Fig 2: Identification of the SLM2-binding motifs in TTN mRNAA. Binding probability of SLM2 to TTN intron 131 predicted from CLIP datasets using the DeepRiPe machine learning model. B. Potential SLM2-binding UWAA repeat motifs in intron 131 of TTN mRNA. C. Occurrence of UWAA motifs in the TTN exons 131–133. D. Sanger sequencing showing individual (Mut1, Mut2, Mut3, and Mut4) and double (Mut1+2 and Mut3+4) mutations of the identified UWAA motifs in intron 131 in the TTN131–133 minigene. E. and F. Analysis of TTN mRNA splicing using splice-PCR targeting exons 131–132 after transfection of HEK293 cell with the WT TTN131–133 minigene or TTN131–133 minigene with individually (E) or doubly (F) mutated UWAA motifs along with a plasmid expressing human SLM2. CLIP, crosslinking and immunoprecipitation; DeepRiPe, Deep RBP binding preference.

Fig 3: SLM2-dependent effects in human iPSC-CMsA. Diagram showing the generation of iPSC-CMs. Fibroblasts of human healthy donors were reprogrammed using non-integrating Sendai virus to generate iPSC clones. iPSC-CMs were then generated by Wnt modulation and metabolic selection. B. IB showing AAV6-mediated overexpression of SLM2 in iPSC-CMs. ß-actin was used as a loading control. Mock indicates iPSC-CMs transfected with control AAV6 virus. C. Gene expression after overexpression of SLM2 by transduction with AAV6 virus (n = 4) analyzed by NGS. D. Immunofluorescence assay of the sarcomeric protein titin Z1/Z2. Scale bar, 50 µm. E. and F. Determination of sarcomere regularity using FFT. Four cardiac differentiation experiments for one cell line were used, and single iPSC-CMs were analyzed (Mock: n = 74 iPSC-CMs; SLM2: n = 85 iPSC-CMs). G. Change of Ca2+ transient rise time. H. Change of Ca2+ half decay time. I. and J. [Ca2+]i (F/F0) after overexpression of SLM2 in iPSC-CMs (Mock, n = 66; SLM2, n = 66). Data are presented as mean ± SEM. Statistical differences were calculated using Student’s t-test (*, P < 0.05). iPSC, induced pluripotent stem cell; iPSC-CM, iPSC-derived cardiomyocyte; AAV6, adeno-associated virus serotype 6; NGS, next-generation sequencing; FFT, Fast Fourier Transform.

Fig 4: slm2 knockdown results in heart failure in zebrafishA. Conservation of the SLM2 proteins in humans, rats, and zebrafish. The KH domain which mediates RNA interaction and the tyrosine (Y) residue which mediates splicing are labeled. B. PCR analysis of zebrafish injected with splice morpholino targeting slm2 mRNA (MO-slm2) or control MO (MO-Ctrl) at 1–2-cell-stage embryos. The 320-bp product indicates the wild-type slm2 and the 120-bp product indicates loss of function by exon skipping. C. The effectiveness of MO-slm2 validated by Sanger sequencing of the PCR products shown in (B). D. Representative images of zebrafish injected with MO-slm2 and MO-Ctrl, respectively. Scale bar, 500 µm. E. Magnification of the hearts of zebrafish shown in (D). F. Quantitative analysis of the phenotypes of zebrafish after MO-slm2 or MO-Ctrl injection. G. Ventricular fractional shortening of zebrafish after MO-slm2 or MO-Ctrl injection. H. Atrial fractional shortening of zebrafish after MO-slm2 or MO-Ctrl injection. I. Heart rate of zebrafish after MO-slm2 or MO-Ctrl injection. *, P < 0.05 (Student’s t-test). A, atrium; V, ventricle; HF, heart failure; WT, wild-type.

Fig 5: Regulation and enrichment of mRBPs in failing human heartsA. Regulation of 421 mRPB-coding genes in the heart of patients diagnosed with DCM (n = 33) compared to controls (n = 24) (Student’s t-test). Size of the bubble indicates a HSS (arbitrary unit) in 53 different human tissues from a total of 11,467 human tissue samples. B.SLM2 mRNA expression levels in heart tissues of patients (DCM and ICM; each n = 3) and controls (n = 5). The SLM2 mRNA expression levels were normalized to the mRNA expression level of RPLP0, which served as the reference gene. C. Representative IB image depicting the protein expression levels of SLM2 in heart tissues of DCM patients and controls. a-actin served as a loading control. D. Densitometric quantification of SLM2 protein expression in heart tissues of DCM patients (n = 4) and controls (n = 3). a-actin served as a loading control and was used for normalization. *, P < 0.05 (Student’s t-test). mRPB, mRNA-binding protein; DCM, dilated cardiomyopathy; ICM, ischemic cardiomyopathy; HSS, heart-specificity score; SLM2, Sam68-like mammalian protein 2; IB, immunoblot.