Reperfusion-Induced Translocation of ␦PKC to Cardiac Mitochondria Prevents Pyruvate Dehydrogenase Reactivation Eric N. Churchill,* Christopher L. Murriel,* Che-Hong Chen, Daria Mochly-Rosen, Luke I. Szweda Abstract—Cardiac ischemia and reperfusion are associated with loss in the activity of the mitochondrial enzyme pyruvate dehydrogenase (PDH). Pharmacological stimulation of PDH activity improves recovery in contractile function during reperfusion. Signaling mechanisms that control inhibition and reactivation of PDH during reperfusion were therefore investigated. Using an isolated rat heart model, we observed ischemia-induced PDH inhibition with only partial recovery evident on reperfusion. Translocation of the redox-sensitive ␦-isoform of protein kinase C (PKC) to the mitochondria occurred during reperfusion. Inhibition of this process resulted in full recovery of PDH activity. Infusion of the ␦PKC activator H2O2 during normoxic perfusion, to mimic one aspect of cardiac reperfusion, resulted in loss in PDH activity that was largely attributable to translocation of ␦PKC to the mitochondria. Evidence indicates that reperfusion-induced translocation of ␦PKC is associated with phosphorylation of the ␣E1 subunit of PDH. A potential mechanism is provided by in vitro data demonstrating that ␦PKC specifically interacts with and phosphorylates pyruvate dehydrogenase kinase (PDK)2. Importantly, this results in activation of PDK2, an enzyme capable of phosphorylating and inhibiting PDH. Thus, translocation of ␦PKC to the mitochondria during reperfusion likely results in activation of PDK2 and phosphorylation-dependent inhibition of PDH. (Circ Res. 2005;97:78-85.) Key Words: pyruvate dehydrogenase 䡲 ␦PKC 䡲 pyruvate dehydrogenase kinase 䡲 free radicals 䡲 mitochondria 䡲 ischemia/reperfusion

P

yruvate dehydrogenase (PDH) is responsible for the conversion of pyruvate derived from glycolysis to acetylCoA for Krebs cycle activity. Enzyme activity is regulated, in part, by phosphorylation- and dephosphorylation-dependent inhibition and activation, respectively.1,2 Phosphorylation is catalyzed by 4 PDH-associated pyruvate dehydrogenase kinases (PDK1– 4) that exhibit tissue-specific expression patterns and differences in specific activity toward 3 phosphorylation sites on the E1␣ subunit of PDH. The PDH complex also contains 2 pyruvate dehydrogenase phosphatases (PDP 1 and PDP 2) responsible for reactivation of PDH.2– 4 PDH therefore represents a highly regulated and critical site for the control of glycolytic flux and ATP production. Cardiac ischemia/reperfusion is associated with alterations in metabolism that, depending on the severity of the ischemic insult, can progress to irreparable myocardial damage.5 Although PDH activity in myocardial tissue has been reported to decline during flow-induced ischemia,6 this is not universally observed.7–9 The effects of reperfusion also exhibit considerable variability, with the majority of studies demonstrating a decrease in PDH activity.7–9

Despite the disparity in evidence regarding PDH activity, cardiac efficiency and recovery of contractile function in postischemic hearts can be improved by pharmacological stimulation of PDH8,10 –16 or infusion of pyruvate.17–23 Identification of factors that regulate PDH activity during ischemia/reperfusion may therefore enhance the potential for therapeutic intervention. Reperfusion of ischemic myocardium is associated with enhanced free radical generation.5,24 Pro-oxidants have been shown to regulate protein function either directly or indirectly through the modulation of other regulatory molecules.25–27 One such example is the novel ␦-isoform of PKC. Exposure of purified ␦PKC to the thiol-specific oxidant diamide and glutathione (GSH) at concentrations that induce inactivation of other PKC isozymes results in ␦PKC activation.28 Additionally, treatment of various cell types with H2O2, glutathione depleting agents, or the general PKC activator PMA results in tyrosine phosphorylation and/or activation and translocation of ␦PKC to the mitochondria where it promotes cytochrome c release and the initiation of apoptosis.29 –35 In contrast, inhibition of ␦ PKC translocation reduces

Original received July 7, 2004; resubmission received February 10, 2005; revised resubmission received May 18, 2005; accepted June 6, 2005. From the Department of Physiology and Biophysics (E.N.C., L.I.S.), Case Western Reserve University, Cleveland, Ohio; and the Department of Molecular Pharmacology (C.L.M., C.-H.C., D.M.-R.), Stanford University School of Medicine, Calif. Dr Mochly-Rosen is a founder of KAI Pharmaceuticals, whose goal it is to bring peptide regulators of protein kinase C to the clinic. *Both authors contributed equally to this work. Correspondence to Luke I. Szweda, Oklahoma Medical Research Foundation, 825 NE 13th St, Oklahoma City, OK 73104. E-mail Luke-Szweda@ omrf.ouhsc.edu © 2005 American Heart Association, Inc. Circulation Research is available at http://circres.ahajournals.org

DOI: 10.1161/01.RES.0000173896.32522.6e

78

Churchill et al

␦PKC Maintains PDH Inhibition During Reperfusion

reperfusion-induced myocardial dysfunction and apoptosis and results in improved regeneration of intracellular ATP, phosphocreatine, and pH.36 – 40 In the present study, we tested the hypothesis that ␦PKC is involved in regulation of PDH during reperfusion. Rat hearts were perfused in a Langendorff fashion, and a specific peptide inhibitor of ␦PKC was used to test the contribution of ␦PKC to ischemia- and reperfusion-induced alterations in PDH activity. In addition, hearts were infused with H2O2 to gain insight into potential mechanisms responsible for concerted regulation of ␦PKC and PDH during ischemia/reperfusion. Finally, in vitro experiments were performed to address potential mechanisms by which ␦PKC influences the phosphorylation state of PDH.

Materials and Methods Rat Heart Perfusion and Isolation of Mitochondria Hearts isolated from male Sprague-Dawley rats (250 to 300 g, Zivic Miller, Pittsburgh, Pa) were perfused according to the Langendorff technique and after each experimental procedure, mitochondria were isolated as described.40

Measurement of PDH and Citrate Synthase Activities Mitochondria (100 ␮g/mL) in 20 mmol/L MOPS, 0.15% Triton, pH 7.4 were incubated with 200 ␮mol/L thiamine pyrophosphate, 40 ␮mol/L CoASH, 2.5 mmol/L pyruvate, 5.0 mmol/L MgCl2, 5.0 mmol/L CaCl2, 1.0 mmol/L NAD⫹, and ⫾0.5 mmol/L NaF. PDH activity was measured at 25°C as the rate of NADH production at 340 nm. Citrate synthase activity was measured as described.41

Evaluation of ␦PKC Translocation

Mitochondrial protein (60 ␮g/lane) was resolved by 4% to 15% SDS-PAGE, transferred to nitrocellulose membrane, and probed with polyclonal anti-␦PKC (Sigma). After incubation with alkaline phosphatase– conjugated anti-IgG rabbit antibody, binding was visualized by chemiluminescence (CSPD system, Tropix).

Analysis of PDH by 2-D Gel Electrophoresis Mitochondria (50 ␮g from each of 3 independent experiments) were pooled and solubilized in a buffer containing 7.0 mol/L urea, 2.0 mol/L thiourea, 4.0% CHAPS, and 0.5% IPG electrophoresis buffer. Protein was resolved by isoelectric focusing using precast Immobiline DryStrips (pI 3 to 10, 13 cm, Amersham) followed by 10% SDS-PAGE. On transfer to nitrocellulose membrane, Western blot analysis was performed using monoclonal antibody to the E1␣ subunit of PDH (Molecular Probes), HRP-conjugated secondary antibody (Amersham), and enhanced chemiluminescence (Sigma). For samples incubated with phosphatase before analysis, mitochondria (50 ␮g) were suspended in 50 ␮L of 50 mmol/L Tris-HCl, 100 mmol/L NaCl, 0.1 mmol/L EGTA, 2 mmol/L dithiothreitol, 0.01% Brij 35, 2.0 mmol/L MnCl2, 0.05% Triton X-100 at pH 7.0, and lysed in a water bath sonicator (Branson 1200) with three 30-s pulses. Lambda protein phosphatase (4000 U, New England Biolabs) was then added to the mitochondria extract and incubated at 30°C for 2 hours.

␦PKC Overlay Interaction Screen Approximately 20 000 lambda phage plaques were screened from a Sprague-Dawley rat heart cDNA library (Stratagene) as previously described.42,43 Recombinant bacterially expressed ␦PKC and partially purified rat brain PKC, in the presence of PKC activators (12 ␮g/mL phoshatidylserine, 2 ␮g/mL diacylglycerol) (Avanti Polar Lipids, Inc), were used as bait proteins. PKC binding was detected using rabbit polyclonal antibodies against ␣, ␤II, ⑀, or ␦PKC (Santa Cruz Biotechnology).

79

PKC and PDK2 Enzymes Recombinant rat ␦PKC was cloned into the pET28 vector, transformed into Escherichia coli BL21(DE3)pLysS cells, and expressed as a His-tagged fusion protein (Dirk Bossemeyer, Heidelberg, Germany). Recombinant rat His-tagged PDK2 protein was obtained from Paresh Sanghani (Indiana University School of Medicine, Indianapolis, Ind). Rat brain PKC enzymes were purified as previously described.44 Human recombinant ⑀ and ␦PKC were purchased from Invitrogen (Carlsbad, Calif).

Column Overlay Affinity Binding Assay 〈 partially purified rat brain PKC preparation (1.0 ␮g/mL in TBS) was incubated with 2.0 ␮g of polyclonal IgG antibodies against the ␣, ␤II, ⑀, or ␦PKC (Santa Cruz) and Protein G agarose beads (Santa Cruz) overnight at 4°C. Beads were washed (20 mmol/L Tris-HCl, pH 7.5, 150 mmol/L NaCl, 0.1% Triton X-100) and recombinant PDK2 protein (30 ␮g) was added and incubated for 1 hour at 4°C. Agarose-immobilized protein complexes were then washed and eluted by boiling the samples in Laemmli buffer. Protein was resolved (12.5% SDS-PAGE), transferred to nitrocellulose membranes, and probed with anti-His conjugated to HRP (Clontech) or with antibodies against ␣, ␤II, ⑀, or ␦PKC followed by anti-rabbit IgG antibodies liked to HRP (Amersham).

Assessment of Interactions Between PDK2 and PKC by ELISA Recombinant PDK2 (20 ng/␮L) in carbonate buffer (4.0 mmol/L Na2CO3, 3.6 mmol/L NaHCO3, pH 9.6) was placed in a 96-well (100 ␮L/well) flat bottom high binding Costar EIA/RIA plate (Corning) and incubated overnight at 4°C. Wells were washed and treated with 10 ␮g of partially purified brain PKC (diluted in 20 mmol/L Tris-HCl, pH 7.5) in the presence of activators (1 hour, 25°C). Binding was assessed using antibodies against ⑀ or ␦PKC (Santa Cruz), alkaline phosphatase (AP)-conjugated secondary antibodies (Boehringer Mannheim), and AP substrate (Pierce).

Assay of PKC-Dependent Phosphorylation of PDK2 Phosphorylation of recombinant PDK2 by purified brain ⑀ and ␦PKC was determined by detecting the incorporation of ␥-32P from [␥-32P]ATP (Amersham) in the presence of PKC activators but in the absence of Ca2⫹. The reactions were conducted at room temperature (20 minutes) and terminated by boiling samples in Laemmli buffer. Proteins were then resolved by 12.5% SDS-PAGE and transferred to nitrocellulose membranes. Densitometric analyses of autoradiograms were performed using NIH ImageJ software program. PDK2 (40 ␮g/mL) was incubated with recombinant human ⑀ or ␦PKC (4.0 ␮g/mL) in 20 mmol/L Tris-HCl, 10 mmol/L EGTA, 20 mmol/L ATP, 20 mmol/L MgCl2, pH 7.5 in the presence of PKC activators for 20 minutes at 37°C. Reactions were terminated by boiling in Laemmli buffer and proteins resolved by 10% SDSPAGE. After transfer to nitrocellulose, blots were probed with anti-PDK2 antibody (Abgent), anti-␦PKC (Santa Cruz), or a mixture of anti-phosphorylated serine PKC substrate, anti-phosphorylated threonine, and anti-phosphorylated threonine-X-arginine antibodies (Cell Signaling). Binding was detected using HRP-conjugated antirabbit IgG antibodies (Amersham).

PDK2 Peptide Activity Assay The activity of purified recombinant rat PDK2 was determined by measuring phosphorylation of the PDH E1␣ subunit tetradecapeptide substrate of PDK2 (YHGHSMSNPGVSYR, SynPep Corporation).45– 47 Phosphorylation was initiated by incubation of [␥-32P]ATP (Amersham) with 1.0 ␮g of PDK2, 100 ␮mol/L peptide, and 100 ng of ⑀ or ␦PKC (Invitrogen) in 20 mmol/L Tris-HCl, 10 mmol/L EGTA, 20 mmol/L ATP, 20 mmol/L MgCl2, pH 7.5. Reactions were conducted at 25°C for 20 minutes and terminated on addition of 25 ␮L of 200 mmol/L ATP/EDTA. The solution was applied to chromatograph paper, dried for 15 minutes at 25°C, rinsed

80

Circulation Research

July 8, 2005

Figure 1. Reactivation of PDH during cardiac reperfusion is prevented by translocation of ␦PKC to the mitochondria. A, Hearts were equilibrated (10 minutes) and then subjected to 150 minutes of perfusion (P150), 30 minutes of no-flow ischemia (I30), or 30 minutes of ischemia followed by 120 minutes of reperfusion (I30R120), in the presence or absence of the ␦PKC specific peptide inhibitor, Tat-␦V1-1 (1.0 ␮mol/L), infused for the first 10 minutes of reperfusion. B, The activity of PDH was measured in the presence of NaF (5.0 mmol/L). Values are presented as the mean⫾SD of 4 to 6 individual experiments. P values (2-tailed t test) where like symbols indicate values compared: * ⱕ0.002, † ⱕ0.001, ‡ ⱕ0.02, § ⱕ0.001. Native lipoic acid on the E2 subunit of PDH was examined by Western blot analyses using polyclonal antilipoic acid antibodies.54 C, Mitochondrial protein was subjected to Western blot analyses (60 ␮g protein per lane) using antibodies specific to ␦PKC, phospho␦PKC (serine/threonine), and adenine nucleotide translocase (ANT). Results of densitometric analyses of Western blots for ␦PKC (n⫽4) are represented as mean⫾SD with the mean for I30R120 intensity assigned a value of 100. P values: * ⱕ0.0005, † ⱕ0.00005, ‡ ⱕ0.0002. Citrate synthase (CS) activity was measured as described in Materials and Methods. D, Protein from total homogenate (T) (10 ␮g per lane) and isolated mitochondria (M) (10 ␮g per lane) were subjected to Western blot analysis using antibodies that recognize BiP (endoplasmic reticulum), GAPDH (cytosolic), the ␤-subunit of NaKATPase (plasma membrane),55 and ANT (mitochondria).

with H2O2 and 70% ethanol, and then dried for 10 minutes. Peptide phosphorylation was measured using a scintillation counter.

Results Translocation of ␦PKC to Mitochondria During Reperfusion Prevents Recovery of Pyruvate Dehydrogenase Activity No-flow ischemia (30 minutes) resulted in a 65% loss in PDH activity relative to activity measured in mitochondria isolated from perfused hearts (Figure 1B). PDH activity remained depressed during reperfusion (120 minutes), with only partial recovery in activity relative to ischemic values. The level of native lipoic acid on the E2 subunit of PDH was unaffected by ischemia or reperfusion indicating no alterations in protein content (Figure 1B). As shown in Figure 1C, ␦PKC translocated to the mitochondria during reperfusion. The relative level of total ␦PKC associated with the mitochondria is reflected by the appearance of phospho-␦PKC (Figure 1C). Infusion of the ␦PKC specific inhibitor Tat-␦V1-136,48 during the first 10 minutes of reperfusion (Figure 1A) reduced translocation of ␦PKC to the mitochondria during reperfusion (Figure 1C). Importantly, inhibition of ␦PKC translocation resulted in recovery of PDH activity to near control values during reperfusion (Figure 1B). Infusion of ␦V1-1 before ischemia failed to diminish ischemia-induced inhibition of PDH indicating that this decrease in activity is ␦PKCindependent. Alterations in PDH activity did not appear to be caused by global changes in mitochondrial function given that citrate synthase activity remained unchanged (Figure 1C). In addition, isolation of mitochondria did not result in significant copurification of contaminating fractions (Figure 1D), and infusion of the Tat carrier alone had no effect on

PDH activity or ␦PKC translocation (not shown). Finally, the PKC inhibitor rottlerin, at a concentration (10 ␮mol/L) specific to ␦PKC, exhibited effects similar to those observed for Tat-␦V1-1 (Figure 2). Therefore, whereas ␦PKC does not appear to be involved in inhibition of PDH activity during ischemia, translocation of the kinase to the mitochondria during reperfusion prevents complete reactivation of PDH.

H2O2 Induces ␦PKC Translocation and Inhibition of PDH

Pro-oxidants have been shown to activate ␦PKC, whereas other isoforms of PKC are inactivated.28,31,32,35 To further test whether ␦PKC translocation to the mitochondria is responsible for inhibition of PDH and to determine whether alterations in redox status may act as the stimulus for ␦PKC translocation in the intact heart, hearts were perfused in the absence and presence of H2O2 (250 ␮mol/L; Figure 3A). This resulted in translocation of ␦PKC to the mitochondria (Figure 3B) and an ⬇50% loss in PDH activity relative to controls (Figure 3C). Treatment of isolated respiring mitochondria with H2O2 had no effect on PDH activity suggesting the requirement for cytosolic factors (not shown). Coinfusion of the ␦PKC inhibitor Tat-␦V1-1 with H2O2 resulted in inhibition of ␦PKC translocation to the mitochondria (Figure 3B) and significant protection of PDH from H2O2-induced inhibition (Figure 3C). Thus, H2O2-induced inhibition of PDH is due, in large part, to ␦PKC translocation.

Phosphorylation-Dependent Inhibition of PDH during Ischemia and Reperfusion To determine whether reperfusion induces phosphorylationdependent inhibition of PDH, enzyme activity was measured

Churchill et al

␦PKC Maintains PDH Inhibition During Reperfusion

81

Figure 2. Rottlerin prevents ␦PKC translocation and PDH inhibition during cardiac reperfusion. Hearts were subjected to 150 minutes of perfusion (P) or 30 minutes of ischemia followed by 120 minutes of reperfusion (I30R120), in the presence or absence of rottlerin (10 ␮mol/L), infused for the first 10 minutes of reperfusion. The activity of PDH and the relative level of ␦PKC associated with the mitochondria were measured as described in the Figure 1 legend. Values for PDH activity are presented as the mean⫾SD of 4 individual experiments. P values: * ⱕ0.0003, † ⱕ0.001.

in the presence and absence of the general phosphatase inhibitor NaF. In mitochondria isolated from reperfused tissue, PDH activity was ⬇25% higher when measured in the absence of NaF (Figure 4A), indicating that ⬇50% of the enzyme activity lost during ischemia/reperfusion may be attributable to phosphorylation. In contrast, NaF had no significant effect on activity in mitochondria isolated from perfused tissue (⫹NaF, 84.9⫾12.1 nmol/min/mg; ⫺NaF, 79.8⫾5.4 nmol/min/mg). Thus, whereas PDH activity remained significantly below control values indicating inhibition/dissociation of enzyme associated phosphatase(s) or alternative mechanisms of inhibition, reperfusion-induced loss in PDH activity appears due, in part, to phosphorylation of the enzyme. PDH can be inhibited to varying degrees by phosphorylation of 3 serine residues on the E1 subunit of the enzyme.2 Two-dimensional Western blot analysis using anti-E1 antibody indicates that PDH migrates at 4 distinct isoelectric points consistent with 4 phosphorylation states of the protein (Figure 4B). In mitochondria isolated from perfused control hearts, the relative abundance of the E1 subunit increased with increasing pI (Figure 4B). On ischemia/ reperfusion, this distribution shifted in the acidic direction consistent with an increase in phosphorylation. Infusion of ␦PKC inhibitor Tat-␦V1-1 during ischemia/reperfusion prevented this shift (Figure 4B). Preincubation of mitochondria from reperfused hearts with phosphatase collapsed the distribution of the E1 subunit consistent with near complete dephosphorylation (⬇90% by densitometry; Figure 4B). Taken together, these results provide further evidence that the 4 isoelectric points represent different phosphorylation states

Figure 3. H2O2-induced translocation of ␦PKC to mitochondria and inactivation of PDH. A, Hearts were perfused for 60 minutes and then infused ⫹/⫺ H2O2 (250 ␮mol/L) for 10 minutes followed by 10 minutes of perfusion in the absence of H2O2. To inhibit ␦PKC translocation, Tat-␦V1-1 (1.0 ␮mol/L) was infused for 20 minutes (5 minutes before H2O2 until 5 minutes after H2O2). B, Mitochondria were subjected to Western blot analyses (60 ␮g protein per lane) using antibodies specific to ␦PKC and ANT. Results of densitometric analyses of Western blots for ␦PKC (n⫽6) are represented as mean⫾SD with the mean for H2O2 intensity assigned a value of 100. P values: * ⱕ0.0001, † ⱕ0.002. C, The activity of PDH was measured in the presence of NaF (5.0 mmol/L). The content of the PDH E2 subunit was assessed by Western blot analyses using antibodies specific to lipoic acid. Values are presented as the mean⫾SD of 8 individual experiments. P values: * ⱕ0.000003, † ⱕ0.005, ‡ ⱕ0.0006.

of the E1 subunit, and that ␦PKC plays a role in phosphorylation and inhibition of PDH.

␦PKC Phosphorylation and Activation of Pyruvate Dehydrogenase Kinase 2

A potential mechanism for ␦PKC-dependent inhibition of PDH is through the activation of specific kinases that phosphorylate and inhibit PDH. An unbiased screen of ⬇20 000 ␭

82

Circulation Research

July 8, 2005

Figure 4. Phosphorylation-dependent inhibition of PDH during ischemia/reperfusion. A, The activity of PDH was measured in the presence or absence of NaF (5.0 mmol/L) using mitochondria prepared after perfusion (P120) or ischemia/reperfusion (I30/ R120) in the absence or presence of Tat-␦V1-1. Values are presented as the mean⫾SD of 4 to 6 individual experiments. P values: * ⱕ0.03, † ⱕ0.02. B, Mitochondrial protein was resolved by 2-D gel electrophoresis followed by Western blot analysis using polyclonal antibodies to the E1␣ subunit of PDH. Where indicated, mitochondria were preincubated with phosphatase as described in Materials and Methods. Blots shown represent pooled mitochondrial samples from 3 separate hearts for each experimental protocol.

phage clones from a rat heart cDNA expression library was conducted using ␦PKC as the bait protein. After secondary and tertiary screens to enrich and validate ␦PKC protein/ protein interactions, 2 clones were isolated, sequenced, and identified as the rat form of pyruvate dehydrogenase kinase 2. Binding was specific to the ␦-isoform of PKC with no binding evident for ␣, ␤II, or ⑀PKC (Figure 5A). The interaction between PDK2 and ␦PKC and specificity of this interaction were confirmed using affinity chromatography with immobilized PKC isoforms (Figure 5B) and ELISA with immobilized PDK2 (Figure 5C). As shown in Figure 6A, purified ␦ and ⑀PKC catalyzed the in vitro phosphorylation of PDK2, with greater levels of phosphorylation evident for ␦PKC. Phosphorylation of PDK2 occurred on a serine/ threonine residue(s), consistent with the catalytic properties of ␦PKC (Figure 6B). To determine whether phosphorylation of PDK2 activates the kinase, a peptide analog of the phosphorylation site on PDH was used as substrate. As shown in Figure 6C, ␦PKC-dependent phosphorylation resulted in

Figure 5. In vitro interactions between PDK2 and ␦PKC. A, Purified recombinant rat heart ␦PKC was used as bait to screen a rat heart ␭ phage cDNA expression library in the presence of 12 ␮g/mL phosphatidylserine (PS) and 2.0 ␮g/mL diacylglycerol (DAG). Two positive clones were isolated and identified as PDK2. The specificity of PDK2/PKC interaction(s) were further evaluated using indicated isoforms of PKC purified from rat brain. B, Partially purified rat brain PKC isoforms (␣, ␤II, ⑀, ␦) were incubated with polyclonal IgG antibodies against the ␣, ␤II, ⑀, or ␦PKC and Protein G agarose beads. Immobilized protein was then incubated with purified recombinant His-tagged PDK2 (rat heart) in the presence of PS and DAG. PDK2 binding to PKC was detected by subjecting the eluted proteins to Western blot analysis with anti–His-HRP antibody. PKC immunoprecipitates were also probed with anti-PKC polyclonal antibodies. Blots shown are representative of 3 independent experiments. C, PDK2 was immobilized onto 96-well plates and incubated with either ␦ or ⑀PKC in the presence of PS and DAG. Protein complexes were probed with polyclonal antibodies against ⑀ and ␦PKC. Values are presented as the mean⫾SD of 6 individual experiments. P values: * ⱕ0.002, † ⱕ0.003.

activation of PDK2. Activation appears specific to ␦PKC in that no appreciable phosphorylation of the peptide analog was observed in the absence of ␦PKC or PDK2 or in the presence of ⑀PKC. It is well documented that PDK2 catalyzes the phosphorylation and inhibition of PDH.2 These results provide a plausible mechanism for ␦PKC-dependent inhibition of PDH during cardiac reperfusion.

Churchill et al

␦PKC Maintains PDH Inhibition During Reperfusion

Figure 6. ␦PKC-dependent phosphorylation and activation of PDK2. A, Purified ␦ and ⑀PKC were incubated with purified recombinant PDK2 in the presence of PS and DAG but in the absence of Ca2⫹. Phosphorylation of PDK2 was measured by incorporation of 32P on incubation with [␥-32P]ATP. Values are presented as the mean⫾SD of 4 individual experiments. P values: *Pⱕ0.02, † ⱕ0.0005. B, Recombinant rat heart PDK2 and human ␦PKC were incubated in the presence of ATP, PS, and DAG for 20 minutes at 37°C. Western blot analysis was then performed using antibodies raised to PDK2, phospho-serine/ threonine, and ␦PKC. C, 32P incorporation into the E1␣ subunit substrate peptide of PDK2 was determined in the presence PDK2, PDK2⫹␦PKC, or PDK2⫹⑀PKC. Values are presented as the mean⫾SD for 4 separate experiments conducted in duplicate. P values: * ⱕ0.001, † ⱕ0.002.

Discussion PDH was shown to decline in activity during cardiac ischemia. Although a fractional regain in PDH activity occurred on reperfusion, the activity of the enzyme remained depressed relative to control values. ␦PKC translocated to the mitochondria during reperfusion. Prevention of ␦PKC translocation resulted in complete recovery in PDH activity. Thus, ␦PKC prevents reactivation or promotes continued inhibition of PDH in response to cardiac reperfusion. Activation of PDH or inclusion of pyruvate during cardiac ischemia/reperfusion improves recovery of hemodynamic function.10 –14,16 –18,21–23

83

Recovery of PDH activity on inhibition of ␦PKC translocation may therefore, in part, provide an explanation for the previously observed cardioprotective role of the ␦PKC specific peptide inhibitor.36 – 40 Reperfusion of myocardial tissue is associated with a rapid increase in the level of various pro-oxidant species.5,24 Purified ␦PKC is sensitive to redox status, increasing in catalytic activity under oxidative conditions that induce inactivation of other PKC isoforms.28 In addition, treatment of cells in culture with H2O2 results in the translocation of ␦PKC to the mitochondria.33 We have demonstrated that perfusion of rat hearts with H2O2 induces the translocation of ␦PKC to the mitochondria and reduction in PDH activity. H2O2-dependent loss of PDH activity was largely prevented by inhibition of ␦PKC translocation. Treatment of mitochondria with H2O2 did not have an effect on PDH activity, indicating that cytosolic component(s) are necessary for inhibition of PDH. Therefore, pro-oxidants produced during cardiac reperfusion may provide the stimulus for ␦PKC translocation and PDH inhibition. PDH is regulated by specific kinases and phosphatases associated with the PDH complex.2– 4,49 In mitochondria isolated from reperfused tissue, PDH activity was partially recovered when assayed in the absence of the phosphatase inhibitor NaF. In addition, reperfusion-induced declines in PDH activity were associated with an acidic shift in the isoelectric point of the ␣E1 subunit of pyruvate dehydrogenase that was prevented on inhibition of ␦PKC translocation. Thus, ␦PKC appears to promote phosphorylation-dependent inhibition of PDH. In vitro data indicates that ␦PKC specifically interacts with PDK2, a kinase that can phosphorylate and inhibit PDH. Importantly, the interaction between ␦PKC and PDK2 leads to the phosphorylation and activation PDK2. Endogenous dephosphorylation of PDH (assayed in the absence of NaF) partially restored the ischemia/reperfusioninduced loss in PDH activity, suggesting additional modes of ␦PKC-dependent inhibition. One potential mechanism is through the inhibition of PDP1 and/or PDP2. However, when mitochondrial samples isolated from reperfused tissue were incubated with alkaline phosphatase to promote dephosphorylation, no further regain in enzyme activity was observed (not shown). In addition, it has been demonstrated that treatment of L6 skeletal muscle cells and immortalized hepatocytes with insulin results in ␦PKC activation and interaction with PDP1/2. This leads to activation of PDP1/2 and stimulation of PDH activity.49 Nevertheless, the effects of ␦PKC are likely to be tissue specific.49 An alternative mechanism of PDH inhibition may involve oxidative modification of PDP1/2 or PDH. Precedence for this possibility is provided by evidence that the phosphatase PTP1B is readily inhibited on glutathionylation in response to receptor stimulated pro-oxidant production50 and that PDH is susceptible to oxidative inhibition.51–53 In the present study, H2O2-dependent loss of PDH activity was partially prevented by inhibition of ␦PKC translocation. In contrast, prevention of ␦PKC translocation to the mitochondria during reperfusion resulted in full reactivation of PDH. This difference may be explained by previous findings that translocation of ␦PKC to the mitochondria results in release of cytochrome c that could

84

Circulation Research

July 8, 2005

in turn amplify mitochondrial free radical production.29 –35 Thus, prevention of ␦PKC translocation to the mitochondria during reperfusion would be expected to prevent ␦PKC- and redox-dependent inhibition of PDH.

Acknowledgments This work was supported by grants from the National Institutes of Health (R01 AG-19357 and R01 AG-16339 to L.I.S and 2RO1 HL-52141 to D.M.-R.). The authors thank Robert Harris, Paresh C. Sanghani and the Indiana Genomics Initiative (INGEN) Protein Expression Core of Indiana University (supported in part by Lilly Endowment Inc) for providing PDK-2.

References 1. Jafri MS, Dudycha SJ, O’Rourke B. Cardiac energy metabolism: models of cellular respiration. Annu Rev Biomed Eng. 2001;3:57– 81. 2. Patel MS, Korotchkina LG. Regulation of mammalian pyruvate dehydrogenase complex by phosphorylation: complexity of multiple phosphorylation sites and kinases. Exp Mol Med. 2001;33:191–197. 3. Simonot C, Lerme F, Louisot P, Gateau-Roesch O. Sub-mitochondrial localization of the catalytic subunit of pyruvate dehydrogenase phosphatase. FEBS Lett. 1997;401:158 –162. 4. Huang B, Gudi R, Wu P, Harris RA, Hamilton J, Popov KM. Isoenzymes of pyruvate dehydrogenase phosphatase. DNA-derived amino acid sequences, expression, and regulation. J Biol Chem. 1998;273: 17680 –17688. 5. Reimer, KA, Jennings, RB. Myocardial ischemia, hypoxia, and infarction. In: Fozzard, H. A., Haber, E., Jennings, R. B., Katz, A. M., Morgan, H. E., eds. The Heart and Cardiovascular System. 2nd ed. New York: Raven;1992:1875–1973. 6. Patel TB, Olson MS. Regulation of pyruvate dehydrogenase complex in ischemic rat heart. Am J Physiol. 1984;246:H858 –H864. 7. Kobayashi K, Neely JR. Effects of ischemia and reperfusion on pyruvate dehydrogenase activity in isolated rat hearts. J Mol Cell Cardiol. 1983; 15:359 –367. 8. Stanley WC, Hernandez LA, Spires D, Bringas J, Wallace S, McCormack JG. Pyruvate dehydrogenase activity and malonyl CoA levels in normal and ischemic swine myocardium: effects of dichloroacetate. J Mol Cell Cardiol. 1996;28:905–914. 9. Terrand J, Papageorgiou I, Rosenblatt-Velin N, Lerch R. Calciummediated activation of pyruvate dehydrogenase in severely injured postischemic myocardium. Am J Physiol Heart Circ Physiol. 2001;281: H722–H730. 10. Barak C, Reed MK, Maniscalco SP, Sherry AD, Malloy CR, Jessen ME. Effects of dichloroacetate on mechanical recovery and oxidation of physiologic substrates after ischemia and reperfusion in the isolated heart. J Cardiovasc Pharmacol. 1998;31:336 –344. 11. Kudej RK, White LT, Kudej AB, Vatner SF, Lewandowski ED. Brief increase in carbohydrate oxidation after reperfusion reverses myocardial stunning in conscious pigs. Circulation. 2002;106:2836 –2841. 12. Lewandowski ED, White LT. Pyruvate dehydrogenase influences postischemic heart function. Circulation. 1995;91:2071–2079. 13. Okuda K, Nohara R, Fujita M, Tamaki N, Konishi J, Sasayama S. Improvement of myocardial ischemic dysfunction with dichloroacetic acid: experimental study by repeated ischemia in dogs. J Cardiovasc Pharmacol. 1995;26:990 –999. 14. Racey-Burns LA, Burns AH, Summer WR, Shepherd RE. The effect of dichloroacetate on the isolated no flow arrested rat heart. Life Sci. 1989; 44:2015–2023. 15. Schoder H, Knight RJ, Kofoed KF, Schelbert HR, Buxton DB. Regulation of pyruvate dehydrogenase activity and glucose metabolism in postischaemic myocardium. Biochim Biophys Acta. 1998;1406:62–72. 16. Wahr JA, Olszanski D, Childs KF, Bolling SF. Dichloroacetate enhanced myocardial functional recovery post-ischemia: ATP and NADH recovery. J Surg Res. 1996;63:220 –224. 17. de Groot MJ, van der Vusse GJ. The effects of exogenous lactate and pyruvate on the recovery of coronary flow in the rat heart after ischaemia. Cardiovasc Res. 1993;27:1088 –1093. 18. Liedtke AJ, Nellis SH. Effects of buffered pyruvate on regional cardiac function in moderate, short-term ischemia in swine heart. Circ Res. 1978;43:189 –199.

19. Mallet RT, Squires JE, Bhatia S, Sun J. Pyruvate restores contractile function and antioxidant defenses of hydrogen peroxide-challenged myocardium. J Mol Cell Cardiol. 2002;34:1173–1184. 20. Mallet RT, Sun J. Mitochondrial metabolism of pyruvate is required for its enhancement of cardiac function and energetics. Cardiovasc Res. 1999;42:149 –161. 21. Mochizuki S, Neely JR. Energy metabolism during reperfusion following ischemia. J Physiol (Paris). 1980;76:805– 812. 22. Saiki Y, Lopaschuk GD, Dodge K, Yamaya K, Morgan C, Rebeyka IM. Pyruvate augments mechanical function via activation of the pyruvate dehydrogenase complex in reperfused ischemic immature rabbit hearts. J Surg Res. 1998;79:164 –169. 23. White LT, O’Donnell JM, Griffin J, Lewandowski ED. Cytosolic redox state mediates postischemic response to pyruvate dehydrogenase stimulation. Am J Physiol. 1999;277:H626 –34. 24. Bolli R, Marban E. Molecular and cellular mechanisms of myocardial stunning. Physiol Rev. 1999;79:609 – 634. 25. Droge W. Free radicals in the physiological control of cell function. Physiol Rev. 2002;82:47–95. 26. Klatt P, Lamas S. Regulation of protein function by S-glutathiolation in response to oxidative and nitrosative stress. Eur J Biochem. 2000;267: 4928 – 4944. 27. Rhee SG. Redox signaling: hydrogen peroxide as intracellular messenger. Exp Mol Med. 1999;31:53–59. 28. Chu F, Ward NE, O’Brian CA. Potent inactivation of representative members of each PKC isozyme subfamily and PKD via S-thiolation by the tumor-promotion/progression antagonist glutathione but not by its precursor cysteine. Carcinogenesis. 2001;22:1221–1229. 29. Domenicotti C, Marengo B, Nitti M, Verzola D, Garibotto G, Cottalasso D, Poli G, Melloni E, Pronzato MA, Marinari UM. A novel role of protein kinase C-delta in cell signaling triggered by glutathione depletion. Biochem Pharmacol. 2003;66:1521–1526. 30. Domenicotti C, Marengo B, Verzola D, Garibotto G, Traverso N, Patriarca S, Maloberti G, Cottalasso D, Poli G, Passalacqua M, Melloni E, Pronzato MA, Marinari UM. Role of PKC-delta activity in glutathione-depleted neuroblastoma cells. Free Radic Biol Med. 2003; 35:504 –516. 31. Konishi H, Tanaka M, Takemura Y, Matsuzaki H, Ono Y, Kikkawa U, Nishizuka Y. Activation of protein kinase C by tyrosine phosphorylation in response to H2O2. Proc Natl Acad Sci U S A. 1997;94: 11233–11237. 32. Konishi H, Yamauchi E, Taniguchi H, Yamamoto T, Matsuzaki H, Takemura Y, Ohmae K, Kikkawa U, Nishizuka Y. Phosphorylation sites of protein kinase C delta in H2O2-treated cells and its activation by tyrosine kinase in vitro. Proc Natl Acad Sci U S A. 2001;98:6587– 6592. 33. Majumder PK, Mishra NC, Sun X, Bharti A, Kharbanda S, Saxena S, Kufe D. Targeting of protein kinase C delta to mitochondria in the oxidative stress response. Cell Growth Differ. 2001;12:465– 470. 34. Majumder PK, Pandey P, Sun X, Cheng K, Datta R, Saxena S, Kharbanda S, Kufe D. Mitochondrial translocation of protein kinase C delta in phorbol ester-induced cytochrome c release and apoptosis. J Biol Chem. 2000;275:21793–21796. 35. Yamamoto T, Matsuzaki H, Konishi H, Ono Y, Kikkawa U. H(2)O(2)induced tyrosine phosphorylation of protein kinase cdelta by a mechanism independent of inhibition of protein-tyrosine phosphatase in CHO and COS-7 cells. Biochem Biophys Res Commun. 2000;273: 960 –966. 36. Chen L, Hahn H, Wu G, Chen CH, Liron T, Schechtman D, Cavallaro G, Banci L, Guo Y, Bolli R, Dorn GW 2nd, Mochly-Rosen D. Opposing cardioprotective actions and parallel hypertrophic effects of delta PKC and epsilon PKC. Proc Natl Acad Sci U S A. 2001;98: 11114 –11119. 37. Hahn HS, Yussman MG, Toyokawa T, Marreez Y, Barrett TJ, Hilty KC, Osinska H, Robbins J, Dorn, GW 2nd. Ischemic protection and myofibrillar cardiomyopathy: dose-dependent effects of in vivo deltaPKC inhibition. Circ Res. 2002;91:741–748. 38. Inagaki K, Chen L, Ikeno F, Lee FH, Imahashi K, Bouley DM, Rezaee M, Yock PG, Murphy E, Mochly-Rosen D. Inhibition of delta-protein kinase C protects against reperfusion injury of the ischemic heart in vivo. Circulation. 2003;108:2304 –2307. 39. Inagaki K, Hahn HS, Dorn GW 2nd, and Mochly-Rosen D. Additive protection of the ischemic heart ex vivo by combined treatment with delta-protein kinase C inhibitor and epsilon-protein kinase C activator. Circulation. 2003;108:869 – 875.

Churchill et al

␦PKC Maintains PDH Inhibition During Reperfusion

40. Murriel CL, Churchill E, Inagaki K, Szweda LI, Mochly-Rosen D. Protein kinase Cdelta activation induces apoptosis in response to cardiac ischemia and reperfusion damage: a mechanism involving BAD and the mitochondria. J Biol Chem. 2004;279:47985– 47991. 41. Nulton-Persson AC, and Szweda LI. Modulation of mitochondrial function by hydrogen peroxide. J Biol Chem. 2001;276:23357–23361. 42. Schechtman D, Murriel C, Bright R, Mochly-Rosen D. Overlay method for detecting protein-protein interactions. Methods Mol Biol. 2003;233:351–357. 43. Csukai M, Mochly-Rosen D. Molecular genetic approaches. II. Expression-interaction cloning. Methods Mol Biol. 1998;88:133–139. 44. Mochly-Rosen D, Koshland, DE Jr. Domain structure and phosphorylation of protein kinase C. J Biol Chem. 1987;262:2291–2297. 45. Stepp LR, Pettit FH, Yeaman SJ, Reed LJ. Purification and properties of pyruvate dehydrogenase kinase from bovine kidney. J Biol Chem. 1983; 258:9454 –9458. 46. Davis PF, Pettit FH, Reed LJ. Peptides derived from pyruvate dehydrogenase as substrates for pyruvate dehydrogenase kinase and phosphatase. Biochem Biophys Res Commun. 1977;75:541–549. 47. Mann WR, Dragland CJ, Vinluan CC, Vedananda TR, Bell PA, Aicher TD. Diverse mechanisms of inhibition of pyruvate dehydrogenase kinase by structurally distinct inhibitors. Biochim Biophys Acta. 2000;1480:283–292. 48. Souroujon MC, Mochly-Rosen D. Peptide modulators of protein-protein interactions in intracellular signaling. Nat Biotechnol. 1998;16:919 –924. 49. Caruso M, Maitan MA, Bifulco G, Miele C, Vigliotta G, Oriente F, Formisano P, Beguinot F. Activation and mitochondrial translocation of

50.

51.

52.

53.

54.

55.

85

protein kinase Cdelta are necessary for insulin stimulation of pyruvate dehydrogenase complex activity in muscle and liver cells. J Biol Chem. 2001;276:45088 – 45097. Barrett WC, DeGnore JP, Keng YF, Zhang ZY, Yim MB, Chock PB. Roles of superoxide radical anion in signal transduction mediated by reversible regulation of protein-tyrosine phosphatase 1B. J Biol Chem. 1999;274:34543–34546. Samikkannu T, Chen CH, Yih LH, Wang AS, Lin SY, Chen TC, Jan KY. Reactive oxygen species are involved in arsenic trioxide inhibition of pyruvate dehydrogenase activity. Chem Res Toxicol. 2003;16:409 – 414. Tabatabaie T, Potts JD, Floyd RA. Reactive oxygen species-mediated inactivation of pyruvate dehydrogenase. Arch Biochem Biophys. 1996; 336:290 –296. Vlessis AA, Muller P, Bartos D, Trunkey D. Mechanism of peroxideinduced cellular injury in cultured adult cardiac myocytes. Faseb J. 1991;5:2600 –2605. Humphries KM, Szweda LI. Selective inactivation of alpha-ketoglutarate dehydrogenase and pyruvate dehydrogenase: reaction of lipoic acid with 4-hydroxy-2-nonenal. Biochemistry. 1998;37:15835–15841. Vance CL, Begg CM, Lee WL, Haase H, Copeland TD, McEnery MW. Differential expression and association of calcium channel alpha1B and beta subunits during rat brain ontogeny. J Biol Chem. 1998;273: 14495–14502.