Intercalated disc protein Xinβ is required for Hippo-YAP signaling in the heart

Abstract

Intercalated discs (ICD), specific cell-to-cell contacts that connect adjacent cardiomyocytes, ensure mechanical and electrochemical coupling during contraction of the heart. Mutations in genes encoding ICD components are linked to cardiovascular diseases. Here, we show that loss of Xinβ, a newly-identified component of ICDs, results in cardiomyocyte proliferation defects and cardiomyopathy. We uncovered a role for Xinβ in signaling via the Hippo-YAP pathway by recruiting NF2 to the ICD to modulate cardiac function. In Xinβ mutant hearts levels of phosphorylated NF2 are substantially reduced, suggesting an impairment of Hippo-YAP signaling. Cardiac-specific overexpression of YAP rescues cardiac defects in Xinβ knock-out mice-indicating a functional and genetic interaction between Xinβ and YAP. Our study reveals a molecular mechanism by which cardiac-expressed intercalated disc protein Xinβ modulates Hippo-YAP signaling to control heart development and cardiac function in a tissue specific manner. Consequently, this pathway may represent a therapeutic target for the treatment of cardiovascular diseases.

Fig. 1. Loss of Xinβ results in alteration of the Hippo-YAP signaling pathway in the heart.

a Smear plot showing the log fold change and average abundance of each gene. Differentially expressed genes are marked red; the expression of Xinβ (also called Xirp2) is among the mostly downregulated in P7.5 (day 7.5) Xinβ^KO hearts; b hierarchical clustering heatmap of differentially expressed genes; c gene set enrichment analysis (GSEA) with gene ontology gene-sets reveals molecular pathways related to cell proliferation and inflammation dysregulated in P7.5 Xinβ^KO hearts; d qRT-PCR of cell-cycle-related genes in P7.5 Xinβ^KO and control hearts. N = 4 biologically independent samples, *P < 0.05. e qRT-PCR of genes related to immune and inflammatory response in P7.5 Xinβ^KO and control hearts. N = 4 biologically independent samples, *P < 0.05. f Enrichment plot showing downregulation of the YAP-conserved pathway; g qRT-PCR of YAP pathway genes in P7.5 Xinβ^KO and control hearts. N = 4 biologically independent samples, *P < 0.05. h Workflow of comparative analysis of dysregulated genes between Xinβ^KO hearts and Ad-YAP cardiomyocytes; i GSEA with MSigDB Hallmark gene sets (v 6.2) displaying differentially regulated pathways between Xinβ^KO hearts and Ad-aYAP cardiomyocytes.

Fig. 2. Decreased cardiomyocyte proliferation in Xinβ^KO hearts.

a Edu incorporation in P7.5 Xinβ^KO and control hearts. DAPI marks nuclei, cTnT labels cardiomyocytes. Quantification in right panel. Scale bars = 100 µm. N = 4 and 5 biologically independent samples, respectively, *P < 0.05. b pH3 staining of P7.5 Xinβ^KO and control hearts. DAPI marks nuclei, α-actinin labels cardiomyocytes. Quantification in right panel. Scale bars = 40 µm. N = 4 and 3 biologically independent samples, respectively, *P < 0.05. c Heart morphology (left panel) and ventricle weight vs. tibial length ratio (right panel) of P7.5 Xinβ^KO and control hearts. Scale bars = 2 mm. N = 9 and 8 biologically independent samples, respectively,*P < 0.05. d Hematoxylin and eosin (red) and Fast Green (green) staining of p7.5 Xinβ^KO and control hearts. Scale bars = 1 mm. e AAV9-based CRISPR/Cas9 strategy to mutate the Xinβ gene in postnatal mouse hearts; f percent fractional shortening (FS) change in 4-week-old AAV-cTnT-Cre and AAV-Xinβ-sgRNA-treated mice. N = 13 and 12 biologically independent samples, respectively, *P < 0.05. g Morphology and immunostaining of isolated cardiomyocytes from hearts of adult Xinβ^KO and control mice. Scale bars = 50 µm.

Fig. 3. Heart-specific YAP^S127A overexpression rescues cardiac defects in Xinβ^KO mice.

a Diagram of the workflow. b Survival curve of WT and Xinβ^KO mice injected with AAV-aYAP (YAP^S127A) or control AAV-GFP. c Gross morphology of P7.5 WT and Xinβ^KO mice injected with AAV-aYAP or control AAV-GFP. Scale bars = 1 cm. d Heart gross morphology and, e histology, of P7.5 WT and Xinβ^KO mice injected with AAV-aYAP or control AAV-GFP. Scale bars = 2 mm (d) and 1 mm (e, f). Principal component analysis of all expressed genes in P7.5 WT and Xinβ^KO mice injected with AAV-aYAP or control AAV-GFP. g Hierarchical clustering heatmap of differentially expressed genes. h Gene set enrichment analysis (GSEA) of differentially regulated pathways P7.5 WT and Xinβ^KO mice injected with AAV-YAP or control AAV-GFP. i Enrichment plot of rescued YAP-conserved signatures in Xinβ^KO hearts injected with AAV-aYAP. j qRT-PCR of cell-cycle-related genes in the hearts of WT and Xinβ^KO mice injected with AAV-aYAP or control AAV-GFP. N = 6 or 7 biologically independent samples, *P < 0.05. k pH3 staining of hearts from P7.5 Xinβ^KO and control mice injected with AAV-aYAP or control AAV-GFP. DAPI marks nuclei, α-actinin labels cardiomyocytes. Quantification in right panel. Scale bars = 40 µm. N = 5 or 6 biologically independent samples, *P < 0.05.

Fig. 4. Xinβ interacts with NF2 to modulate the Hippo-YAP signaling in the heart.

a Western blots of indicated proteins using heart samples from P7.5 Xinβ^KO and control mice. b Quantification of p-NF2 and p-YAP protein levels in P7.5 Xinβ^KO and control hearts. N = 3 biologically independent samples, *P < 0.05. c Immunohistochemistry detecting p-NF2 in P7.5 Xinβ^KO and control hearts. DAPI marks nuclei, cTnT labels cardiomyocytes. Scale bars = 20 µm. d Immunohistochemistry detecting total YAP (t-YAP) in P7.5 Xinβ^KO and control hearts. DAPI marks nuclei, α-actinin labels cardiomyocytes. Scale bars = 20 µm. Quantification of nuclear located YAP-positive cardiomyocytes is shown in the right panel. N = 5 and 4 biologically independent samples, respectively, *P < 0.05. e Immunohistochemistry detecting NF2 and Xinβ expression in the intercalated discs of adult cardiomyocytes. Scale bars = 20 µm. Quantification is presented in the right panel. N = 11 biologically independent samples, Data are presented as mean values ± SEM. f Co-immunoprecipitation assays detecting interactions between GFP-tagged NF2 and Flag-tagged Xinβ protein fragments. 5% cell lysate was used as input to demonstrate the expression of tagged proteins. g Co-immunoprecipitation assay detecting the interaction between GFP-tagged NF2 and Flag-tagged Xinβ protein fragment (amino acids 334-680). Five percent cell lysate was used as input to demonstrate the expression of tagged proteins. h Co-immunoprecipitation assays detecting the interaction between GFP-tagged NF2 and Flag-tagged Xinβ protein fragment (amino acids 334-1075) using antibodies to Flag (top) and GFP (bottom). Five percent cell lysate was used as the input to demonstrate the expression of tagged proteins. i Summary of the Xinβ domains that interact with NF2. j Interaction of endogenous Xinβ with NF2 protein. k qRT-PCR of human Xinβ (hXinβ) and cardiomyopathy marker genes NPPA and NPPB in heart samples from dilated cardiomyopathy (DCM) and control patients. N = 5 biologically independent samples, *P < 0.05.

Fig. 5. Model of Xinβ regulation of the Hippo-YAP signaling pathway in cardiomyocytes.

Xinβ negatively modulates Hippo-YAP signaling by recruiting phosphorylated NF2 to the sarcolemmal membrane of cardiomyocytes, leading to the reduction of total NF2 protein, repression of Hippo signaling, and the activation of the downstream mediator YAP (left panel). Loss of Xinβ results in activation of Hippo signaling and a reduction of functional YAP levels, which reduces expression of downstream targets.