A titin truncating variant linked to atrial fibrillation increases atrial profibrotic signalling and cholinergic sensitivity

Figures

• 

  Figure 1. Differential sarcomere integrity and TTN gene expression in WT, het and hom atrial hiPSC-CMs. (A) hiPSC-CMs were stained for α-actinin (red), Titin (z-disc region, green) and DAPI (nuclei, blue). Scale bar = 20 μm. (B) RT-PCR was performed using primers that annealed at exons 303 and 304. The mutant TTN product can be distinguished from the WT allele (552 bp) by its larger amplicon (637 bp) (see Supplementary material online, Figure S6). (C) WT and Het atrial hiPSC-CMs were analysed for expression of TTN by qPCR using primer pairs that bind within different exons. Primers designed to amplify Exon 11–12 were specific for the N2-BA isoform of TTN. The primers designed to amplify Exon 49–50 were specific for both N2-B and N2-BA isoforms. N2-BA/N2-B+N2-BA correlates with the proportion of compliant TTN isoforms (N = 4 Differentiation Batches, n = 12 Technical Replicates). A Welch’s t test was used to ascertain significance (*denotes P < 0.05).

• 

  Figure 2. Bulk RNA-Seq of WT and het atrial hiPSC-CMs. (A) Volcano Plot of WT and Het hiPSC-CM Gene Expression. The x-axis represents the log2 fold change in gene expression between Het and WT hiPSC-CMs, while the y-axis displays the -log10 P-value for each gene. Genes with a log2 fold change > 1.5 and a P-value below 0.05 (denoted by dashed lines) are highlighted in either red (up-regulated in Het hiPSC-CMs) or black (down-regulated in Het hiPSC-CMs). (B) MA plot (average expression vs. log fold change) in WT and Het hiPSC-CMs. Genes with a log2 fold change > 1.5 and an average expression > 2 (denoted by dashed lines) are highlighted in either red (up-regulated in Het hiPSC-CMs) or black (down-regulated in Het hiPSC-CMs). (C) Heatmap of sample gene expression for 35 differentially expressed genes in WT and Het hiPSC-CMs. (D) Gene pathways which are up and down-regulated in Het hiPSC-CMs. Pathways are listed along the x-axis, while the y-axis represents the normalized enrichment score (NES), indicating the degree of enrichment for each pathway. Bubble size corresponds to the number of genes involved in each pathway, and the colour gradient represents the adjusted P-value, with red…

• 

  Figure 3. Sarcomere shortening and augmented cardiac stress in het atrial hiPSC-CMs. (A) Top Left: atrial hiPSC-CMs were stained for α-actinin, F-Actin and DAPI (Nuclei). Scale bar = 20 μm. Top Right: The Z-lines and thin myofilaments of the cells were mapped and analysed using a Z-line detection script38 to ascertain sarcomere length, continuous Z line length and Z line and Actin myofilament Organization (orientation order parameter). WT N = 3, n = 50, Het N = 3, n = 47. (B) RT-qPCR demonstrated a significant increase in NPPB (N = 3, n = 9), PDGFA and PDGFB (N = 4, n = 12) expression. (C) Gene (N = 6, n = 18) and protein expression (ELISA) (N = 4, n = 8) of the pro-fibrotic cytokine TGF-β1 was unchanged in the Het cells. A Welch’s t test was used to ascertain significance (**** denotes P < 0.0001, ** denotes P < 0.01, ns = P > 0.05).

• 

  Figure 4. Pro-fibrotic gene expression and remodelling in Het EHTs. (A) Co-cultured EHTs constructed using atrial hiPSC-CMs and quiescent hiPSC-CFs (Scale Bars: hiPSC-CF = 150 μm, hiPSC-CMs = 20 μm, EHT = 2 mm, EHT IHC inlet = 20 μm). (B) Left: Representative beat curves showing patterns of contraction and relaxation of WT and Het EHTs. Right: Contractile Force and Beat Duration (WT N = 5 EHT batches, n = 17 EHTs; Het N = 5, n = 21). Linear mixed models were used to ascertain significance (** denotes P < 0.01). (C) qPCR was performed on EHTs for pro-fibrotic and extracellular matrix genes: COL1A1, COL3A1 and PDGFB (N = 3 batches of EHTs, n = 7 EHTs for WT and n = 9 EHTs for Het, 3 technical replicates performed per EHT). A Mann–Whitney U test was used to ascertain significance for COL1A1 and COL3A1. A Welch’s t test was used to ascertain significance of PDGFB (**denotes P < 0.01). (D) Supernatant was taken from the EHTs and ELISAs were performed for Fibronectin-1, TGF-β1and Pro-Collagen 1. A Mann–Whitney U test was used to ascertain significance for Fibronectin-1. Welch’s t test was used to ascertain significance for TGF-β1 and Pro-Collagen 1 (*denotes P < 0.05, **denotes P < 0.01,…

• 

  Figure 5. Alterations in spontaneous action potential frequency and ion channel subunit gene expression in WT and het atrial hiPSC-CMs. (A) Left: A representative spontaneous action potential trace from WT and Het atrial hiPSC-CMs. Right: Spontaneous action potential frequencies observed in WT (N = 3, n = 21) and Het hiPSC-CMs (N = 3, n = 23). A Mann–Whitney U test was used to ascertain significance (*denotes P < 0.05). (B) RT-qPCR was performed using primers designed to amplify ion channel subunit genes KCNJ3, KCNK1 and KCND2 (N = 3, n = 9). A Mann–Whitney U test was used to ascertain significance of KCNJ3 and KCND2 (****denotes P < 0.0001). A Welch’s t test was used to ascertain significance of KCNK1 (**denotes P < 0.01). (C): Left: Time dependent recovery kinetics of INa Right: INa Time Recovery P50 (WT, N = 3, n = 33 and Het, N = 3, n = 19). A Mann–Whitney U test was used to ascertain significance (** denotes P < 0.01). (D) RT-qPCR was performed using primers designed to amplify ion channel subunit genes SCN1A and SCN9A (N = 3, n = 9). A Welch’s t test was used to ascertain significance (*denotes P < 0.05, **denotes P < 0.01, ***P < 0.001. P < 0.0001).

• 

  Figure 6. Enhanced electrophysiological sensitivity to carbachol in WT and het atrial hiPSC-CMs. Spontaneous action potentials (WT = black, Het = red) were recorded prior to and after the treatment of the cells with 2 μM of the acetylcholine receptor agonist carbachol (CCh) [in figure WT/Het label; control/carbachol]. (A) Representative traces of spontaneous action potentials recorded prior to and after the addition of 2 μM carbachol in WT and Het atrial hiPSC-CMs. (B) The effect of carbachol on AP amplitude, maximum diastolic membrane potential, spontaneous upstroke velocity, APD90 spontaneous AP frequency and relative APD90 post-carbachol treatment showing the mean and SD (N = 3, n = 8, N = 3, n = 8). A mixed effects one-way ANOVA was used to ascertain significance of carbachol treatment within each cell line (*P < 0.05, ***P < 0.001).

• 

  Figure 7. In silico action potential properties following carbachol and acetylcholine simulation. (A) Initial and calibrated population of atrial cardiomyocyte (CM) models. (B) APD at 90% of cellular repolarisation (APD90) for control conditions (CTRL) and after 2.0 μM carbachol (CCh) in the calibrated population of atrial CM models (C). Comparison of the relative APD90 (prior and after 2.0 μM CCh) in-vitro vs. in-silico for WT and Het atrial hiPSC-CMs (in-vitro, mixed effects one-way ANOVA, *P < 0.05 ) and atrial CM models with baseline (≤1, WT) or up-regulated (≥1.7, Het) IKACh density (in silico) (D). In-silico comparison of the effects of 2.0 μM CCh vs. 0.1 μM acetylcholine (ACh) for the same subgroups shown in (C). The blue circle highlights the atrial CM model used for multi-scale simulations.

• 

  Figure 8. Sinus tachycardia-induced atrial fibrillation (AF) for WT and het conditions. (A) In-silico atrial cardiomyocyte model used for the WT, and a schematic representation of the human whole-atria model used for multi-scale simulations. Consecutive snapshots of the atrial transmembrane voltage (Vm) show that under sinus tachycardia AF is not invoked. (B) In-silico atrial cardiomyocyte model used for Het TTNtv. Snapshots of Vm show AF induction and maintenance under sinus tachycardia. The initiation of AF for the Het condition is illustrated and explainedin the box at the bottom of the figure. SAN, sinoatrial node; SCV-ICV, superior and inferior cava vein; rs-ri-ls-li-PV, right superior, right inferior, left superior and left inferior pulmonary vein; CS, coronary sinus. The release of 0.1 μM acetylcholine only occurs at the ganglionated plexuses (GPs), arranged following an octopus configuration.