Proteomic and metabolomic analysis of cardioprotection: Interplay between protein kinase C epsilon and delta in regulating glucose metabolism of murine hearts

Abstract

We applied a combined proteomic and metabolomic approach to obtain novel mechanistic insights in PKCvarepsilon-mediated cardioprotection. Mitochondrial and cytosolic proteins from control and transgenic hearts with constitutively active or dominant negative PKCvarepsilon were analyzed using difference in-gel electrophoresis (DIGE). Among the differentially expressed proteins were creatine kinase, pyruvate kinase, lactate dehydrogenase, and the cytosolic isoforms of aspartate amino transferase and malate dehydrogenase, the two enzymatic components of the malate aspartate shuttle, which are required for the import of reducing equivalents from glycolysis across the inner mitochondrial membrane. These enzymatic changes appeared to be dependent on PKCvarepsilon activity, as they were not observed in mice expressing inactive PKCvarepsilon. High-resolution proton nuclear magnetic resonance ((1)H-NMR) spectroscopy confirmed a pronounced effect of PKCvarepsilon activity on cardiac glucose and energy metabolism: normoxic hearts with constitutively active PKCvarepsilon had significantly lower concentrations of glucose, lactate, glutamine and creatine, but higher levels of choline, glutamate and total adenosine nucleotides. Moreover, the depletion of cardiac energy metabolites was slower during ischemia/reperfusion injury and glucose metabolism recovered faster upon reperfusion in transgenic hearts with active PKCvarepsilon. Notably, inhibition of PKCvarepsilon resulted in compensatory phosphorylation and mitochondrial translocation of PKCdelta. Taken together, our findings are the first evidence that PKCvarepsilon activity modulates cardiac glucose metabolism and provide a possible explanation for the synergistic effect of PKCdelta and PKCvarepsilon in cardioprotection.

Figure 1. Subcellular fractionation

Mitochondrial and cytosolic extracts were prepared from murine hearts as described in the Material and Methods section. The purity of the fractions was assessed by Western blotting for VDAC, a mitochondrial outer membrane protein, and myoglobin, an abundant cytosolic protein (A). Panel B illustrates the depletion of nuclear proteins (NuMA) and the enrichment of mitochondrial (prohibitin) compared to other membrane proteins (Na⁺/K⁺ ATPase).

Figure 2. 2-DE map of cardiac proteins in the mitochondrial and cytosolic fraction

Subcellular protein extracts were pre-labelled with Cy3 and Cy5 using the DIGE approach and co-separated on large format 2-DE gels using pH 3-10NL IPG strips followed by 12% SDS polyacrylamide gels. Images were acquired on a fluorescence scanner and counterstained with silver. A silver-stained image of mitochondrial extracts and a DIGE image of cytosolic extracts is shown in panel A and B, respectively. Analyses using DeCyder® software revealed the spots showing a significant difference in AE or DN hearts. Proteins were numbered and identified by nano-LC MS/MS (Supplemental Table I–III).

Figure 3. Differential expression of cytosolic enzymes in PKCε transgenic hearts

Principal Component Analysis on the set of differentially expressed cytosolic proteins (Anova<0.05) allowed clear discrimination of controls (grey), AE (green) and DN PKCε hearts (red) (A). The differentially expressed enzymes contributing to this discrimination formed a cluster of similar expression profiles (B) and are highlighted on the representative 2-DE gel shown below (C).

Figure 4. Nuclear magnetic resonance spectra of murine hearts

Representative high-resolution ¹H-NMR spectra of AE PKCε transgenic and control hearts. Within the aliphatic region of the NMR spectra (−0.05 to 4.2 parts per million), resonances have been assigned to 25 metabolites, including creatine (Cr), glycine (Gly), taurine (Tau), phosphocholine (PC), succinate (Succ), glutamate (Glu), alanine (Ala), and lactate (Lac). Sodium 3-trimethylsilyl-2,2,3,3-tetradeuteropropionate (TSP) was added into the samples for chemical shift calibration. The bottom panel shows the first principal component (PC1) and the differences between the two means of metabolites in hearts with and without PKCε activation.

Figure 5. Comparison of metabolites during normoxia (A) and after ischemia/reperfusion injury (B)

The histogram shows the fold change of metabolites in AE PKCε hearts normalized to controls. Note the inverse pattern of most cardiac metabolites before (A) and after ischemia/reperfusion injury (B), but the consistent elevation of the adenosine nucleotide pool in AE PKCε hearts. * Significant difference p<0.05, ** p<0.01

Figure 6. Metabolic recovery after ischemia/reperfusion

Average metabolite concentration after 30 min of reperfusion in control, AE and DN transgenic hearts (A). The short reperfusion time was adapted to assess the early recovery of cardiac metabolism upon restoration of blood flow. Note that concentrations of metabolites related to glucose metabolism (arrows) correlate with PKCε activity (B). * Statistically significant difference from DN hearts, ** statistically significant difference from control as well as DN hearts.

Figure 7. Mitochondrial translocation of PKCδ

Mitochondrial extracts of control, AE and DN PKCε hearts were analyzed by Western blotting and probed with antibodies for phospho-PKCδ (Thr⁵⁰⁵) and total PKCδ. Ponceau red staining served as a loading control. Note the increase in phosphorylation and mitochondrial translocation of PKCδ upon inactivation of PKCε. pT denotes phosphothreonine (A). Verification of the antibody specificity by using hearts from PKCδ-deficient mice (B, C). The antibody for PKCδ-pThr⁵⁰⁵ showed good specificity. The antibody for total PKCδ recognizes three bands, of which the middle one is specific for PKCδ.

Figure 8. Model for PKCε-initiated metabolic changes linked to cardioprotection

PKCε activation resulted in differential expression of pyruvate kinase (PK), enolase (Enol) and lactate dehydrogenase (LDH) with a corresponding reduction of glucose and lactate in murine hearts. Notably, a pronounced effect was also observed on cytosolic malate dehydrogenase (MDH) and aspartate aminotransferase (AAT), the two enzymatic components of the malate-aspartate shuttle. The latter transfers electrons from the cytoplasm to mitochondria and is important in allowing maximum release of the free energy in glycolysis under aerobic conditions. In contrast, inhibition of PKCε stimulated loop phosphorylation and translocation of PKCδ to cardiac mitochondria, where it targets the pyruvate dehydrogenase complex (PDC), the enzyme responsible for converting the glycolytic endproduct pyruvate to acetyl-CoA. The pyruvate dehydrogenase complex is located at the inner mitochondrial membrane and inhibited when phosphorylated by pyruvate dehydrogenase kinase (PDK) and activated upon dephosphorylation by pyruvate dehydrogenase phosphatase (PDP). Thus, PKCε and PKCδ activities influence key enzymatic reactions bridging aerobic and anaerobic glucose metabolism.