Fig 2: Quantification of ROS formation was performed on freshly isolated SSM (a–c) and IFM (d) treated with carbenoxolone (CBX), S-nitroso-N-acetyl-dl-penicillamine (SNAP), S-nitrosoglutathione (GSNO), a combination of a NO donor and hemichannel blocker (SNAP + CBX; GSNO + CBX), or dH2O (a). Additionally, ROS formation was measured with lower concentrations (1 µmol and 48 nmol) of GSNO (b) and was measured after application of 250 µmol Cx43 mimetic peptide Gap26 (c). Mitochondrial H2O2 production was measured during ADP stimulated complex 1 respiration for 4 min and the increase of H2O2 was expressed in nmol/min/mg protein by comparing the data to a standard curve. As a control, respiration chain uncoupler FCCP (e, f) and respiration chain complex 1 inhibitor rotenone (g, h) were applied to ROS measurements. Data correspond to mean ± SEM of 6–20 replicates per group. Asterisk (p < 0.05) indicates significant differences between marked groups

Fig 3: Phosphoproteomics identification of ErbB2 and ERRa post-translational control of cardiomyocyte structure and metabolism.a Volcano plots illustrating the significantly up-regulated (red) and down-regulated (blue) phosphopeptides from cardiac phosphoproteomics profiling of mouse models relative to WT (limma, p < 0.05, |FC | = 1.5, n = 5). b Consolidated phosphomotifs generated by PHOSIDA79 of pSer-modified phosphopeptides found differentially expressed in the mouse models compared to WT showing site-specific amino acid preferences adjacent to the central serine phosphorylated residue. c Phosphopeptide sequence motif discovery by the MoMo27 software tool based on the Motif-x28 algorithm along with PhosphoMotif Finder29 predicted phosphomotif-targeting kinases. d Bar chart showing the number of identified DEPPs from (a) harboring up- and/or down-regulated site-specific phosphorylation (limma, p < 0.05, |FC| = 1.5, n = 5). e Schematic showing the cardiomyocyte localization of 25% of DEPPs identified in KI:KO hearts from (d). Both the number and directional change of protein phosphosites are indicated. Sarcomere regions are shown: M, M line; Z, Z line; I, I band; A, A band, H, H zone. f Box plot of relative phosphorylation levels of DEPPs important for cardiomyocyte structural integrity across the genotypes relative to WT (n = 5). g Immunoblot analysis of total and phospho-Cx43 (S368) levels in heart tissue extracts (n = 3). Vinculin levels are shown as a loading control. h Box plot of relative phosphorylation levels of DEPPs important for glucose uptake and catabolism across the genotypes relative to WT (n = 5). i Immunoblot analysis of total and phospho-Pdha1 (S232) levels and its upstream kinase Pdk1 in heart tissue extracts (n = 3). Clock levels are shown as a loading control. j Scatter dot plot of mouse cardiac pyruvate dehydrogenase (PDH) activities (n = 5). Data in (f) and (h) are shown as box and whiskers plots: center line denotes median, box extends from 25th to 75th percentiles, and whiskers extend to the lowest and highest values; *p < 0.05, |FC| = 1.5 by limma relative to WT controls. Data in (j) represent means ± SEM; *p < 0.05 by ANOVA. See also Supplementary Fig. 2.

Fig 4: Expression profile of interesting genes and correlated morphological changes during the recovery period.The mRNA expression of (A) RANKL, (B) OPG, (C) RANKL/OPG, (D) Gja1, (E) Sost, (F) Dmp1, (G) PDNP and (H) OCN during the recovery period and the correlations of (I) OPG with Circularity, (J) of Gja1 with Circularity and (L) of Dmp1 and Area. Independent Student’s t-test was performed to test the differences of the mRNA expression of interesting genes between MLO-Y4 with and without loading history at each time point during the recovery period. *P < 0.05; **P < 0.01; ***P < 0.001. The dash lines in (I), (J), (K) and (L) indicate the 95% confidence interval. The values of each replicate are presented as the dots. The results are shown as mean ± standard deviation.

Fig 5: Effects of diabetes on structural remodeling and protein levels in the Wt and Hif1a+/-LV. (A) Representative of immunofluorescence confocal and light microscopy images of the hearts stained with anti-connexin 43 (Cx43, red) together with fluorescein-labeled wheat germ agglutinin (WGA, green); with picrosirius red (PSR, polarizing microscopy); and with anti-collagen 1 (Col1, red), autofluorescence (auto) are shown. The nuclei were counterstained with Hoechst 33342 (blue). Confocal images are stacked Z-plane sections from confocal microscopy. Scale bar: Cx43/WGA 25 μm, PSR 50 μm, Col1 25 μm. (B) Representative Western blot analyses of protein lysates from the isolated LVs of Wt and Hif1a+/- non-diabetic and diabetic hearts are shown. (C-E) A relative quantification of protein levels of Cx43, phosphoCx43 (pCx43), and Col1 (n = 3 per group) was performed. The bar graphs show the mean values of relative protein levels normalized to the loading control (GAPDH) ± SEM from densitometric analysis. *P < 0.05 (2-way ANOVA with Tukey’s post-test).