A novel Notch and WNT signaling mechanism contribute to paediatric DCM: a pathway to new therapeutics

Abstract

Background: Paediatric Idiopathic dilated cardiomyopathy (iDCM) is a life-threatening disease. The lack of disease-specific animal models limits our understanding of its mechanisms. We previously demonstrated that paediatric iDCM serum-circulating proteins promote pathologic remodeling in vitro, and that secreted frizzled related protein 1 (sFRP1) increases stiffness in cardiomyocytes. Here we investigated the mechanisms by which sFRP1 contributes to iDCM.

Methods: The effect of sFRP1 in combination with isoproterenol (ISO) (to recapitulate the increase in circulating catecholamine observed in paediatric iDCM) was evaluated in neonatal rat ventricular myocytes (in vitro), and in neonatal rats through intraperitoneal injections (in vivo). Function and molecular mechanisms were investigated through echocardiography and next-generation-sequencing. Protein levels and localization were determined by Western blot. Tissue stiffness was measured by Atomic Force Microscopy. In vitro and in vivo data were compared to explanted human heart tissue.

Results: We show that ISO+sFRP1 reactivates the fetal gene program in vitro, and promotes cardiac dysfunction, dilation and stiffness in vivo. Importantly, we show stiffness is also increased in paediatric iDCM hearts. We identified co-activation of Notch and WNT signaling in both ISO+sFRP1-treated rats and paediatric iDCM hearts. Mechanistically, in vitro inhibition of Notch or β-catenin prevented pathological remodeling, and Notch inhibition improved cardiac function, myocardial stiffness and ventricular dilation in ISO+sFRP1-treated rats.

Conclusion: We identified alterations in Notch and WNT signaling in paediatric iDCM hearts and in our model. Notch inhibition abrogated pathologic changes in vitro and in vivo. These findings provide novel mechanistic insights and a potential therapeutic target for paediatric iDCM.

Keywords: Heart failure; Notch; Paediatric; WNT; dilated cardiomyopathy; stiffness.

Figure 1.

ISO+sFRP1 promotes pathological remodeling in NRVMs. (A) RT-qPCR of NPPA expression in NRVMs treated with vehicle (Control), ISO, sFRP1 or ISO+sFRP1. Gene expression was normalized to 18S, and data are presented as a relative fold change to Controls. NRVMs treated with 100nM Isoproterenol (ISO) +/− 1 μg/mL human recombinant sFRP1 for 72 hours. n = 8 independent NRVM preps. All groups are log2 transformed. (B) RT-qPCR of NPPB expression in NRVMs treated with vehicle (Control), ISO, sFRP1 or ISO+sFRP1. Gene expression was normalized to 18S, and data are presented as a relative fold change to Controls. NRVMs treated with 100nM Isoproterenol (ISO) +/− 1 μg/mL human recombinant sFRP1 for 72 hours. n = 8 independent NRVM preps. All groups are log2 transformed. (C) RT-qPCR of MYH6/7 expression in NRVMs treated with vehicle (Control), ISO, sFRP1 or ISO+sFRP1. Gene expression was normalized to 18S, and data are presented as a relative fold change to Controls. NRVMs treated with 100nM Isoproterenol (ISO) +/− 1 μg/mL human recombinant sFRP1 for 72 hours. n = 8 independent NRVM preps. All groups are log2 transformed. (D) RT-qPCR of sFRP1 gene expression in NRVMs treated with ISO+sFRP1 compared to vehicle-treated controls. Gene expression was normalized to 18S, and data are presented as a relative fold change to controls. n = 6 independent NRVM preps.

Figure 2.

DCM phenotype in ISO+sFRP1 treated rats and cardiac stiffness is increased in paediatric DCM hearts and ISO+sFRP1 rats. (A) Schematic of injection plan, timeline and duration. Rats were treated with 0.1mg/kg/day ISO every other day for the first 5 treatments and every day for the subsequent 7 days. sFRP1 (50μg/kg/day) was co-administered with ISO (or vehicle (PBS, 0.5mM ascorbic acid). (B) EF of neonatal rats treated with vehicle (Control), ISO, sFRP1 and ISO+sFRP1. (C) LV vol(s) of neonatal rats treated with vehicle (Control), ISO, sFRP1 and ISO+sFRP1. (D) LVID(s) of neonatal rats treated with vehicle (Control), ISO, sFRP1 and ISO+sFRP1. Control n = 20, ISO n = 9, sFRP1 n = 9, ISO+sFRP1 n=19. Only significant p-values are notated in the Figure. (E) Elasticity/stiffness (Young’s modulus) in rats treated with ISO+sFRP1 compared to controls. Quantification is shown. n =3 per group (F) Elasticity/stiffness (Young’s modulus) in explanted paediatric DCM hearts compared to non-failing controls. Quantification is shown. NF n =3, DCM n =4. Only significant P values (p<0.05) are notated in the Figure. Fitting a mixed model, Tukey’s multiple comparisons test was used for multiple comparison. Unpaired t-test was used for two-group analysis. Error bar denotes mean± SEM. NRVMs, neonatal rat ventricular myocytes; FGP, fetal gene program; ISO, Isoproterenol; sFRP1, secreted frizzled protein 1; EF, ejection fraction; LV vol(s), Left ventricular volume in systole; LVID. Left ventricular internal diameter in systole; LVAW(s), left ventricular anterior wall in systole; LVPW (s), left ventricular posterior wall in systole.

Figure 3.

Notch and WNT signaling are predicted to be altered in ISO+sFRP1 rats and in paediatric DCM hearts via Next generation sequencing. (A) Schematic of the experimental plan for next generation sequencing (bulk RNA-seq and Single nuclei RNA-seq.) (B) Heatmap representing unsupervised hierarchical clustering of the top 50 significantly differentially expressed genes (DEGs) in bulk RNA sequencing. n = 6/group; Welch’s t test p < 0.05. (C) Pathway analysis using the 113 upregulated genes in ISO+sFRP1 vs Control rats using Reactome pathway analysis tool (left) and Hallmark pathway analysis tool (right). (D) Representative UMAP plot from single-nuclei RNA sequencing of LV tissue from Controls, ISO-, sFRP1- or ISO+sFRP1-treated rats, n = 3/group. A total of 9214 nuclei in control, 11588 nuclei in ISO-, 13758 nuclei in sFRP1-, and 11127 in ISO+sFRP1-treated rats were identified after cells were pooled. 26 distinct cell clusters were identified including 8 CMs and 5 FBs. (E) Reactome pathway analysis using upregulated genes in Ventricular cardiomyocyte sub-cluster 2 of ISO+sFRP1 vs Control rats (left). Enrichr analysis identifies transcription factors predicted to control expression of upregulated genes in ventricular cardiomyocyte sub-cluster 2 (right). (F) Reactome pathways analysis using upregulated genes in ventricular cardiomyocyte sub-cluster 6 of ISO+sFRP1 vs Control rats (left). Enrichr analysis identifies transcription factors predicted to control expression of upregulated genes in ventricular cardiomyocyte sub-cluster 6 (right). Clusters were annotated into major cell populations, and the ventricular cardiomyocyte cluster was sub-clustered into eight different sub-clusters using R statistical software. Differentially expressed genes for each cluster were calculated with the FindAllMarkers function. UMAP: uniform manifold approximation and projection, CM: cardiomyocytes; FB: fibroblast; NK: Natural Killer cells; B/T: B cells and T cells. DCM, dilated cardiomyopathy; ISO-Isoproterenol, sFRP1- secreted frizzled protein 1; NRVMs, neonatal rat ventricular myocytes; LV, left ventricle.

Figure 4.

Levels of NICD are increased in nuclear and cytoplasmic fractions in vivo, in vitro and in explanted human DCM hearts. (A, B) Western blot showing increased protein levels of NICD in the cytoplasm (A) and nuclear fractions (B) of NF vs DCM human LV tissue with quantification. Cytoplasmic n=6/group, nuclear n=7/group. (C, D) Western blot showing increased protein levels of NICD in the cytoplasm (C) and nuclear fractions (D) of ISO+sFRP1 vs vehicle-treated control rat LV tissue with quantifications. Cytoplasmic n=6/group, nuclear n=7/group. (E, F) Western blot showing the increased protein levels of NICD in the cytoplasm (E) and nuclear fractions (F) of ISO+sFRP1 vs vehicle-treated NRVMs with quantifications. Cytoplasmic n = 4 independent NRVM preps, nuclear n = 5 independent NRVM preps. NICD. Notch Intracellular domain, NRVMs, neonatal rat ventricular myocytes. NF, non-failing; DCM, dilated cardiomyopathy; NICD, Notch Intracellular domain. Unpaired 2-tailed t test was used in analysis. P-values are notated in each graph.

Figure 5.

Levels of β-catenin are increased in nuclear and cytoplasmic fractions in vivo, in vitro and in human DCM hearts. (A, B) Western blot showing increased protein levels of β-catenin in the cytoplasm (A) and nuclear fractions (B) of NF and DCM explanted human LV tissue with quantifications. NF, non-failing; DCM, dilated cardiomyopathy; cytoplasmic n=3/group, nuclear n=7/group. (C, D) Western blot showing increased protein levels of β-catenin in the cytoplasm (C) and nuclear fractions (D) of ISO+sFRP1 vs vehicle-treated control rat LV tissue with quantifications. Cytoplasmic n=3/group, nuclear, n=7/group. (E, F) Western blot showing increased protein levels of β-catenin in the cytoplasm (E) and nuclear fractions (F) of ISO+sFRP1 vs vehicle-treated NRVMs with quantifications. Cytoplasmic and nuclear fractions from n=4 independent NRVM preps. NRVMs, neonatal rat ventricular myocytes. Unpaired 2-tailed t test was used in analysis. p-values are notated in each graph.

Figure 6.

β-catenin contributes to Notch activation and inhibiting β-catenin prevents pathological remodeling in ISO+sFRP1-treated cells. (A) RT-qPCR of Notch target gene expression in NRVMs transfected with scrambled control siRNA (scr) or siCTNNB1, and treated with vehicle (Control) or ISO+sFRP1. Expression of Hey1, HEY2, HES1 and HEYL genes. Gene expression was normalized to 18S, and data are presented as a relative fold change to Controls. n=6 independent NRVM preps/group (B) RT-qPCR of FGP gene expression in NRVMs transfected with scrambled control siRNA (scr) or siCTNNB1 and treated with vehicle (Control) or ISO+sFRP1. Expression of NPPA, NPPB, and ratio of MYH6 to MYH7. Gene expression was normalized to 18S, and data are presented as a relative fold change to Controls. n=4 independent NRVM preps/group (C, D) Western blot showing reduced protein levels of NICD in the cytoplasm (C) and nuclear fractions (D) of NRVMs treated with ISO+sFRP1 and transfected with si-CTNNB1 when compared to ISO+sFRP1-treated cells with quantifications. n=3 independent NRVM preps/group. NICD, Notch Intracellular domain, NRVMs, neonatal rat ventricular myocytes.

Figure 7.

Notch contributes to β-catenin activation, and inhibiting Notch prevents pathological remodeling. (A) RT-qPCR of Notch target gene expression (HEY1, HEY2, HES1 and HEYL) in NRVMs treated with vehicle (Control), ISO+sFRP1, DAPT (20μM) or ISO+sFRP1+DAPT. Gene expression was normalized to 18S, and data are presented as a relative fold change to Controls. n = 7 independent NRVM preps/group (B) RT-qPCR of FGP in NRVMs treated with vehicle (Control), ISO+sFRP1, DAPT (20μM) or ISO+sFRP1+DAPT. Gene expression was normalized to 18S, and data are presented as a relative fold change to Controls. n = 5 independent NRVM preps/group (C) Western blot showing the reduced protein levels of β-catenin in the cytoplasm (C) and nuclear fractions (D) of ISO+sFRP1+DAPT or ISO+sFRP1-treated NRVMs with quantifications. P-values are notated in each graph. n = 4 independent NRVM preps/group. NRVM, neonatal rat ventricular myocytes (E) RT-qPCR of CTNNB1 gene expression in NRVMs treated with vehicle (Control), ISO+sFRP1, DAPT (20μM) and ISO+Sfrp1+DAPT. Gene expression was normalized to 18S, and data are presented as a relative fold change to Controls. Tukey’s multiple comparisons test was used for all data sets. Control, n = 4, ISO+sFRP1 n = 8, DAPT n = 4 and ISO+sFRP1+DAPT n = 7

Figure 8.

Inhibiting Notch prevents cardiac dysfunction in ISO+sFRP1-treated rats. (A) Schematic of the experimental plan for rats treated with ISO+sFRP1 +/−DAPT showing a prevention of cardiac dysfunction (B) EF of neonatal rats treated with Control, ISO+sFRP1 and DAPT ISO+sFRP1. (C) LV vol(s) of neonatal rats treated with Control, ISO+sFRP1 and DAPT ISO+sFRP1. (D) LVID(s) of neonatal rats treated with Control, ISO+sFRP1 and DAPT ISO+sFRP1. Control n=13, ISO+sFRP1 n=13. DAPT+ISO+sFRP1 n=6. (E) Elasticity/stiffness (Young’s modulus) in rats treated with DAPT+ISO+sFRP1 compared to ISO+sFRP1 treated rats and controls. Quantification is shown. Control n=3, ISO+sFRP1 n=3 and DAPT+ISO+sFRP1 n=4 (F) Simple linear regression of cardiac function (EF) and stiffness (Young’s modulus in kPa). EF: ejection fraction; LV vol(s), Left ventricular volume in systole; LVID. Left ventricular internal diameter in systole. Tukey’s multiple comparisons test was used for all data sets. P-values are notated in the graph.