Titin-truncating variants in hiPSC cardiomyocytes induce pathogenic proteinopathy and sarcomere defects with preserved core contractile machinery

Figures

• 

  Figure 1. Generation of hiPSC-CMs carrying titin-truncating variants (A) Schematic representation of one cardiac sarcomere unit with major components. (B) Sanger sequencing confirming the generation of titin A-band truncation hiPSCs. (C) Schematic workflow of hiPSC cardiac differentiation. (D) CM differentiation in WT and TTNtv mutant hiPSCs: ventricular (MLC2v), atrial (COUP-TFII), pan-CM (cTnT), nuclei (Hoechst). Scale bars, 100 μm. (E) Contraction profiles of WT and TTNtv mutant hiPSC-CMs by 2D motion tracking analysis. See quantification results in Figure S1I. (F) Representative titin immunofluorescence images of WT, tv/+, and tv/tv hiPSC-CMs: Z-line specific (red); M-line specific (green); Hoechst (blue). Scale bars, 20 μm. Panels on the right (I, II, and III) show zoom-in views of boxed areas. Scale bars, 5 μm.

• 

  Figure 2. Transcriptomic profiling reveals DCM disease signature of TTN tv/+ hiPSC-CMs (A) PCA analysis of WT and TTN tv/+ hiPSC-CMs with different culture time (days 30, 90, and 150). For each time point, samples from three independent differentiation batches were sequenced. (B) Venn diagram showing unique and shared DEGs (FDR < 0.05) in time-point-paired analysis. See also Figure S2A. (C) GO biological process terms enriched in hierarchically clustered gene subsets that changed consistently in WT and TTN tv/+ hiPSC-CMs with extended in vitro culture time. (D) Heatmap showing expression patterns of DEGs (FDR < 0.001) between WT and TTN tv/+ hiPSCs-CMs on three time points combined. Color scale represents Z score. (E and F) GO biological process terms (E) and IPA metabolic and tox pathway terms (F) enriched in gene clusters from (D). (G) Heatmap showing expression patterns of selected genes related to key functional aspects of cardiomyocyte physiology on two later time points (day 90 and day 150). Color scale represents Z score. (H) STRING database analysis revealing protein-protein interaction between titin and selected dysregulated genes from (G). Blue circles, sarcomere components; red…

• 

  Figure 3. TTNtv accumulates in cytoplasm for autophagy-lysosomal degradation (A) Titin total transcript count in WT and TTN tv/+ hiPSC-CMs. Data are shown as mean ± SD. (B) Allele frequency of titin transcripts determined by all heterozygous SNPs detected in RNA-seq. (C) SNP coverage by log-transformed reads up- and downstream of TTNtv stop codon. (D) Titin isoform (N2BA and N2B) transcript count in WT and TTN tv/+ hiPSC-CMs. Isoform-specific exon-exon junction reads were measured: N2B (exons 50–220), N2BA (exons 50–51). (E) Protein-stain and titin western blot with Z- and M-specific titin antibodies. (F) Gene-targeting strategy to generate TTNtv-mCherry/+ reporter hiPSCs. See also Figures S3C–S3F. (G) Pie chart showing different sizes of TTNtv-mCherry fusion protein accumulation observed in hiPSC-CM cytoplasm from N = 3 independent differentiation experiments. Size cutoff: negative (<10 pixels2), puncta (10–100 pixels2), aggregate (>100 pixels2). 1 pixel2 = 0.036 μm2. (H) Co-staining of TTNtv-mCherry spots with lysosome (LysoTracker and LAMP-1), autophagosome (LC3), and 26S proteasome (PSMD3) markers. Scale bars, 20 μm. (I) Accumulation of cytoplasmic TTNtv-mCherry spots under…

• 

  Figure 4. TTN tv/+ hiPSC-CMs exhibit normal calcium cycling but aberrant sodium channel activities (A) Representative Ca2+ transients in WT and TTN tv/+ hiPSC-CMs. See Figure S4A for quantification of output parameters. (B) Representative action potential (AP) traces in WT and TTN tv/+ hiPSC-CMs (paced at 1.5 Hz) on the multi-electrode array (MEA) platform. (C and D) Action potential and field potential output parameters. Additional parameters are shown in Figure S4D. Data are shown as mean ± SD. Each data point represents readings from one well of MEA 96-well plates in N = 3 independent experiments. ∗∗∗∗p < 0.0001; ∗∗∗p < 0.001; n.s., not significant; one-way ANOVA with Bonferroni’s multiple comparison test. (E and F) Pharmacological responses of WT and TTN tv/+ hiPSC-CMs. Changes in Na+ depolarization spike amplitude in field potential recordings are reported. Values are normalized to baseline readings and presented as mean ± SEM. Data were collected from N = 3 independent experiments.

• 

  Figure 5. TTN tv/+ hiPSC-CMs display sarcomere disorganization and deficiency at single-cell level (A) Schematic workflow of automated sarcomere organization analysis (SOA). (B) Representative Haralick plots and SOA analysis outputs of hiPSC-CM sarcomeres. Categorization of sarcomere organization is based on the combinations of sarcomere length, organization score, and sarcomere alignment. Immunofluorescence markers: α-actinin (green), cTnT (red), and Hoechst (blue). (C) Scatterplots of single-cell SOA analysis of WT, TTN tv/+ (clone 1 and clone 2 combined) and TTN tv/tv hiPSC-CMs. Cells were gated on quadrants of organization score and sarcomere length and categorized as exemplified in (B): organized (top left); intermediate (bottom left); and disorganized (bottom right). Percentage values of cells in each quadrant are shown on the plots. n represents the total number of cells analyzed in three independent differentiation experiments. (D) Histogram showing the distribution of sarcomere alignment (0°–90°) in WT and TTN tv/+ hiPSC-CMs. (E and F) Histogram comparisons of sarcomere density (E) and functional sarcomere coverage (F) between WT and TTN tv/+ hiPSC-CMs. (G) Positive correlation…

• 

  Figure 6. TTN tv/+ CMWs show contractile defects but preserve an intact core contractile machinery for sarcomere modulator therapy (A) Illustration of the 3D cardiac micro-wire (CMW) platform for contractility assay. (B) Immunofluorescence images of CMW: hiPSC-CMs (α-actinin, red); cardiac fibroblasts (vimentin, green); nuclei (DAPI, blue). Bottom panel shows zoom-in view. Scale bars, 2 mm (top) and 200 μm (bottom). (C–E) Contractile measurements of WT and TTN tv/+ CMWs: (C) representative raw traces (left) and contractile force (right), (D) contraction (left) and relaxation (right) velocity, and (E) contraction (left) and relaxation (right) time. Data are shown as scatterplots with bar graphs representing mean ± SD. Each data point represents measurements of one CMW from N = 3 independent batches. ∗∗∗∗p < 0.0001; ∗∗p < 0.01; n.s., not significant; one-way ANOVA with Bonferroni’s multiple comparison test. (F) Comparison of contractile kinetics between WT and TTN tv/+ CMWs using contraction velocity and relaxation velocity. The coefficient of determination R2 and linear regression slope are reported. (G) The cardiac core contractile machinery and mechanisms of action of sarcomere…

• 

  Figure 7. Sarcomere modulator-treated TTNtv CMWs indicated reversal of DCM disease signatures (A) UMAP projection of CMW cell clusters identified by scRNA-seq (n = 29,515) from two independent differentiation batches. (B) Violin plot showing normalized expression of representative signature genes of cell clusters. (C) List of representative cardiometabolic disease-associated MeSH terms enriched in TTNtv CMW DEGs (CM clusters). The size of the dot reflects the GeneRatio and the color of the dot indicates increasing significance of the enriched pathways. Top 20 significantly enriched terms (p < 0.05) are presented. (D) Expression patterns of top 50 informative genes in non-dividing cardiomyocyte clusters (CM1–4). (E) Bar graph of diffusion map pseudotime component 3 distribution across clusters CM1–4. (F) UMAP projection of clusters CM1–4 identified by scRNA-seq (n = 15,802). Arrow indicates the direction of cardiomyocyte functional maturity by pseudotime analysis. (G) Relative cell composition of clusters CM1–4 across genotypes and treatment groups. (H) Heatmap showing expression of representative genes decreased in TTNtv CMWs but restored under sarcomere modulator treatment. (I and J)…