Reducing RBM20 activity improves diastolic dysfunction and cardiac atrophy

Figures

• 

  Fig. 1. Reduced RBM20-dependent alternative splicing restores cardiac dimensions in titin N2B knockout hearts. a Titin isoform expression of left ventricles from wildtype (WT) vs. heterozygous RBM20-deficient (R) and N2B knockout (N) vs. compound RBM20/N2B-deficient animals (NR). b Protein and RNA expression analysis of RBM20. Representative western blot of RBM20 and the quantification normalized to β-tubulin in WT, RBM20-HET, N2B-KO, and splice-rescue mice (n = 6 for each group). Quantification of Rbm20 mRNA expression normalized to Eef1a1 (n = 5–6 for each group). c Heart weight (HW) to body weight (BW) ratios (n = 8–10 for each group). d Representative longitudinal images of Masson’s trichrome-stained hearts with ventricular area indicated by the dashed line. Scale bar, 1 mm; original magnification, ×5. *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001 (b, c)

• 

  Fig. 2. Heterozygous loss of RBM20 normalizes cardiac function in titin N2B-KO mice. Data in (a–c) were obtained by echocardiography and (d) by Doppler echocardiography. a Left ventricular inner diameter in diastole (LVIDd) and systole (LVIDs) in the four experimental groups (n = 9–10 for each group). b Calculated heart weight (HW) to body weight (BW) ratios (n = 9–10 for each group). c Left ventricular fractional shortening (LVFS) and left ventricular ejection fraction (LVEF) between genotypes (n = 9–10 for each group). d Mitral valve E-wave to A-wave ratio as a parameter of diastolic stiffness (n = 8–10 for each group). *P < 0.05; **P < 0.01; ***P < 0.001 ; ****P < 0.0001

• 

  Fig. 3. Reduced RBM20 activity decreases diastolic stiffness in N2B-KO hearts. Data in (a–d) were obtained by cardiac catheterization. a End-systolic (ESP) and end-diastolic (EDP) pressures in the four experimental groups (n = 8–10 for each group). b Changes in maximum velocity of contraction (dP/dt max) and maximum velocity of relaxation (−dP/dt max) between genotypes (n = 8–10 for each group). c Representative PV loops of the N2B-KO indicating end-systolic (ESPVR, dark blue) and end-diastolic pressure-volume relationships (EDPVR, light blue) and the quantification of ESPVR as a measure of end-systolic elastance and the EDPVR as a measure of diastolic stiffness (n = 6–7 for each group). d Preload recruitable stroke work (PRSW) as a measure of myocardial contractility (n = 6–9 for each group). *P < 0.05; **P < 0.01

• 

  Fig. 4. Whole transcriptome analysis by RNAseq. a Matrix of differentially regulated genes between WT, RBM20-HET, N2B-KO, and splice-rescue mice (P < 0.05; n = 2 for N2B-KO, n = 3 for other groups). Each square provides the number of differentially expressed genes per comparison. Color intensity reflects the gene count as indicated by the heat map. b Of the six possible comparisons between experimental groups, we focused differences of wildtype to the three mutants (WT/N, WT/R, WT/NR) and the comparison of N2B-KO with the splice rescue (N/NR). Genes differentially expressed between WT and N2B-KO but not between WT and splice rescue were normalized by reduced RBM20 activity (reverted genes). Genes differentially expressed between WT and splice rescue but not between WT and N2B-KO were additionally regulated by reduced RBM20 levels (misregulated genes). c Enrichment analysis of reverted genes (top 5 biological processes ranked by P value). d The reverted genes and misregulated genes were classified by their gene ontology

• 

  Fig. 5. RBM20-dependent isoform expression. Differential inclusion of RBM20 exons 6 and 7 (a), titin N2B region (b), and genotype-dependent isoform expression of full-length Rbm20 (c), Ttn (d). genotype-dependent isoform expression of Camk2d (e), Ldb3 (f), Ttc17 (g), Obscn (h), Ank3 (i), and Fhl1 (j). PSI scores (percentage spliced in) are indicated on the y-axis for each experimental group (n = 2 for N2B-KO, n = 3 for other groups). Exon labels are below; size bar, 10 kb. ****P < 0.0001

• 

  Fig. 6. FHL1 and FHL2 levels are inversely regulated in splice-rescue mice. a Representative western blot analysis of FHL1 and FHL2 expression levels in LV tissue of WT, RBM20-HET (R), N2B-KO (N), and splice-rescue mice (NR). b Quantification of FHL1A, FHL1B, and FHL2 protein expression normalized to β-tubulin (n = 6 for all groups). (c) Quantification of Fhl1 and Fhl2 mRNA levels by TaqMan normalized to Eef1a1 (n = 5–6) *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001

• 

  Fig. 7. Splice-directed therapy in N2B KO mice with diastolic dysfunction and cardiac atrophy. Loss of the elastic N2B region (blue) leads to cardiac atrophy with reduced FHL2 and increased FHL1B expression. Reducing RBM20 activity to 50 % (splice rescue) increases exon retention both in titin’s I-band region (green) and in FHL1B. Together the RBM20-dependent adaptation of cardiac splicing restores titin-based elasticity and cardiac size and adapts FHL protein levels and fatty acid (FA) metabolism