Fig. 6: Ectopic PITX2c expression in human iPSC-derived PCs disrupts transcriptional state and function.

a UMAP representation of the single-cell transcriptomics of hiPSC-derived PCs transduced with AAV6-mCherry or AAV6-PITX2c alongside the distribution of PC and working myocardium-associated marker expression. PC C0 and C7 are outlined. b Heat-map showing the proportion of cells per cluster expressing PC and working myocardium-associated genes and the average expression of those genes per cluster. c The contribution of AAV6-mCherry (red) or AAV6-PITX2c (blue) PCs to each cluster. d Heat-map showing the distribution of EGFP expression in AAV6-mCherry and AAV6-PITX2c cells. PC C0 and C7 are outlined. e Scatterplot showing the relative expression of genes in the mouse wild-type HCN4+ SAN and delB/delB HCN4^low SAN compared to that of AAV6-PITX2c PC C0 and C7 and AAV6-mCherry C0 and C7 demonstrating that the most significantly deregulated PC genes are deregulated similarly in delB mice and AAV6-PITX2c hiPSC-PCs. Purple dots indicate genes that are significantly differentially expressed (p_adj < 0.05) in both comparisons. Wald test corrected for multiple comparisons using the Benjamini-Hochberg method. f Expression analysis of AAV9-PITX2c PC C0 (light blue) and C7 (dark blue) indicates a dichotomous response in AAV9-PITX2c hiPSC-PCs that correlates with PITX2c expression level. Wald test corrected for multiple comparisons using the Benjamini-Hochberg method. g Scatterplots showing the relative expression of genes in hiPSC-PCs and the mouse SAN, indicating that C0 (PITX2) is comparable to the delB/delB HCN4^high subdomain while C7 is comparable to the delB/delB HCN4^low subdomain. Key affected genes behave similarly in hiPSC-PCs and the mouse SAN. Purple dots indicate genes that are significantly differentially expressed (p_adj < 0.05) in both comparisons. Wald test corrected for multiple comparisons using the Benjamini-Hochberg method. h Typical examples of spontaneous activity (top panel) and the AP upstroke velocity (V_max; bottom panel) in AAV6-mCherry (red) and AAV6-PITX2c (blue) hiPSC-PCs. i While all measured AAV6-mCherry PCs (n = 9) showed spontaneous activity, only 50% of AAV6-PITX2c PCs (n = 12) are spontaneously active (two-tailed Fisher’s exact test, p = 0.0186). j Of the AAV6-PITX2c PCs that retained spontaneous activity (n = 6), PITX2 increased the coefficient of variation (CoV; SD/average) (two-tailed Mann-Whitney test, p = 0.0004) while differences in cycle length did not reach the threshold of significance. k Dot plot showing that the MDP is lower in AAV6-PITX2c PCs (n = 6) compared to AAV6-mCherry controls (n = 9) (two-tailed Mann-Whitney test, p = 0.0076). (l) V_max (p = 0.009) and AP duration at 20% (p = 0.0033), 50% (p = 0.0009) and 90% (p = 0.0013) repolarization following Kir2.1 injection under dynamic clamp conditions (two-tailed Mann-Whitney test). AP parameters were established for 9 AAV6-mCherry control and 12 AAV6-PITX2c PCs. Exact p and p_adj values and L2FC values are listed in Supplementary Data 1, 2, 11 and 12. Source data are provided in the Source Data file. PC pacemaker cardiomyocyte, SAN sinus node, C cluster, CM cardiomyocyte, L2FC Log₂Fold-change, AP action potential, MDP maximal diastolic potential.