Multi-Omics Approach Profiling Metabolic Remodeling in Early Systolic Dysfunction and in Overt Systolic Heart Failure

Figures

• 

  Figure 1. Visualization of the MOD and SHF proteomic datasets. (A) Heat maps showing differential log2 fold change in expression of identified proteins that changed in MOD vs. Sham (left), and in SHF vs. Sham (right). The same Sham biological samples were used as a reference for both TMT-label proteomic runs; n = 3 biological samples per group. (B) Volcano plots of the identified proteins that changed in MOD (left) and SHF (right) groups vs. Sham. The x-axis displays the log2 fold change in group means for down-regulated proteins (green) and up-regulated proteins (red) in MOD or SHF vs. Sham. The y-axis displays the -Log10 p-value of identified proteins that were statistically significant in MOD (left) and SHF (right) vs. Sham. A -Log10 p-value cutoff of 1.3 (p = 0.05) and higher was considered significant. (C) Venn diagrams showing the number of identified proteins that changed in SHF (red) and MOD (green) groups, divided into those whose relative expression decreased (left) or increased (right) vs. Sham. The blue and yellow diagrams show the total number of identified mitochondrial (mt) proteins in the MOD and SHF proteomic runs, respectively. (D) Heat maps showing differential log2 fold…

• 

  Figure 2. Top-ranked Canonical Pathways related to metabolism and Upstream Regulators, resulting from IPA’s ‘Comparison Analyses’ of the experimental groups. (A) ‘Core Analyses’ of the statistically analyzed proteomic datasets, by both Qlucore (Q) and Scaffold (S) bioinformatics software, were performed in IPA for each of the two-group comparisons, i.e., MOD vs. Sham and SHF vs. Sham, with a cutoff p-value < 0.05. Analyzed datasets were then subsequently compared with each other in IPA’s ‘Comparison Analyses’ function to yield the most enriched Canonical Pathways (left) related to metabolism and Upstream Regulators (right) that were shared among the two-group comparisons. The respective z-score-based heat maps indicate shared Canonical Pathways and Upstream Regulators that were up-regulated/activated or down-regulated/inhibited in the two-group comparisons, with orange and blue color intensities representing the z-score-based extent of up-regulation/activation or down-regulation/inhibition, respectively. Canonical Pathways and Upstream Regulators with a z-score outside the cutoff value are represented by a dot on the heat map. (B) Western blot in left ventricular myocardium showing the…

• 

  Figure 3. Oxidative phosphorylation in MOD and SHF. (A) Oxidative phosphorylation (OXPHOS) network of the electron transport chain (ETC) complexes I-IV transport system and ATP synthase are presented in MOD vs. Sham (upper panel) and SHF vs. Sham (lower panel). The network highlights mt OXPHOS proteins in each ETC complex that were downregulated (green) vs. Sham, were unchanged (grey) vs. Sham, or were not identified (white) in the MOD and SHF proteomic runs. The darker the green color, the higher the degree of downregulation. (B) Expression of the ETC complexes I-IV and ATP synthase (complex V) in MOD and SHF by immunoblotting, * p < 0.05 vs. Sham and † p < 0.05 vs. MOD.

• 

  Figure 4. Visualization of the non-targeted metabolomics dataset in Sham, MOD, and SHF groups. (A) Heat maps showing differential change in log2 fold expression of metabolites that changed in MOD vs. Sham (left), and in SHF vs. Sham (right); n = 3 samples per group. (B) Volcano plots showing the log2 fold change in group means, on the x-axis, and the corresponding -Log10 (p-value), on the y-axis, for the metabolites that were downregulated (green) or upregulated (red) in MOD (left) and SHF (right) vs. Sham. A -Log10 (p-value) cutoff of 1.3 and higher was considered significant. (C) PCA plots showing the variance in biological samples within and between the Sham, MOD, and SHF groups for the metabolites that changed in MOD (left) and SHF (right) vs. Sham. (D) Venn diagram showing the number of metabolites that changed in SHF (red) and MOD (blue) vs. Sham and those that changed in SHF vs. MOD (green). (E) Volcano plot of the 11 common identified metabolites that changed in MOD and SHF vs. Sham (intersection of red and blue diagrams excluding the green diagram). Of those, there was a trend for further downregulation of pantetheine 4′-phosphate and 2′,3′ cyclic AMP in SHF compared with MOD. (F)…

• 

  Figure 5. Fatty-acyl-carnitine abundance, top-ranked Metabolic Pathways, and Upstream Regulators in the MOD and SHF. (A,B) Heat map and PCA plot showing the differential log2 fold change in expression and variance of biological samples for fatty-acyl-carnitines that changed in MOD and SHF groups vs. Sham. (C) ‘Core Analyses’ of the statistically analyzed dataset, by Qlucore bioinformatics software, was performed in IPA for each of the two-group comparisons and then were compared in IPA’s ‘Comparison Analyses’ function. The generated heat maps show the most enriched Metabolic Pathways (upper panel) by p-value and Upstream Regulators (lower panel) by z-score. (D) Mechanistic network showing connection between calcium, as a second messenger, and metabolites implicated in calcium cycling and homeostasis. Please refer to prediction legend for details.

• 

  Figure 6. Visualization of the MOD and SHF phospho (p)-proteomic datasets. (A) Heat maps showing differential log2 fold change in the expression of p-sites that changed in MOD vs. Sham (left), and in SHF vs. Sham (right). The same biological samples in the Sham group were used as a reference for both TMT-label p-proteomic runs; n = 3 biological samples per group. Note: there was a technical problem in one of the Sham biological samples in the MOD p-proteomic run, which was eliminated as it was skewing the data. (B) Volcano plots showing the degree of log2 fold change, on the x-axis, with the corresponding -Log10 p-value, on the y-axis, for the p-sites that were downregulated (green) or upregulated (red) in MOD (left) and SHF (right) groups compared with Sham. (C) PCA plots showing the variance of biological samples in the Sham, MOD and SHF groups for the p-sites that changed in MOD (left) and SHF (right) vs. Sham.

• 

  Figure 7. Top-ranked Signaling Pathways in the MOD and SHF groups. (A) ‘Core Analyses’ of the statistically analyzed p-proteomic datasets by Qlucore and Scaffold bioinformatics software was performed in IPA for each of the two-group comparisons (MOD vs. Sham and SHF vs. Sham), which were then compared in IPA’s ‘Comparison Analyses’ function to yield the most enriched signaling Pathways that were shared among the two-group comparisons. The generated heat maps are presented by p-value (left) and by z-score (right). (B) Western blot showing post-translational modification of AMPK and PKA downstream protein targets as well as protein phosphatase 1 (PP1) expression in the MOD and SHF groups, * p < 0.05 vs. Sham and † p < 0.05 vs. MOD.

• 

  Figure 8. Schematic drawing highlighting metabolic pathways, calcium cycling, and implicated signaling pathways in Sham (A), and their directional change in MOD vs. Sham (B), and SHF vs. Sham (C). Green arrows promote pathway or activity. Rectangles show signaling pathways or proteins that were inhibited or down-regulated (green), respectively, and those that were activated or up-regulated (red), respectively. The deeper the color, the higher the intensity of activation/up-regulation or inhibition/down-regulation, as shown by the scale bar at the bottom of the figure. Oxidative phosphorylation (OXPHOS), fatty acid (FA) metabolism, and mitochondrial (mt)-FA β-oxidation were the most enriched/inhibited metabolic pathways in MOD vs. Sham, followed by BCAA catabolism and enzymes of the tricarboxylic acid (TCA) cycle. Glycolysis and pyruvate metabolism were the least enriched/inhibited in MOD vs. Sham. The metabolites taurine and 2′,3′cAMP were downregulated in MOD vs. Sham. Endoplasmic reticulum (ER)–mt interface protein kinase A (PKA) signaling and calcium cycling as well as PGC-1α signaling were attenuated in MOD vs. Sham. Glycogen degradation and protein kinase C isoform delta (PKCδ)…