSC-CMs differentiation, transduction, and blasticidin selection. C. Representative data showing percentage of cTNT + cells in human PSC-CMs (transduced 105 vg/cell AAV6 on day 4 then treated with 2.5 µg/mL blasticidin on days 7 and 9). Please click here to view a larger version of this figure.
Movie 1: Fluorescent time-lapse video of mouse PSC-CMs generated from the Myom2-TagRFP cell line. RFP signals showed a sarcomere pattern after PSC-CM culture for 28 days. The cells showed beating synchronously when stimulated with electricity at 1 Hz. The time-lapse images were acquired every 20 ms with a 100X lens. Scale bar = 5 μm. Please click here to download this movie.
Movie 2: Fluorescent time-lapse video of the human PSC-CMs with ACTN2-mCherry cultured on a non-patterned culture dish. The PSC-CMs expressing ACTN2-mCherry on a non-patterned culture dish not only showed disorganization of sarcomere but also presented a waving contraction, for which it is difficult to determine sarcomere shortening. The cells were stimulated with electricity at 0.5 Hz and images acquired every 10 ms with a 100X lens. Scale bar = 10 μm. Please click here to download this movie.
Movie 3: Fluorescent time-lapse video of the human PSC-CMs with ACTN2-mCherry cultured on a patterned culture dish. The patterned culture promoted alignment of the sarcomere and forced the cells to a rod shape. This method allowed the sarcomere shortening in PSC-CMs to be determined more easily. The video was obtained by stimulating the cells with electricity at 0.5 Hz. The framerate was 100 frames per second. Scale bar = 10 μm. Please click here to download this movie.
Supplemental CAD files. CAD files for creating stamps with strips of 200 μm width and grooves of 10 μm (Stamp_200x10.dxf), 25 μm (Stamp_200x25.dxf), and 50 μm (Stamp_200x50.dxf). Please click here to download this file.
PSC-CMs have great potential to be utilized as an in vitro platform to model heart disease and to test the effects of drugs. Nevertheless, an accurate, unified method to assess PSC-CMs functions must first be established. Most of functional tests work with PSC-CMs, e.g., electrophysiology, calcium transient, and metabolism26, and one of the first patient-derived PSC-CM studies was about long-QT syndrome27. However, contractility, one of the most important functions of a cardiomyocyte, is still difficult to assess. With adult cardiomyocytes, sarcomere shortening is widely used. In contrast, due to the underdeveloped and disorganized sarcomere of PSC-CMs, the standard method for sarcomere shortening does not work with these cells. Therefore, we have presented an alternative method for examining sarcomere shortening in PSC-CMs using fluorescent-tagged sarcomere proteins. As demonstrated, proteins localized to the M-line (MYOM2) or Z-line (ACTN2, TCAP, and PDLIM3) fused with fluorescent proteins can be used for this approach. We have also shown that fluorescent-tagged proteins can be expressed from endogenous loci or by AAVs. AAVs provide more flexibility for expressing fluorescent-tagged proteins than do endogenous loci, as AAVs can be applied to any type of patient-derived PSC-CMs. In contrast, expressing proteins from endogenous loci may have a lesser effect on sarcomere function, as the expression level of the genes is tightly regulated and can also be used for monitoring the maturation of PSC-CMs18.
Even though Myom2-TagRFP, ACTN2-mCherry, and Lifeact were all used to examine the sarcomere shortening16,18,19, it is still unclear if these proteins interfere with sarcomere function. Recently, Lifeact was reported to disturb actin organization and cellular morphology28. It is also important to note that fusion patterns (i.e., the GFP fusion site at N-term or C-term of target protein) also affect the sarcomere function29. Therefore, before being used widely, it is important to thoroughly assess whether fluorescent-tagged sarcomere proteins interfere with sarcomere function and whether protein-protein interactions occur in sarcomeres in vitro, in vivo, and/or in adult cardiomyocytes. This repertoire of fluorescent-tagged sarcomere proteins provides a good starting point for future protein-engineering options (i.e., shortening the sarcomere proteins to only localization signals). Selecting proteins to tag is another key to success. We have tagged another Z-line protein with GFP, however, this protein displayed only a cytoplasmic distribution rather than localizing to the sarcomere. For live-imaging, protein stability may also play a role, as, for example, if a tagged protein is unstable, the signal level will be lower. The photostability of the fluorescent protein is also important, as unstable protein signals will be easily quenched during imaging.
To examine the contractility of PSC-CMs using methods other than described, indirect measurements of force generated by PSC-CMs (e.g., micro-post arrays, traction force microscopy) or motion (e.g., high-resolution motion detection using SI8000)11,12,13,14 can be used. Our method can be combined with these methods or with dye-based action potential/calcium transient measurements. The combinatorial approach may provide further information on how a disease causes dysfunction in patient-derived PSC-CMs.
One of the challenges in sarcomere shortening in PSC-CMs is to find a good sarcomere that moves linearly, otherwise, the cells may easily come off the line for sarcomere detection of SarcOptiM and cause unstable sarcomere shortening results. Here, we demonstrate that using a patterned culture generated with PDMS stamps may provide a more stable and linear sarcomere movement. A patterned culture is also known to support the maturation of PSC-CMs16, which is important for sarcomere function.
The authors have nothing to disclose.
We would like to acknowledge all the lab members in the Division of Regenerative Medicine at the Jichi Medical University for the helpful discussion and technical assistance. This study was supported by the grants from the Japan Agency for Medical Research and Development (AMED; JP18bm0704012 and JP20bm0804018), the Japan Society for the Promotion of Science (JSPS; JP19KK0219), and the Japanese Circulation Society (the Grant for Basic Research) to H.U.
1-Thioglycerol | Sigma-Aldrich | M6145-25 | |
2-Mercaptoethanol (55mM) | Thermo Fisher Scientific | 21985-023 | |
2-methacryloyloxyethyl phosphorylcholine (MPC) polymer, | NOF Corp. | LIPIDURE-CM5206 | |
2-Propanol | Fujifilm wako | 166-04836 | |
35-mm imaging dish with a polymer coverslip (µ-Dish 35 mm, high) | ibidi | 81156 | |
AAVproR Helper Free System (AAV6) (vectors; pHelper, pRC6, pAAV-CMV-Vector) |
Takara | 6651 | |
ACTN2-mCherry (AR12, AR21) hiPSCs | N.A. | We inserted IRES-puromycin resistant casette to 3' UTR of TNNT2 locus and mCherry around the stop codon of ACTN2 in 610B1 hiPSC line, following a method describe elsewhere (Anzai, Methods Mol Biol, in press) | |
B-27 Supplement (50X), serum free | Thermo Fisher Scientific | 17504-044 | |
B-27 Supplement, minus insulin | Thermo Fisher Scientific | A18956-01 | |
B27 supplement (50X), minus Vitamin A | Thermo Fisher Scientific | 12587-010 | |
Benzonase (25 U/µL) | Merck Millipore | 70746 | |
Blasticidin S Hydrochloride | Fujifilm wako | 029-18701 | |
BMP-4, Human, Recombinant, | R&D Systems, Inc. | 314-BP-010 | |
Bovine Serum Albumin | Sigma-Aldrich | A4503-100g | |
C59, Wnt Antagonist (WntC59) | abcam | ab142216 | |
CAD drawing software, | Robert McNeel and Associates, WA, USA | Rhinoceros 6.0 | |
Centrifugal ultrafiltration unit (100k MWCO), Vivaspin-20 | Sartorius | VS2042 | |
CHIR99021 | Cayman | 13122 | |
Chromium etchant | Nihon Kagaku Sangyo Co., Ltd., Japan | N14B | |
Chromium mask coated with AZP1350 | Clean Surface Technology Co., Japan | CBL2506Bu-AZP | |
Dr. GenTLE Precipitation Carrier (20mg/mL Glycogen, 3 M Sodium Acetate (pH 5.2)) | Takara | 9094 | |
Dulbecco’s Modified Eagle’s Medium (DMEM) – high glucose | Sigma-Aldrich | D6429-500 | |
Dulbecco’s Modified Eagle’s Medium (DMEM) – high glucose, without sodium pyruvate | Sigma-Aldrich | D5796 | |
Ethanol (99.5) | Fujifilm wako | 057-00456 | |
Fetal Bovine Serum | Moregate | 59301104 | |
FGF-10, Human, Recombinant, | R&D Systems, Inc. | 345-FG-025 | |
Fibroblast Growth Factor(basic), human, recombinant | Fujifilm wako | 060-04543 | |
Gelatin from porcine skin powder | Sigma-Aldrich | G1890-100g | |
Glasgow Minimum Essential Medium (GMEM) | Sigma-Aldrich | G5154-500 | |
GLASS BOTTOM culture plates | MatTek | P24G-1.5-13-F/H | |
Ham’s F-12 | Thermo Fisher Scientific | 11765-062 | |
Iscove's Modified Dulbecco's Medium (IMDM) | Thermo Fisher Scientific | 12440-061 | |
L-alanine-L-glutamine (GlutaMAX Supplement, 200mM) | Thermo Fisher Scientific | 35050-061 | |
L(+)-Ascorbic Acid Sodium Salt | Fujifilm wako | 196-01252 | |
Laminin-511 E8 fragment (LN511-E8, iMatrix-511) | Nippi | 892012 | |
Mask aligner | Union Optical Co., Ltd., Japan | PEM-800 | |
Maskless lithography tool | NanoSystem Solutions, Inc., Japan | D-Light DL-1000 | |
MEM Non-Essential Amino Acids Solution (100X) | Thermo Fisher Scientific | 11140-050 | |
Millex-HV Syringe Filter Unit, 0.45 µm, PVDF (0.45-µm filter) | Merck Millipore | SLHVR33RS | |
Myom2-RFP (SMM18) | N.A. | Developed in our previous paper (Chanthra, Sci Rep, 2020) | |
N-2 Supplement (100X) | Thermo Fisher Scientific | 17502-048 | |
ORCA-Flash4.0 V3 digital CMOS camera | Hamamatsu | C13440-20CU | |
PD0325901 | Stemgent | 04-0006-10 | |
Penicillin-Streptomycin (10,000 U/mL) | Thermo Fisher Scientific | 15140-122 | |
Petri dish | Sansei medical co. Ltd | 01-004 | |
Phenol/Chloroform/Isoamyl alcohol (25:24:1) | Nippon Gene | 311-90151 | |
Polydimethylsiloxane (PDMS) elastomer | Dow Corning Corp., MI, USA | SILPOT 184 | |
polyethylenimine MAX (MW. 40,000) | Polyscience | 24765-1 | |
Positive photoresist developer | Tokyo Ohka Kogyo Co., Ltd., Japan | NMD-3 | |
PowerUp SYBR Green Master Mix | Thermo Fisher Scientific | A25742 | |
Proteinase K | Takara | 9034 | |
Puromycin Dihydrochloride | Fujifilm wako | 166-23153 | |
Recombinant Human/Mouse/Rat Activin A Protein | R&D Systems, Inc. | 338-AC-050 | |
Recombinant trypsin-like protease (rTrypsin; TrypLE express) | Thermo Fisher Scientific | 12604-039 | |
RPMI1640 Medium | Thermo Fisher Scientific | 11875-119 | |
Silicon wafer | Matsuzaki Seisakusyo Co., Ltd., Japan | N.A. | |
Sodium Pyruvate (100 mM) | Thermo Fisher Scientific | 11360-070 | |
Spin-coater | Mikasa Co., Ltd., Japan | MS-A100 | |
Spininng confocal microscopy | Oxford Instruments | Andor Dragonfly Spinning Disk System | |
StemSure LIF, Mouse, recombinant, Solution (10^6U) | Fujifilm wako | 195-16053 | |
SU-8 3010 | Kayaku Advanced Materials, Inc., MA, USA | SU-8 3010 | |
SU-8 developer | Kayaku Advanced Materials, Inc., MA, USA | SU-8 developer | |
Tris-EDTA | Nippon Gene | 314-90021 | |
Vascular Endothelial Growth Factor-A165(VEGF), Human, recombinant | Fujifilm wako | 226-01781 |