Self-assembled chamber-like cardiac organoids for modeling cardiac chamber formation and cardiotoxicity assessment

Figures

• 

  Fig. 1. Stable chambered structure of cardiac organoids (CCOs) derived from hPSCs. a Protocol and time-lapse of unstable chamber-like cardiac organoids formation captured in live brightfield images using Hofbauer’s protocol9. Scale bar, 500 μm. Middle: The key molecules used at each step are listed beneath the arrows. Molecules and their concentrations unique to Hofbauer’s protocol, compared to ours, are highlighted in different colors. Bottom: Bright-field time-lapse images captured throughout the process. b Protocol and time-lapse of CCOs formation captured in images using our protocol. Top: Heatmap of key genes corresponding to each developmental stage from time-course bulk RNA sequencing. Middle: Schematic of the CCOs formation protocol, highlighting six key developmental stages: (i) pluripotency, (ii) mesoderm, (iii) cardiac mesoderm, (iv) cardiomyocyte formation, (v) chamber formation, and (vi) chamber maintenance. The key molecules used at each step are listed beneath the arrows. Molecules and their concentrations unique to our protocol compared to Hofbauer’s are highlighted in different colors. Bottom: Bright-field time-lapse images captured throughout the process. VST,…

• 

  Fig. 2. Inner endothelium and outer myocardium layers and markers in the CCOs. a Expression levels of endothelial-specific markers during CCOs formation from bulk sequencing data. All bar graphs show mean ± SD (n = 2). FPKM, Fragments Per Kilobase of transcript per Million mapped reads. b Representative immunofluorescence images illustrating the architecture of outer myocardial and inner endothelial layers in CCOs on day 17.5. Scale bar, 200 μm. c Representative 3D reconstruction of a CCO (D21.5) acquired with Zeiss Lightsheet 7 microscopy. Endothelial cells are visualized by CDH5-Scarlet fluorescence (red, CDH5-P2A-Scarlet H9 line), and cardiomyocytes by TNNT2 immunostaining (green). Scale bar, 200 μm. d Heatmap displaying gene expression of markers associated with chamber formation and heart looping during CCOs development from day 0 to day 25. e Representative brightfield images of CCOs treated with or without 0.5 μM SB431542 for 20 days, captured on day 48.5, alongside brightfield images of ETS1 knockout (KO) cardiac organoids captured on day 28.5. Scale bar, 500 μm. f qPCR analysis showing reduced expression of TWIST1 and SNAI1 in CCOs following SB431542 treatment, suggesting…

• 

  Fig. 3. Analysis of cell diversity and communication in cultures on day 25.5. a Cell type annotation: UMAP visualization of cell types derived from single-cell RNA sequencing data on day 25.5, with distinct colors representing different cell types. b Marker gene expression: UMAP scatter heatmap showing marker gene expression across cell types, with color gradients indicating expression levels. Annotation of cell types was performed based on marker gene expression in clusters. c Cell communication: The top panel depicts signaling pathways where the endothelium acts as the sender, influencing other cell types (e.g., fibroblasts, atrial cardiomyocytes, and ventricular cardiomyocytes). The bottom panel shows pathways where the endothelium is the receiver. Arcs represent individual pathways, with direction indicating sender and receiver relationships. Arc colors denote different signaling pathways (e.g., BMP, WNT, NOTCH). d Heatmap of interaction probabilities: Displays the probabilities of key signaling pathway interactions among different cell types aCM: Atrial Cardiomyocytes; Endo: Endothelial Cells; vCM: Ventricular Cardiomyocytes; Fibro: Fibroblasts).

• 

  Fig. 4. The roles of activin A and retinoic acid in the chamber formation and stability of CCOs. a Time-lapse brightfield images showing the development of CCOs under different concentrations of AA and RA. A4R50: AA 4 nM and RA 50 nM, A4R500: AA 4 nM and RA 500 nM, A50R500: AA 50 nM and RA 500 nM. Scale bar, 500 μm. b Representative brightfield images of cardiac organoids induced with varying concentrations of AA and RA on day 25.5. Scale bar, 500 μm. c Representative brightfield images of cardiac organoids induced with varying concentrations of AA on day 8.5. Scale bar, 500 μm. d Measurement of chamber areas in cardiac organoids on days 15.5 and 25.5 after treatment with different concentrations of AA and RA. All bar graphs show mean ± SEM (N = 3, n = 95). Statistical significance was determined by two-tailed unpaired t test (for time-course comparisons) or one-way ANOVA with Tukey’s test (for multi-group comparisons). ****p < 0.0001. Source data are provided as a Source Data file. All images are representative of at least 3 independent experiments with similar results.

• 

  Fig. 5. TFAP2B, NR2F2 and HAND1 are essential for the formation of CCOs. a Venn diagram illustrating upregulated or downregulated transcription factor genes in cardiac organoids on different days, without PD173074 as control. b Bar chart of log2 fold change of 13 selected transcription factors in cardiac organoids without PD173074 as the control on day 3.5. c Expression levels of the selected transcription factors on day 3.5 of cardioids. FPKM, Fragments Per Kilobase of transcript per Million mapped reads. d TFAP2B, NR2F1 and NR2F2 gene expression analysis (log2 fold-change vs. day 1.5) with or without PD173074 in cardioids. The gene expression data of TFAP2B, NR2F2, and HAND1 shown in panel d are also presented in Fig. S8A as part of 13 selected genes, providing a more complete view of gene expression changes during CCO formation with or without PD173074 treatment. e Confocal cryosection images of TFAP2B and NR2F2 KO cardiac organoids stained with TNNT2. Scale bar, 200 μm. f Representative brightfield images of TFAP2B, NR2F1, and NR2F2 KO cardiac organoids from day 15.5 to day 30. Scale bar, 500 μm. g Beating rate of WT control, NR2F1, and NR2F2 KO cardiac organoids from days 15.5 to…

• 

  Fig. 6. Development of a CCOs-based platform for cardiotoxicity screening. a Schematic representation of CCOs treatment with isoproterenol (ISO) or doxorubicin (DOX). This schematic was created by the authors for this study. b Representative heatmaps showing the averaged motion magnitude at 0 and 1 s. The vector field overlay with red arrows indicates motion velocity. Scale bar, 170 μm. c Beats per minute of CCOs treated with drugs. Pretreatment with propranolol (2.5 μM, 30 min) reduced spontaneous beating rate, and subsequent ISO stimulation did not restore the beating frequency. Mean ± SEM. Two-tailed unpaired t test (DMSO, N = 3, n = 10, p = 0.6783; ISO, N = 3, n = 7, ***p = 6.94×10−4) or one-way ANOVA with Tukey’s test (PROP + ISO, N = 3, n = 11). d Average displacement of contractile activity in CCOs over time, before and after treatment with 1 μM isoproterenol. e Representative brightfield images of CCOs treated with 5 μM doxorubicin (DOX) or daunorubicin (DNR) after 24 and 48 h. Scale bar, 500 μm. f Quantification of CCOs chamber areas after treatment with 5 μM DOX or DNR for 24 and 48 h. Mean ± SEM, one-way ANOVA with Tukey’s test (DMSO, N = 3, n = 14; DOX, N = 3, n = 9; DNR, N =…

• 

  Fig. 7. Establishment of a CCOs-based platform for teratogenicity screening. a Live brightfield images of CCOs treated with different drugs. All drugs were added into the CCOs from day 0. Aspirin, 30 μM; thalidomide, 10 μg/ml; acitretin, 50 nM. Scale bar, 500 μm. b Quantification of the chamber areas of CCOs treated with or without drugs. Mean ± SEM, one-way ANOVA with Tukey’s test (N = 3, n = 91). c Beats per minute of CCOs treated with or without drugs. Mean ± SEM, one-way ANOVA with Tukey’s test (N = 3, n = 65). d Enlarged brightfield images of CCOs treated with different drugs on day 1.5 and day 3.5. Enlarged brightfield images of CCOs treated with different drugs on day 1.5 and day 3.5. Images are magnified views of corresponding time points shown in (a). Scale bar, 500 μm (full image), 125 μm (magnified area). e Live brightfield images of CCOs captured on day 36.5, treated with thalidomide starting from day 15.5. Scale bar, 500 μm. For one-way ANOVA analysis. Source data are provided as a Source Data file. ****p < 0.0001. All images are representative of at least 3 independent experiments with similar results.