ERRγ agonist under mechanical stretching manifests hypertrophic cardiomyopathy phenotypes of engineered cardiac tissue through maturation

Abstract

Engineered cardiac tissue (ECT) using human induced pluripotent stem cell-derived cardiomyocytes is a promising tool for modeling heart disease. However, tissue immaturity makes robust disease modeling difficult. Here, we established a method for modeling hypertrophic cardiomyopathy (HCM) malignant (MYH7 R719Q) and nonmalignant (MYBPC3 G115^∗) pathogenic sarcomere gene mutations by accelerating ECT maturation using an ERRγ agonist, T112, and mechanical stretching. ECTs treated with T112 under 10% elongation stimulation exhibited more organized and mature characteristics. Whereas matured ECTs with the MYH7 R719Q mutation showed broad HCM phenotypes, including hypertrophy, hypercontraction, diastolic dysfunction, myofibril misalignment, fibrotic change, and glycolytic activation, matured MYBPC3 G115^∗ ECTs displayed limited phenotypes, which were primarily observed only under our new maturation protocol (i.e., hypertrophy). Altogether, ERRγ activation combined with mechanical stimulation enhanced ECT maturation, leading to a more accurate manifestation of HCM phenotypes, including non-cardiomyocyte activation, consistent with clinical observations.

Keywords: Disease modeling; Engineered cardiac tissue; Estrogen related receptor gamma; Fibrosis; Hypertrophic cardiomyopathy; Maturation; Mechanical stress; Sarcomere gene mutation; Stem cell-derived cardiomyocytes.

Graphical abstract

Figure 1

Functional characterization of mature ECTs (A) Representative tissue image and schematic of mechanical stretching. (B) Phase contrast (upper), EmGFP fluorescence (TNNI1, middle), and mCherry fluorescence (TNNI3, lower) images of day-15 ECTs treated with DMSO or 3 μM T112, in static condition or with mechanical stretching (10% elongation at 1 Hz) (DMSO-static, T112-static, DMSO-mech, and T112-mech). Scale bar, 500 μm. See also Figures S1A and S1B. (C) Quantification of mCherry expression of day-15 ECTs with DMSO or 3 μM T112, in static condition or with mechanical stretching (2.5% or 10% elongation at 1 Hz) from three independent experiments (n = 4–8). See also Figures S1C, S1D, S2A, and S2B. (D) Representative immunostaining images of day-15 DMSO-static, T112-static, DMSO-mech, or T112-mech ECTs for wheat germ agglutinin (WGA) (red), ACTN2 (green), and DNA (blue). Scale bar, 50 μm. See also Figure S1E. (E) Quantification of cardiomyocyte area of day-15 DMSO-static, T112-static, DMSO-mech, and T112-mech ECTs from three independent experiments (n = 110–164). (F) Quantification of sarcomere alignment of day-15 DMSO-static, T112-static, DMSO-mech, and T112-mech ECTs from five independent experiments (n = 5–9). (G) Quantification of sarcomere length of day-15 DMSO-static, T112-static, DMSO-mech, or T112-mech ECTs from three independent experiments (n = 90–120). (H) Quantification of contractile force in DMSO-static, T112-static, DMSO-mech, or T112-mech ECTs from four independent experiments (n = 9–12). (I) Quantification of mitochondrial DNA content of day-15 DMSO-static, T112-static, DMSO-mech, or T112-mech ECTs from nine independent experiments (n = 15–18). Data are shown relative to the DMSO-static group in each experimental batch. (J) Quantification of glucose consumption of DMSO-static, T112-static, DMSO-mech, or T112-mech ECTs from six independent experiments (n = 6–12). Data presented as the mean ± SD (C, F, H, and J) or in a boxplot (E, G, and I). All statistical analyses were performed using two-way ANOVA with Tukey’s multiple comparison test.

Figure 2

RNA-seq analysis of ECTs (A) PCA of RNA-seq of day-15 DMSO-static, T112-static, DMSO-mech, or T112-mech ECTs from three independent experiments (n = 3). (B) Gene Ontology (GO) cell component terms enriched in upregulated differentially expressed genes (DEGs) compared with DMSO-static (log2 FC > 0, q < 0.05). The color indicates the p value. Circle size indicates the ratio of the number of genes containing each GO term to the number of DEGs. See also Figures S3A and S3B and Data S1 and S2. (C) Heatmap showing the expression of selected genes. The color bar indicates log2 FC compared to the DMSO-static group. Asterisk (^∗) indicates significant expression compared to the DMSO-static group (q value <0.05). (D) Heatmap showing expression of selected genes. The color bar indicates the Z score of the normalized count. Asterisk (^∗) indicates significant expression compared to the T112-static group (q value <0.05). See also Figure S6D.

Figure 3

Phenotype characterization of MYH7 R719Q ECTs (A) Representative immunofluorescence images of day-15 WT or MYH7 R719Q ECTs with DMSO-static, T112-static, DMSO-mech, or T112-mech treatment for WGA (red), ACTN2 (green), and DNA (blue). Scale bar, 50 μm. See also Figures S4A–S4D. (B) Quantification of sarcomere alignment of day-15 WT or MYH7 R719Q ECTs with DMSO-static, T112-static, DMSO-mech, or T112-mech treatment from five independent experiments (n = 5–6). (C) Representative immunofluorescence images of day-15 WT or MYH7 R719Q ECTs with T112-mech treatment for ACTN2 (green) and DNA (blue). Scale bar, 200 μm. (D) Quantification of cardiomyocyte area of day-15 WT or MYH7 R719Q ECTs with DMSO-static, T112-static, DMSO-mech, or T112-mech treatment from four independent experiments (n = 100–154). (E and F) Quantification of contractile force (E) and 90% relaxation time (F) in WT or MYH7 R719Q ECTs with DMSO-static, T112-static, DMSO-mech, or T112-mech treatment from three to five independent experiments (n = 3–6). See also Figure S4E. Data are presented as the mean ± SD (B, E, and F) or in a boxplot (D). All statistical analyses were performed using two-way ANOVA with Tukey’s multiple comparison test.

Figure 4

Phenotype characterization of MYBPC3 G115^∗ ECTs (A) Representative western blotting images of day-29 WT or MYBPC3 G115^∗ EB lysate using anti-MYBPC3 (upper) and anti-β-actin antibodies (lower). The molecular-weight ladder is shown on the left side. See also Figures S4D and S5A–S5D. (B) Quantification of MYBPC3 expression normalized by β-actin expression from three independent experiments (n = 3). (C) Representative immunofluorescence images of day-15 WT or MYBPC3 G115^∗ ECTs with T112-mech treatment for WGA (red), ACTN2 (green), and DNA (blue). Scale bar, 50 μm. (D) Quantification of cardiomyocyte area of day-15 WT and MYBPC3 G115^∗ ECTs with T112-mech treatment from three independent experiments (n = 70–88). See also Figure S5F. (E) Quantification of sarcomere alignment on day-15 WT and MYBPC3 G115^∗ ECTs with T112-mech treatment from three independent experiments (n = 3–4). (F and G) Quantification of contractile force (F) and 90% relaxation time (G) in WT or MYBPC3 G115^∗ ECTs with T112-mech treatment from four independent experiments (n = 5–10). See also Figure S5E. Data are presented as the mean ± SD (B and E–G) or in a boxplot (D). All statistical analyses were performed using unpaired t tests.

Figure 5

RNA-seq analysis of HCM model using isogenic iPSCs (A) PCA of RNA-seq data of day-15 WT, MYH7 R719Q, or MYBPC3 G115^∗ ECTs with T112-mech treatment from three independent experiments (n = 3). (B) GO biological process terms enriched in upregulated DEGs compared to WT (log2 FC > 0, q < 0.05). The color indicates the p value. Circle size indicates the ratio of the number of genes containing each GO term to the number of DEGs. See also Data S3. (C) Heatmap showing the expression of selected genes. The color bar indicates log2 FC compared to WT. Asterisk (^∗) indicates significant expression change compared to WT (q value <0.05). (D) Quantification of glucose consumption of WT, MYH7 R719Q, or MYBPC3 G115^∗ ECTs cultured for 4 days under static conditions after T112 treatment and mechanical stimulation from three independent experiments (n = 5–6). (E) Representative immunostaining images of day-15 WT, MYH7 R719Q, or MYBPC3 G115^∗ ECTs with T112-mech treatment for fibronectin (green), cardiac troponin T (red), and DNA (blue). Scale bar, 50 μm. (F) Quantification of fibronectin-positive area of day-15 WT, MYH7 R719Q, or MYBPC3 G115^∗ ECTs with T112-mech treatment from three independent experiments (n = 3). See also Figure S5G. Data are presented as the mean ± SD (D and F). Statistical analyses were performed using one-way ANOVA with Dunnett’s multiple comparison test (D and F).

Figure 6

Robustness of T112 treatment and mechanical stretching (A) Representative immunostaining images of day-15 ECTs with DMSO-static or T112-mech treatment for WGA (red), ACTN2 (green), and DNA (blue). Scale bar, 50 μm. (B) Quantification of cardiomyocyte area of day-15 DMSO-static or T112-mech ECTs from three independent experiments (n = 105–287). (C) Quantification of sarcomere length of day-15 of ECTs with DMSO-static or T112-mech treatment from three independent experiments (n = 72). (D) Quantification of sarcomere alignment of day-15 DMSO-static or T112-mech ECTs from three independent experiments (n = 4–5). (E) Quantification of contractile force in DMSO-static or T112-mech ECTs from four independent experiments (n = 9–10). (F) Quantification of sarcomere alignment of day-15 WT or MYH7 R719Q ECTs with DMSO-static or T112-mech treatment from four independent experiments (n = 5–7). See also Figures S6A and S6C. (G) Quantification of cardiomyocyte area of day-15 WT or MYBPC3 G115^∗ ECTs with DMSO-static or T112-mech treatment from three to four independent experiments (n = 95–159). See also Figures S6B and S6C. (H) Quantification of fibronectin-positive area of day-15 WT or MYBPC3 G115^∗ ECTs with DMSO-static or T112-mech treatment from three independent experiments (n = 3–4). (I) Representative immunostaining images of day-15 WT or MYBPC3 G115^∗ ECTs with T112-mech treatment for fibronectin (green), cardiac troponin T (red), and DNA (blue). Scale bar, 50 μm. Data are presented in a boxplot (B, C, and G) or the mean ± SD (D–F and I). Statistical analyses were performed using unpaired t tests (B–E) or two-way ANOVA with Šídák’s multiple comparison test (F and I) or Tukey’s multiple comparison test (G).