Human iPSC-derived cardiac-specific extracellular matrix scaffolds for cardiomyocyte maturation and post-myocardial infarction repair

Fig. 1. Fabrication and characterization of ECM scaffolds. (A) Schematic representation illustrating cardiac ECM (hiPSC-CF-ECM, Pri-CF-ECM) and non-cardiac ECM (hDF-ECM) fabrication for engineering hiPSC-CMs containing cardiac patches. (B) Immunofluorescence staining reveals the anisotropic and nanofibrous architecture of collagen-I, laminin and fibronectin within ECM sheets following surface topographical cues. Verhoeff Van Gieson staining revealed the organization of elastin fibers within ECM sheets. Anisotropic nano-scale ECM bundles organized in ECM sheets observed by FE-SEM images. (C, D) ECM thickness measured by Z-stacking in confocal imaging. (E) The thickness of nanofibrous ECM bundles quantified using FE-SEM micrographs. Results displayed as mean ± standard deviation and were considered statistically significant for ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001.

Fig. 2. Characterization of major ECM structural components and embedded growth factors. Quantification of ECM-specific structural macromolecules including (A) insoluble collagen, (B) soluble collagen, (C) elastin, and (D) GAGs within ECM sheets post decellularization. Quantification of ECM embedded growth factors, including (E) VEGF, (F) IGF, (G) FGF, (H) endothelin I and (I) angiotensin II detected post decellularization. Results displayed as mean ± standard deviation and were considered statistically significant for ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001.

Fig. 3. Compositional evaluation of ECM scaffolds determined by mass spectrometry. (A) Heat map comparing the fold difference in the expression of cardiac proteins between cardiac and non-cardiac ECM scaffolds. Pie graph of the relative abundance of (B) fibrillar collagens, (C) non-fibrillar collagens and (D) matricellular proteins, respectively. Pie graph of the most abundant 20 proteins in cardiac (E) hiPSC-CF-ECM, (F) Pri-CF-ECM, and non-cardiac (G) hDF-ECM, based on the LFQ intensities of each protein to indicate their abundance within the scaffold. LFQ intensities of the rest of the proteins identified were binned under other (black).

Fig. 4. Organization of hiPSC-CMs on ECM scaffolds. Anisotropic and matured organization of (A) sarcomeric α-actinin (SAA, green) and F-actin (red) and (B) cardiac troponin T (cTnT, green) and connexin 43 (CX43, red)) (Scale bar: 50 μm).

Fig. 5. Evaluating maturation of hiPSC-CMs cultured on ECM constructs via bulk RNA sequencing. Differential gene volcano map indicating the number of up-regulated and down-regulated genes in hiPSC-CMs cultured on (A) hiPSC-CF-ECM Vs. collagen-I, (B) Pri-CF-ECM Vs. collagen-I, (C) hiPSC-CF-ECM Vs. Pri-CF-ECM, (D) hiPSC-CF-ECM Vs. hDF-ECM. (E) Hierarchical clustering of hiPSC-CMs cultured on hiPSC-CF-ECM, Pri-CF-ECM, hDF-ECM, and collagen-I based on mRNA expression obtained from RNA-seq. Heatmap indicating differential expression gene clustering from hiPSC-CMs cultured of various ECM constructs and collagen-I. (F) Gene ontology (GO) enrichment analysis reveals up/down regulation of cardiac-specific structural and functional genes that are activated in hiPSC-CMs cultured on hiPSC-CF-ECM versus Pri-CF-ECM. GO enrichment analysis scatter plot indicating expression of GO terms associated with cardiac-function in hiPSC-CMs cultured on (G) hiPSC-CF-ECM vs. Pri-CF-ECM, and (H) hiPSC-CF-ECM vs. hDF-ECM.

Fig. 6. Evaluating contractile properties of hiPSC-CMs. Digital image correction (DIC)-based analysis. (A)maximum principal strain variation and(B)beating cycle of hiPSC-CMs cultured on ECM scaffolds and their comparison with collagen-I-coated cover glass measured via DIC.(C)immunofluorescence staining-based sarcomere length measurement. hiPSC-CMs cultured on hiPSC-CF- ECM(D), Pri-CF-ECM(E), hDF-ECM(F), and(G)collagen-I-coated cover glass was stained with sarcomeric α−actinin (SAA) to measure the distance between the two SAA signals. Results displayed as mean ± standard deviation and were considered statistically significant for ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001.

Fig. 7. Optical action potentials recorded from hiPSC-CM monolayers cultured on different ECM scaffolds. (A–B) Representative action potentials shown are shown for monolayers seeded on PDMS with hiPSC-CF-ECM, Pri-CF-ECM, hDF-ECM and Collagen-I (glass coverslips) coatings with electrical pacing 2 Hz. (C) Summarized data for action potential duration at 80 % repolarization. (D) Spontaneous beating rate. hiPSC-CF-ECM n = 5, Pri-CF-ECM n = 4, hDF-ECM n = 4, Collagen-I n = 3. P values were determined by one-way ANOVA with multiple comparisons to the collagen-I group. Data represented as mean ± SD. ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.005.

Fig. 8. Influence of the nanofibrous ECM scaffolds on the electrical conduction properties of hiPSC-CM monolayers. (A–B) Representative voltage activation maps are shown for monolayers seeded on PDMS with hiPSC-CF-ECM, Pri-CF-ECM, hDF-ECM, and Collagen-I (glass coverslips) coatings. Activation maps were reconstructed during electrical pacing (1, 2, 3 Hz). Arrows show the directions of CV measurements (in cm/s), including longitudinal (aligned with PDMS chips pattern – CVL) and transversal (CVT). All the maps are plotted within the same time scale for comparison. (C–E) Summarized data for CVL, CVT, and anisotropy ratio (CVL/CVT) measured in all groups at different frequencies. hiPSC-CF-ECM n = 3∼5, Pri-CF-ECM n = 3∼4, hDF-ECM n = 2∼4, Collagen-I n = 3. P values were determined by one-way ANOVA with multiple comparisons to Collagen-I (C–D) and one sample t-test (E). Data represented as mean ± SD. ∗p < 0.05, ∗∗p < 0.01.

Fig. 9. Echocardiographic assessment of cardiac function in a rat myocardial infarction (MI) model following hiPSC-CF-ECM implantation. (A) Ejection fraction (EF) quantified from B-mode imaging in PSLAX; (B) Left ventricular internal diameter during systole (LVID; s) and (C) Left ventricular internal diameter during diastole (LVID; d), both assessed via PSLAX M-mode; (D) Global longitudinal strain (GLS) calculated using speckle-tracking echocardiography; (E) Isovolumic relaxation time (IVRT) and (F) mitral valve pressure half-time (PHT), both measured from pulsed-wave Doppler spectra. Data are presented as mean ± standard deviation; statistical significance (∗p < 0.05) was determined by one-way ANOVA with post hoc multiple comparisons.