Modelling arrhythmogenic cardiomyopathy fattyfibro pathology with PKP2-deficient epicardial cells derived from human iPSCs

Figures

• 

  Fig. 1. Generation and characterization of patient-derived and gene-edited PKP2-mutant iPSC lines. A Pedigrees and genotypes of ACM patient families 1025 and 398, showing inheritance of PKP2 mutations. Family 1025 carries a nonsense mutation (PKP2 c.1849C>T, p.Q617X), and Family 398 carries a frameshift mutation (PKP2 c.2013delC, p.P672fsX683). Squares represent males and circles represent females. Shaded symbols indicate individuals from whom iPSC lines were derived. Gene-corrected control lines were generated using CRISPR/Cas9 and are denoted by dashed lines. Predicted protein structures show the location and potential structural impact of the mutations. Sanger sequencing traces confirm the presence of the mutant allele (top) and successful correction (bottom) for both patient-derived lines (1025–100 and 398–110). B Schematic of the human PKP2 gene locus, including isoforms Pkp2a and Pkp2b. Arrows indicate transcriptional direction. The position of exon 2, which harbors the mutations used in this study, is highlighted in red. Generation of isogenic control, ACM, and PKP2KO human pluripotent stem cells (hPSCs).

• 

  Fig. 2. Epicardial cells generated from isogenic control, ACM, and PKP2KO human pluripotent stem cells (hPSCs). A Differentiation protocol used to generate hPSC-EPCs. B Phase-contrast images of hPSC differentiating into hPSC-EPCs from day 0 up to day 12 with epicardial cobblestone morphology (scale bar= 100 µm). C Phase contrast images of ACM, PKP2KO, and isogenic control hPSC-EPCs (scale bar= 100 μm). D Immunofluorescence images of WT1 in ACM, PKP2KO, and isogenic control hPSC-EPCs (scale bar= 100 μm). E qPCR of the epicardial marker WT1 in ACM, PKP2KO, and isogenic control hPSC-EPCs relative to undifferentiated hPSCs (mean ± SEM, N = 6 of hPSCs, hPSC-EPCs and ACM/PKP2KO hPSC-EPCs). Statistical significance was evaluated using a one-way ANOVA test with Bonferroni correction. F qPCR of the epicardial marker TBX18 in ACM, PKP2KO, and isogenic control hPSC-EPCs relative to undifferentiated hPSCs (mean ± SEM, N = 4 of hPSCs, hPSC-EPCs and ACM/PKP2KO hPSC-EPCs). Statistical significance was evaluated using a one-way ANOVA test with Bonferroni correction. G Immunofluorescence images of PKP2 in ACM, PKP2KO, and isogenic control hPSC-EPCs (scale bar= 50 μm). H Representative western blot image…

• 

  Fig. 3. ACM and PKP2KO hPSC-EPCs demonstrate enhanced EMT properties compared to isogenic control hPSC-EPCs. A qPCR of the EMT marker CDH1 (E-cadherin) and CDH2 (N-cadherin) with or without TGFβ treatment relative to isogenic control hPSC-EPCs, ACM hiPSC-EPCs, and PKP2KO hESC-EPCs (mean ± SEM, N = 3). Statistical significance was evaluated using a one-way ANOVA with Bonferroni correction. B Representative images of 48-hour cell migration assay, showing the start of the migration of hiPSC-EPCs and ACM hiPSC-EPCs or C hESC-EPCs and PKP2KO hESC-EPCs from the introduced gap at hour 0 and the final migration results of hiPSC-EPCs and ACM hiPSC-EPCs at 48 hour (scale bar = 500 μm). D Gap closure quantification of ACM hiPSC-EPCs relative to control (hiPSC-EPCs) over 48 hours post introduction of the gap (mean ± SEM, N = 10 of hiPSC-EPCs and ACM hiPSC-EPCs with ** at 24 hour (p = 0.0071), with *** at 48 hour (p = 0.0072)). Statistical significance was evaluated using a two-way ANOVA test with Bonferroni correction. E Quantification gap closure of PKP2KO hESC-EPCs relative to control (hESC-EPCs) over 48 hours post introduction of the gap (mean ± SEM, N = 5 of hESC-EPCs and PKP2KO hESC-EPCs with *…

• 

  Fig. 4. ACM and PKP2KO hPSC-EPCs displayed increased lipid accumulation compared to isogenic control hPSC-EPCs. A Representative fluorescence images of Nile Red staining of lipids in ACM hiPSC-EPCs and hiPSC-EPCs or B hESC-EPCs and hESC-EPCs during adipogenic induction at days 0, 4, and 8 (scale bar = 100 μm). C Quantification of Nile Red fluorescent intensity per cell in each field view of ACM hiPSC-EPCs and hiPSC-EPCs relative to the control (hiPSC-EPCs at day 0) at days 0, 4, and 8 (mean ± SEM, N = 10 of hiPSC-EPCs and ACM hiPSC-EPCs). Quantification of Nile Red fluorescent intensity per cell in each field view of PKP2KO hESC-EPCs and hESC-EPCs relative to the control (hESC-EPCs at day 0) at days 0, 4, and 8 (mean ± SEM, N = 6 of hESC-EPCs and PKP2KO hESC-EPCs). Statistical significance was evaluated using a one-way ANOVA test with Bonferroni correction. D qPCR of CEBPA in ACM hiPSC-EPCs and hiPSC-EPCs with or without adipogenic induction treatment relative to untreated hiPSC-EPCs (mean ± SEM, N = 6 of hiPSC-EPCs and ACM hiPSC-EPCs). qPCR of CEBPA in PKP2KO hESC-EPCs and hESC-EPCs with or without adipogenic induction treatment relative to untreated hESC-EPCs (mean ± SEM, N = 8 of…

• 

  Fig. 5. A more fibrotic phenotype is observed in fibroblast derived from hPSC-EPCs when compared to fibroblasts which did not go through an epicardial cell state. A qPCR of fibrotic markers Col1α1 (mean ± SEM, N = 3) and B PDGFRA in hiPSC-EPCs and ACM hiPSC-EPCs (mean ± SEM, N = 4), and in hESC-EPCs and PKP2KO hESC-EPCs (mean ± SEM, N = 3). Statistical significance was evaluated using a one sample t test. C Phase contrast images of non-epicardial derived fibroblasts (ACM, PKP2KO, and isogenic control hPSC-CFs) and D fibroblasts derived from epicardial cells (ACM, PKP2KO, and isogenic control hPSC-EpiCFs). E qPCR of Col1α1 and F PDGFRA in hiPSC-CFs, ACM hiPSC-CFs, hiPSC-EpiCFs, ACM hiPSC-EpiCFs as well as in hESC-CFs, PKP2KO hESC-CFs, hESC-EpiCFs, and PKP2KO hESC-EpiCFs (mean ± SEM, N = 6). Statistical significance was evaluated using a one-way ANOVA test with Bonferroni correction. G, H Representative fluorescence images of Nile red staining of lipids in ACM, PKP2KO, and isogenic control hPSC-EpiCFs and hPSC-CFs during adipogenic induction at days 0 and 8 (scale bar = 100 μm). I Quantification of Nile red fluorescent intensity per cell in ACM hiPSC-CFs and hiPSC-CFs relative to control…

• 

  Fig. 6. ACM hiPSC-EPCs differentially expressed genes related to the Rho-GTPAse, immune (interleukins, interferons), metabolism, and Wnt signaling pathways. A Heatmap representing log2 fold changes in 1202 differentially expressed genes (p < 0.05) between ACM and isogenic control hiPSC-EPCs (N = 5, transcription factors and growth factors annotated on right panel). B Reactome pathways called using differentially expressed genes (Rho-GTPase, interleukins, and IGF signaling terms highlighted in red). C Motif sequences for TEAD2/4, FOXD3/FOXA1, and progesterone receptor (PR) were enriched within 2 kb of the differentially expressed gene transcriptional start sites. D STRING network analysis highlighting the interaction between members of the Wnt Pathway (WNT5B, WNT16, FZD1), IGF (IGF2), interleukin (IL1B), interferon, and metabolic genes.

• 

  Fig. 7. IGF2 signaling promotes adipogenic gene expression and lipid accumulation in hPSC-derived epicardial cells, while PKP2KO cells exhibit reduced sensitivity to IGF-1R inhibition. A qPCR analysis of human heart tissue shows elevated expression of IGF2 (p = 0.0526, N = 4), PPARG (p = 0.0827, N = 3), and CEBPA (p = 0.0972, N = 3) in ACM patients compared to control hearts. B Phase-contrast images of control and PKP2KO hPSC-derived epicardial cells (hPSC-EPCs) after six days of adipogenic induction, with or without IGF2 treatment. IGF2 induces morphological changes consistent with adipogenic differentiation. C qPCR analysis of CEBPA and PPARG expression in control (gray) and PKP2KO (purple) hPSC-EPCs treated with IGF2. In control hPSC-EPCs, IGF2 treatment significantly increases CEBPA (p = 0.0082, N = 6) and PPARG (p = 0.0186, N = 5). In PKP2KO hPSC-EPCs, IGF2 also elevates CEBPA (p = 0.0112, N = 6) and PPARG (p = 0.0.0681, N = 5). D Nile Red staining reveals lipid accumulation after 8 days of adipogenic induction with or without IGF-1R inhibitor GSK1904529A. Lipid content is visibly reduced in control hPSC-EPCs following inhibitor treatment, with no obvious change in PKP2KO hPSC-EPCs.…