Phosphoproteomic Analysis Reveals Downstream PKA Effectors of AKAP Cypher/ZASP in the Pathogenesis of Dilated Cardiomyopathy

Figures

• 

  Figure 1. Phosphoproteomics workflow. Neonatal wild-type and Cypher-knockout mice were administered isoproterenol (15 mg/kg, 30 min) before the hearts were harvested. Heart tissue lysates were digested into peptides with trypsin overnight, followed by phosphopeptide enrichment using titanium dioxide spin tips. After washing, bound phosphopeptides were eluted and desalted with alkaline C18 for direct LC-MS analysis. Raw data were analysed using the MaxQuant software. DCM, dilated cardiomyopathy; isoproterenol, a PKA activator; TiO2, titanium dioxide; LC-MS, liquid chromatography–mass spectrometry.

• 

  Figure 2. Bioinformatic analysis of the phosphoproteomic results. (A) Distribution of differential phosphopeptides in Cypher-knockout (KO) mice compared to that in wild-type (WT) mice. (B) Distribution of single and double phosphosites based on differential phosphopeptides. (C) Distribution of serine (Ser) and threonine (Thr) phosphorylation based on differential phosphopeptides. (D,E) GO enrichment for biological process, cellular component, and molecular function in downregulated (D) and upregulated (E) phosphoproteins. (F,G) KEGG analysis of the top 10 signalling pathways that converge on downregulated (F) and upregulated (G) phosphoproteins.

• 

  Figure 3. Phosphorylation-specific motif analysis and candidate PKA substrates. (A) Grouping of differentially phosphopeptides. The 117 hypophosphorylated peptides in Cypher-KO mice were categorised into two groups: Group 1, containing phosphopeptides with PKA consensus motifs and Group 2, without the motifs. The 99 hyperphosphorylated peptides were divided into the other two groups: Group 3, containing phosphopeptides with SP motif and Group 4, without the motif. (B–E) The logo motif of phosphosites in Group 1 (B), Group 2 (C), Group 3 (D), and Group 4 (E). Sequence logos were created using WebLogo. (F) The list of candidate PKA substrates regulated by Cypher in Group 1. Sites marked in red are the reported PKA phosphorylation sites. Sites selected for further study are marked with an asterisk (*).

• 

  Figure 4. Cypher interacts with β-catenin and facilitates its phosphorylation at Ser675 by PKA. (A) Western blotting for β-catenin and its phosphorylation using heart lysates from neonatal WT and Cypher-KO mice. Isoproterenol (15 mg/kg, 30 min) was administered by intraperitoneal infusion before harvesting. (B) Quantification of p-β-catenin (Ser675 and Ser552) protein levels in heart lysates. N = 4 biologically independent samples. (C,D) Western blotting results showing p-β-catenin (Ser675 and Ser552) in Cypher-overexpressing HEK293T or empty vector control cells. The cells were treated with forskolin and IBMX (3-isobutyl-1-methylxanthine) for 30 min before collection. Results of quantitative analysis are shown (D). N = 4 biologically independent samples. (E,F) β-catenin or its mutant S675A was coexpressed with Cypher in HEK293T cells. The phosphorylated β-catenin at Ser675 with forskolin treatment was assessed and quantified. N = 3 biologically independent samples. (G) β-catenin interacted with Cypher in vivo. β-catenin was immunoprecipitated from neonatal mouse heart lysates, and the interaction with Cypher was validated by anti-Cypher antibody. Rabbit IgG served as the negative control.…

• 

  Figure 5. Cypher activates β-catenin via Ser675 phosphorylation and modulates cardiomyocyte proliferation. (A,B) Myc-tagged Cypher was overexpressed in non-cardiac HEK293T cells with no endogenous Cypher. Empty vector served as a control. The expression of p-β-catenin, β-catenin and its target genes, including cyclin D1, c-Myc, and c-jun, were detected using western blotting and quantified. N = 3 biologically independent samples. (C,D) H9C2 cells with endogenous Cypher were transfected with scramble control (SiScramble) or Cypher siRNA (SiCypher). Cypher, p-β-catenin, and β-catenin and the abundance of its target genes were examined and quantified. N = 3 biologically independent samples. (E) Dual-luciferase reporter assays were performed to evaluate β-catenin transcriptional activity. HEK293T cells were cotransfected with Cypher or empty vector and TOP or FOP plasmid. The renilla plasmid was used as an internal control. N = 8 biologically independent samples. (F) The mRNA levels of the cell cycle-related genes (cyclin A2, cyclin B1, cyclin D1, and P21) were measured in Cypher-knockdown H9C2 cells using real-time quantitative polymerase chain reaction (RT-qPCR). N = 3 biologically…

• 

  Figure 6. Validation of vimentin and troponin I as PKA substrates regulated by Cypher. (A,B) Heart lysates from neonatal WT and Cypher-KO mice were analysed for total and phosphorylated vimentin. The mice were administered isoproterenol (15 mg/kg, 30 min) prior to sacrifice. Quantitative p-vimentin (Ser72) are shown (B). N = 4 biologically independent samples. (C,D) HEK293T cell lysates with Cypher overexpression were analysed for total and phosphorylated vimentin. Forskolin (100 μM, 30 min) and IBMX were used for activating PKA. p-Vimentin (Ser72) was quantitatively measured (D). N = 4 biologically independent samples. (E) Vimentin interacted with Cypher in vivo. Vimentin was enriched from neonatal mouse heart lysates, and interaction with Cypher was verified using anti-Cypher antibody. Rabbit IgG served as the negative control. (F,G) Coimmunoprecipitation (Co-IP) and reverse Co-IP experiments were performed to test the interaction between Cypher and vimentin in HEK293T cells. (F) Myc-tagged Cypher was enriched using anti-Myc antibody, and the interacting protein was probed with anti-Flag antibody. (G) Flag-tagged vimentin was enriched using anti-Flag antibody, and the interacting protein…