HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) All Versions of this Article: 293/4/H2219    most recent 01306.2006v1 Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in ISI Web of Science Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Guo, D. Articles by Johnson, J. A. Search for Related Content PubMed PubMed Citation Articles by Guo, D. Articles by Johnson, J. A.

Protein kinase C- coimmunoprecipitates with cytochrome oxidase subunit IV and is associated with improved cytochrome-c oxidase activity and cardioprotection

¹Department of Pharmacology and Toxicology, School of Medicine, and ²The Program in Regenerative Medicine, Institute of Molecular Medicine and Genetics, Medical College of Georgia, Augusta, Georgia

Submitted 29 November 2006 ; accepted in final form 8 July 2007

	ABSTRACT

TOP ABSTRACT METHODS RESULTS DISCUSSION REFERENCES

 
We have utilized an in situ rat coronary ligation model to establish a PKC- cytochrome oxidase subunit IV (COIV) coimmunoprecipitation in myocardium exposed to ischemic preconditioning (PC). Ischemia-reperfusion (I/R) damage and PC protection were confirmed using tetrazolium-based staining methods and serum levels of cardiac troponin I. Homogenates prepared from the regions at risk (RAR) and not at risk (RNAR) for I/R injury were fractionated into cell-soluble (S), 600 g low-speed centrifugation (L), Percoll/Optiprep density gradient-purified mitochondrial (M), and 100,000 g particulate (P) fractions. COIV immunoreactivity and cytochrome-c oxidase activity measurements estimated the percentages of cellular mitochondria in S, L, M, and P fractions to be 0, 55, 29, and 16%, respectively. We observed 18, 3, and 3% of PKC-, -, and - isozymes in the M fraction under basal conditions. Following PC, we observed a 61% increase in PKC- levels in the RAR M fraction compared with the RNAR M fraction. In RAR mitochondria, we also observed a 2.8-fold increase in PKC- serine 729 phosphoimmunoreactivity (autophosphorylation), indicating the presence of activated PKC- in mitochondria following PC. PC administered before prolonged I/R induced a 1.9-fold increase in the coimmunoprecipitation of COIV, with anti-PKC- antisera and a twofold enhancement of cytochrome-c oxidase activity. Our results suggest that PKC- may interact with COIV as a component of the cardioprotection in PC. Induction of this interaction may provide a novel therapeutic target for protecting the heart from I/R damage.

cardiac; ischemia-reperfusion; protein kinase C; oxidative phosphorylation; coronary ligation; mitochondria

Numerous studies have implicated individual PKC isozymes in PC and I/R (1, 2, 6, 9, 13–15, 26, 42, 45, 47, 48, 51–53, 55). Recently, mitochondrial mechanisms involving PKC isozymes have received considerable attention (1, 2, 8, 10, 36, 42). Most studies of PC indicate that the PKC- isozyme is a central player, but there have also been reports of other PKC isozymes contributing to cardiac PC and I/R. For example, Wang and colleagues (58) used confocal analyses in tissue sections taken from Langendorf hearts to colocalize PKC- immunofluorescence and mitochondrial staining by tetramethyl rhodamine ethyl ester following diazoxide-induced PC. In a series of papers by Mayr et al., it was reported that PKC- knockout (KO) mice demonstrated decreased glycolysis and an increased lipid metabolism under baseline conditions (33) and were unable to demonstrate a PC response (34). In contrast, Mochly-Rosen and colleagues demonstrated that inhibition of PKC- with a peptide-based PKC- translocation inhibitor reduced I/R injury in transgenic mice (21) and isolated heart preparations (21, 37). PKC- has also been reported to induce pathological hypertrophy and cardiac apoptosis involving translocation of the PKC- isozyme to mitochondria and interaction with the proapoptotic protein Bad (36). Recently, the PKC- isozyme has been proposed to delay the reactivation of pyruvate dehydrogenase following I/R injury, which slows the resupply of acetyl coenzyme A to the Krebs cycle (8). Ping and coworkers reported that the PKC- isozyme forms signaling clusters between c-Src and MAPK inside cardiac mitochondria (2, 26, 49, 55). Baines et al. (1) showed that the PKC- isozyme translocated to cardiac mitochondria and bound to the mitochondrial permeability transition pore to inhibit its function in cardiac PC. Costa and coworkers (10) identified a role for the PKC- isozyme in PKG-mediated opening of the mitochondrial ATP-sensitive K⁺ channel in PC cardioprotection.

A primary goal of this study was to determine which PKC isozymes existed in cardiac mitochondria before and after cardiac PC and I/R injury. It has been known for 30 years that the heart has two primary populations of mitochondria: those that exist just below the sarcolemmal membrane [subsarcolemmal mitochondria (SSM)], and those that are deeply imbedded in the myofibrils [interfibrillar mitochondria (IFM)] (44). Each subgroup of mitochondria can utilize the same substrates for energy production, but some differences in enzymatic activities have been reported. For example, SSM appear to have higher carnitine palmitoylase and -glycerophosphate dehydrogenase activities than IFMs. Conversely, IFMs display greater specific activities of succinate dehydrogenase and citrate synthase (44). The SSM are easier to extract and have been more widely studied than IFM, which generally require protease treatment of heart tissue during tissue homogenization and even then are usually incompletely liberated. Reflective of the difficulties in isolating IFM, very few cardiac PC or I/R injury studies even consider them (7, 24, 25). We, therefore, used subcellular fractionation studies, which clearly demonstrate that IFM are present in high abundance and necessitate further study with regard to the mitochondrial roles of PKC isozymes in PC and I/R.

Cytochrome oxidase (CO) is a 13-subunit enzyme complex comprising the final step in the ETC. It is intricately linked to maintenance of the mitochondrial proton gradient and the production of ATP (54). Our laboratory recently demonstrated in neonatal cardiac myocytes that the PKC- isozyme interacts with the number IV subunit of cytochrome-c oxidase (COIV) to enhance CO activity during hypoxic PC (42). In this study, we present evidence for the existence of this mechanism in an in situ rat model of cardiac PC. We hypothesize that, following reperfusion, this protective response allows a more rapid recovery of ATP production (20), reduced reactive oxygen species (ROS) production, and is responsible in part for smaller infarctions. These studies identify another mechanism of protection for PKC- and suggest additional therapeutic opportunities for attenuating I/R injury.

	METHODS

TOP ABSTRACT METHODS RESULTS DISCUSSION REFERENCES

 
Myocardial PC and ischemia and reperfusion. Rats were anesthetized with an intraperitoneal injection of ketamine HCl (100 mg/kg) and xylazine (10 mg/kg). The right jugular vein was cannulated for the delivery of saline. The left carotid artery was cannulated for the measurement of blood pressure and heart rate using a transducer connected to a polygraph. The trachea was then cannulated and connected to a rodent ventilator (model 683, Harvard Apparatus, South Natick, MA). Rats were ventilated at 65 breaths/min with room air supplemented with O₂. Atelectasis was prevented by maintaining a positive end-expiratory pressure of 10 cmH₂O. Body temperature was maintained at 37°C using a heating pad. Once heart rate and blood pressure stabilized, a left thoracotomy was performed at the fifth intercostal space. A pericardiotomy was then performed, and a 4-0 silk suture was passed below the left descending coronary artery close to the origin immediately below the left atrial appendage to the right portion of the left ventricle. The ends of the suture were threaded through a polyethylene tube to form a snare, and then the snare was clamped onto the epicardial surface using a hemostat. For PC, two cycles of a 5-min occlusion followed by a 5-min reperfusion were induced. For I/R injury, the occlusion was elicited for 30 min by pulling on the snare. Next, the snare was released for a 120-min reperfusion period. Coronary artery occlusion was confirmed by epicardial cyanosis, decrease in blood pressure, and eventually by infarct staining methods (see below).

All experimental protocols involving the use of animals were approved by The Medical College of Georgia Institutional Animal Care and Use Committee and conformed with The Helsinki Agreement for the humane care and use of laboratory animals.

Determination of infarct size. Left ventricle infarct size and region at risk (RAR) were determined as described previously (4, 31, 32). After 120 min of reperfusion, the coronary artery was again occluded. The RAR was determined by a lack of staining with Evans blue dye (2%), which was injected into the left ventricular cavity and allowed to perfuse the left ventricle. The entire heart was excised, the atria and great blood vessels of the heart were removed, remaining tissue was rinsed of excess Evans blue dye, and left ventricular tissue was cut into 2-mm transverse sections. Slices were incubated in a 1% solution of 2,3,5-triphenyltetrazolium chloride (TTC) in PBS buffer for 12 min to stain viable myocardium brick red. Infarcted tissue was not stained and appeared lighter or white in color. The slices were then fixed in 10% formalin for 24 h and photographed. The ischemic RAR (unstained by Evans blue dye) and the infarcted area (unstained by TTC) were measured using Metamorph Image software. Infarct size was expressed as a percentage of the RAR for I/R damage. We used Evans blue to identify the RAR and regions not at risk (RNAR) in all experiments. However, separate hearts were used for TTC staining and biochemical analyses. For this reason, we also confirmed myocardial infarction and PC using a second marker [cardiac troponin I (cTnI) release] in all experiments (see below).

Quantitation of cTnI release. cTnI release was monitored using the high-sensitivity microplate spectrophotometric assay kit optimized for rat cTnI (Life Diagnostics, West Chester, PA). Serum samples were taken from rats before and after PC, I/R, and PC + I/R exposures and processed according to the manufacturer's instructions. Briefly, the assay involves immobilized cTnI antisera on a microplate and a second fluorescent antibody, which binds cTnI captured on the microplate. Analyses included a standard cTnI curve, and all serum samples were read in triplicate within the linear range of detection of the assay.

Subcellular fractionation. Unless otherwise indicated, all steps were conducted at 4°C or on ice using chilled buffers. The left ventricles were isolated, and 2 mm pieces of tissue from the RAR and RNAR for I/R injury were incubated in 4-ml isotonic MSE buffer [10 mM Tris·HCl, pH 7.5, 220 mM mannitol, 70 mM sucrose, 1 mM EGTA, 0.025% fatty acid-free bovine serum albumin (BSA), 1.6 mM carnitine, 2 mM taurine, and 10 µg/ml each of aprotinin, leupeptin, and phenylmethylsulfonyl fluoride] containing 0.25 mg/ml of trypsin (2,550 U/ml). Next, BSA was added to a final concentration of 10 mg/ml to terminate trypsin proteolysis, and the tissue was washed twice with MSE without trypsin. The resulting tissue was subjected to homogenization using a motorized homogenizer (30 strokes). The homogenate was centrifuged twice at 600 g to obtain a low-speed pellet. The two 600 g pellets were combined and resuspended in a total volume of 1.5 ml to produce the low-speed centrifugation (L) fraction in Figs. 4–7. The supernatant from this spin was then subjected to a 12,000 g centrifugation to obtain a crude mitochondrial fraction pellet. This pellet was resuspended in 2.2 ml of MSE buffer and loaded onto a Percoll/Optiprep gradient (described below) to obtain a highly purified mitochondrial (M) fraction. The 12,000 g supernatant was then subjected to a 100,000 g spin for 20 min. The supernatant from this spin was used as the cell-soluble (S) fraction, and the pellet was resuspended in 0.45 ml MSE buffer and used as the crude particulate (P) fraction. One hundred fifty micrograms of each fraction were then subjected to SDS-PAGE and Western blot analysis, as described below.

View larger version (21K): [in this window] [in a new window]   Fig. 1. Ischemic preconditioning (PC) reduces infarct size induced by ischemia-reperfusion injury. Sprague-Dawley rats were exposed to mock surgery without coronary ligation (control), ischemic PC (2 cycles of 5-min ischemia followed by 5-min reperfusion) only, 30 min of ischemia followed by 120 min of reperfusion (I/R), or PC + I/R, as described in METHODS. The region at risk (RAR) and infarct size were established using standard Evans blue and tetrazolium staining [2,3,5-triphenyltetrazolium chloride (TTC)] techniques (4, 31, 32) and methods. Shown are the infarct sizes as a percentage of the RAR for each corresponding group. Representative sections are shown from control (A), PC (B), I/R (C), and PC + I/R (D) groups. E: mean ± SE of tetrazolium staining results (open bars, left axis) and a second marker of infarction cardiac troponin I (cTnI) levels in rat serum (hatched bars, right axis) for control (Con), PC, I/R, and PC + I/R groups; n = 4 (tetrazolium) or 8 (cTnI) animals in each experimental group.

View larger version (25K): [in this window] [in a new window]   Fig. 2. Detection of mitochondria in subcellular fractions by cytochrome-c oxidase subunit IV (COIV) immunoreactivity and cytochrome-c oxidase activity. Untreated control rats were subjected to anesthesia, sham surgery, and removal of hearts, as described in METHODS. Tissue from the cardiac left ventricle was then homogenized and fractionated into cell-soluble (S), low-speed (600 g) centrifugation (L), Percoll/Optiprep purified mitochondrial (M), and 100,000 g particulate (P) fractions, as described in METHODS. A: each fraction (150 µg) was subjected to Western blot analyses using anti-COIV antisera. Detection of COIV served as a preliminary indicator of interfibrillar mitochondria remaining in the L fraction and the P fraction following extraction of the mitochondria recovered in the M fraction. B: to confirm that COIV immunoreactivity represented intact, viable mitochondria, we assayed cytochrome-c oxidase activity in each fraction in the presence (open bars) and absence (solid bars) of the detergent n-dodecyl-,D-maltoside. All cytochrome-c oxidase activities were completely inhibited by inclusion of 1 mM KCN in the assay (right side). Results shown are means ± SE from a single experiment representative of 4 independent experiments, each performed in triplicate and taken from a different rat heart (ventricle).

View larger version (35K): [in this window] [in a new window]   Fig. 3. Purity of mitochondrial fractions. Adult rat cardiac S, L, M, and P fractions were isolated from the left ventricle of Sprague-Dawley rats, as described in Fig. 2. Western blot analyses were then conducted using antisera directed against proteins known to exist in the golgi apparatus (furin), plasma membrane (cadherans), endosomes (EEA1 = early endosomal antigen), and mitochondria (COIV = the number IV subunit of cytochrome oxidase). Lanes 1–2 were included as positive controls. Lane 1 contains 150 µg of 100,000 g particulate fraction protein isolated from neonatal cardiac myocytes (NCM), and lane 2 contains 50 µg of Hela cell total homogenate protein. As is shown, only immunoreactivity to COIV was found in our Percoll/Opti-prep gradient purified mitochondria (M fraction). Results are means ± SE from 3 independent mitochondria preparations.

View larger version (41K): [in this window] [in a new window]   Fig. 4. Basal cardiac mitochondrial distributions of PKC isozymes. Sprague-Dawley rats were anesthetized and subjected to mock surgery without coronary ligations, as described in Fig. 1. Since there was no left ventricular RAR for ischemia, comparable anatomical tissue regions were taken from control group as in the PC and I/R hearts used in Figs. 5–8. Tissue was homogenized and S, L, M, and P fractions were isolated. Each fraction (150 µg) was subjected to SDS-PAGE and Western blot analysis with PKC isozyme-selective antisera. Typical autoradiographs for each PKC isozyme are shown in the top portion of each histogram. Histograms represent mean ± SE data from 5 animals.

View larger version (27K): [in this window] [in a new window]   Fig. 7. Mitochondrial PKC- (A), - (B), and - (C) isozymes are present in mitochondria following I/R injury. All conditions and analyses are similar to Fig. 4, except I/R alone was the treatment instead of control (mock surgery). Ischemia was for 30 min, and reperfusion was conducted for 120 min, as described in METHODS. All other details are as in Fig. 5.

Immunoprecipitations. Following isolation, mitochondria were solublized in immunoprecipitation (IP) buffer [10 mM Tris·HCl, pH 7.4, 0.125% (wt/vol) BSA, 5 mM EDTA, 5 mM EGTA, 5 mM Na₄P₂O₇, 1% (vol/vol) Triton X-100, 10 nM calyculin A, 20 µg/ml each of phenylmethylsulphonyl fluoride, leupeptin, aprotinin, and soybean trypsin inhibitor]. Samples were then subjected to standard IP procedures (17). All subsequent steps were conducted at or below 4°C. Briefly, stopped reactions were placed on ice for three 5-min incubations vortexing between each incubation. They were then subjected to a 600 g centrifugation to precipitate detergent-insoluble material. Supernatants were then incubated with primary antisera (anti-CO subunits, Molecular Probes, or anti-PKC isozymes, BD Transduction Laboratories) coupled to Bio-Rad Affi-gel for 4 h overnight. Control experiments using Affi-gel alone (without coupled antisera) yielded no IP of CO subunits or PKC isozymes. Immune complexes were collected by centrifugation, and the resulting Affi-gel pellets were washed three times with wash buffer (10 mM Tris·HCl, pH 7.4, 1 mM EDTA, 1 mM EGTA, 5 mM Na₄P₂O₇, 10 nM calyculin A, 20 µg/ml each of phenylmethylsulphonyl fluoride, leupeptin, aprotinin, and soybean trypsin inhibitor, and 1% Triton X-100). Proteins were liberated from protein A agarose by heating in Laemmli sample buffer at 85°C for 10 min, with vortexing every 3 min. Samples were then subjected to SDS-PAGE on 13.5% acrylamide gels and transferred onto nitrocellulose paper using standard Western blot techniques. The resulting blots were probed for COIV or PKC isozymes using ¹²⁵I-protein A detection, as previously described (22, 23, 41, 42).

CO activity assays. Intact mitochondria were subjected to the spectrophotometric CO assay, according to manufacturer's instructions (Sigma Chemical) (41, 42). The kit measures the oxidation of ferrocytochrome c to ferricytochrome c via the activity of CO. CO activity is, therefore, monitored as a decrease in absorbance at 550 nM. Aliquots from the S, L, M, and P fractions were incubated in assay buffer (10 mM Tris·HCl, pH 7.0, 120 mM KCl) in the absence and presence of 1 mM n-dodecyl-,D-maltoside (D--M). We observed no measurable activity in the 100,000 g S fractions in any of our assays (see Fig. 2B). As specified in the manufacturer's instructions, activity was maximal during the first minute of the assay, and only minor sustained activity remains after that time. We, therefore, reported our results using the initial reaction rates (1 min). Approximately 95% of the CO activity was found in the 1 mM D--M solublized mitochondrial pellet. The average activities observed were 710 ± 88 mU·min^–1·mg protein^–1, where 1 unit is the amount of enzyme required to oxidize 1 µM of ferrocytochrome c per minute at pH 7.0 at 25°C. The kit comes with a purified CO standard, which allows determination of the linear range of CO activity. All assays were conducted in triplicate within the linear range of the assay. On average, we found 6 ± 2% of the total CO activity in non-D--M solubilized mitochondrial fractions, and this percentage was similar between treatment groups. We, therefore, believe that, using the current isolation methods, most mitochondria have an intact outer mitochondrial membrane, and that differences in CO activities between different treatment groups did not reflect differences in outer mitochondrial membrane integrity. Equivalent amounts of protein for each cell fraction or mitochondrial extract were assayed.

Western blot analyses. Samples were subjected to SDS-PAGE on 12–13.5% acrylamide gels and transferred onto nitrocellulose paper. The resulting blots were probed for PKC isozymes, COIV, or other proteins using ¹²⁵I-protein A detection, as previously described (22, 23). PKC antisera were from BD Transduction Laboratories and were used at 1:300 dilutions. Sources/dilutions for other antisera are as follows: furin (Santa Cruz Antibody, 1:300), pan-cadherans (Sigma, 1:400), early endosomal antigen (Abcam, 1:1,000), retinoblastoma (Santa Cruz, 1:500), and COIV (Invitrogen, 1:500), and phospho-PKC- serine 729 and phospho-PKC- threonine 507 antisera (Santa Cruz Antibody, 1:300).

Statistical analyses. Statistical significance was tested using the Student's T-test. A P value of 0.05 was considered significant.

	RESULTS

TOP ABSTRACT METHODS RESULTS DISCUSSION REFERENCES

 
Baseline hemodynamic parameters. Adult Sprague-Dawley rats were subjected to control, ischemic PC, 30-min ischemia/120-min reperfusion (I30/R120), or PC + I30/R120 using standard coronary ligation techniques, as described in METHODS. Table 1 summarizes data demonstrating that rats from each experimental group had comparable body and heart weights, heart rate, and mean systolic blood pressure at the initiation of experiments. We observed no statistically significant differences in any of these parameters between experimental treatment groups.

Distribution of mitochondria in subcellular fractions. It is well known that there are two major pools of mitochondria in the myocardium, SSM and IFM (44). Historically, SSM have been used more frequently due to their ease of isolation, but a substantial number of IFM exist. In Fig. 2A, we obtained cardiac left ventricular tissue from untreated control rats using anatomic regions of the myocardium that were similar to those used for our PC and infarct studies. Tissue was homogenized and fractionated into S, L, M, and P fractions, as described in METHODS. Protein (150 µg) from each fraction was then subjected to SDS-PAGE and Western blot analysis with antisera directed against the COIV. Since COIV is a known mitochondrial marker protein, we used its immunoreactivity as an initial, crude estimate of the percentage of total cellular mitochondria in each of our subcellular fractions. As predicted, we found no COIV in the cell-soluble (predominantly cytosolic) fraction (Fig. 2A). Conversely, when 150 µg of protein were used for each fraction, we observed 12 ± 2, 45 ± 2, and 43 ± 3% of the total mitochondrial marker (COIV) immunoreactivity in the L, M, and P fractions. This was well supported by our assays of CO activity (conducted using equivalent concentrations of protein for each fraction) shown in Fig. 2 and Table 2. As with all of our Western blot analyses, we used 150 µg of protein in each fraction to maximize detection of antigens, which may exist in low abundance. However, this was not representative of the in situ ratios of protein found in these fractions. On average, we determined the percentage of total homogenate protein found in the S, L, M, and P fractions to be 18.8 ± 1.0, 66.8 ± 1.1, 9.4 ± 0.5, and 5.4 ± 0.3 (n = 11), respectively. Establishment of these ratios was necessary to estimate (based on COIV immunoreactivity and CO activity measurements) the percentage of mitochondria in each fraction. When we normalized the data shown in Fig. 2 to account for the actual percentages of homogenate protein present in the S, L, M, and P fractions, we estimated the percentage of total mitochondria in these fractions to be 0, 55, 29, and 16%, respectively. These findings are significant as the L and P fractions are often discarded when evaluating mitochondrial roles in PC and I/R, which means that, in most studies to date, less than one-third of the total mitochondrial pool was evaluated. In addition, the L fraction has been considered to be predominantly unlysed cells, nuclei, and broken cell debris, and the 100,000 g spin has been considered to be enriched in plasma membrane proteins. Our results clearly show that there are substantial numbers of mitochondria in each of these fractions (even after trypsin exposures and homogenization), which should be considered in interpretation of experimental results using these types of subcellular fractionation protocols in myocardium. Our study is the first to characterize this in situ rat model of PC + I/R using this level of subcellular fractionation in analyses of mitochondrial PKC isozymes. While this study focuses on PKC isozymes in SSM, it also reveals the need for future work targeted at the roles of PKC isozymes in IFM.

Purity of mitochondrial fractions. Since individual PKC isozymes have been reported to translocate to many nonmitochondrial cell structures, some of which can contaminate mitochondrial preparations, we felt it was important to determine the level of contamination present from other subcellular fractions in the M fraction. Therefore, before initiating biochemical studies to correlate changes in mitochondrial PKC isozymes with PC or I/R injury, we first confirmed the purity of mitochondria isolated from Percoll/Optiprep density gradients using antibodies directed against known subcellular marker proteins (Fig. 3). Protein (150 µg) from S, L, M, and P fractions were analyzed in Western blot analyses for the presence of cell compartment-specific antigens (Fig. 3). Hela cell homogenates and particulate fractions from neonatal cardiac myocytes were used as positive controls (Fig. 3, left). The only positive immunoreactivity observed in our Percoll/Optiprep-purified M fraction was for the mitochondrial inner membrane protein COIV (Fig. 3). Despite positive immunoreactivity in controls (Fig. 3, left), antisera directed against known marker proteins for the plasma membrane (cadherans, 120–130 kDa), golgi (Furin 60 kDa), endosomes (EEA1, 162 kDa), and nuclei (Rb protein, 110–116 kDa, not shown) consistently failed to yield detectable signals in Western blots of the M fraction, confirming a high degree of mitochondrial purity (Fig. 3).

Mitochondrial distributions of PKC isozymes under basal conditions in rat cardiac ventricle. There have been numerous studies implicating individual PKC isozymes in mitochondrial responses related to cardiac PC (1, 2, 10, 33, 34, 37, 42, 58, 60) and I/R injury (8, 36, 37), yet there has not been a comprehensive analysis in in situ rat ventricle monitoring all PKC family members simultaneously in gradient-purified mitochondria under these conditions. We felt this was crucial for identifying which PKC isozymes should be the focus of future studies, for investigating potential opposing relationships between individual PKC isozymes, and for accurately determining changes in the mitochondrial distribution of individual PKC isozymes following PC and I/R treatments. We first determined the baseline mitochondrial distributions of PKC isozymes in the S, L, M, and P fractions (Fig. 4). Proteins (150 µg) from S, L, M, and P fractions were subjected to SDS-PAGE and electrotransfer onto nitrocellulose paper. These blots were then probed with antisera directed against the PKC-, -, -, -, -, -, -, and - isozymes using ¹²⁵I-protein A detection, as previously described (22, 23, 41, 42). In analyses conducted with 150 µg of protein in each (S, L, M, and P) fraction, the PKC- isozyme was found predominantly in the S fraction under basal conditions, with virtually no detectable levels in the M fraction, and on average only 17 ± 8% of PKC- was found in the combined L and P fractions (Fig. 4). Using antisera capable of recognizing either PKC-_I or -_II, we found only trace levels of the PKC- isozymes, all of which existed in the S fraction (not shown). We also did not detect the PKC- isozyme in any of our cell fractions (not shown). With the use of 150 µg of protein in each fraction, the PKC- isozyme was determined to exist 25 ± 1% in the S fraction, 21 ± 1 in the L fraction, 15 ± 3% in the M fraction, and 39 ± 5% in the P fraction. If we normalize these values based on the percentage of total homogenate protein found in each fraction, we obtain 19.6, 58.8, 17.6, and 4.0% of PKC- in the S, L, M, and P fractions, respectively. These results confirmed a clear mitochondrial distribution of the PKC- isozyme in cardiac mitochondria under basal conditions.

PKC- was also detected in cardiac mitochondria under basal conditions, confirming results originally reported by Baines and coworkers in mouse hearts (1, 2) using nongradient purified mitochondria preparations. In analyses using 150 µg of protein in each (S, L, M, and P) fraction, the relative proportions of this enzyme were 43 ± 1 (S), 3 ± 0 (L), 9 ± 4 (M), and 33 ± 9% (P). When we normalized our data to correct for the percentage of total homogenate protein in each fraction (as was done for Fig. 2), the percentage of PKC- in each fraction was found to be 71.8 (S), 17.8 (L), 3.3 (M), and 7.1% (P). It is important to note that, had we not conducted these Western blot analyses with 150 µg of protein, we would not have been able to detect the low levels of PKC- in the M fraction.

In Western blots utilizing 150 µg of protein in each fraction, the PKC- isozyme was also found to be present in mitochondria under basal conditions, with a distribution of 55 ± 6, 21 ± 3, 18 ± 7, and 6 ± 3% in the S, L, M, and P fractions, respectively. When these values were corrected for the percentage of homogenate protein found in each fraction, the values were 40.8 (S), 55.6 (L), 3 (M), and 0.6% (P). In analyses using 150 µg of protein in the S, L, M, and P fractions, the PKC- and - isozymes were present mostly in the S fraction (65 ± 3%), with only 21 ± 2% in the L and P fractions. We observed no mitochondrial distributions of the PKC- and - isozymes. Our results clearly demonstrate mitochondrial prelocalizations for the PKC-, -, and - isozymes under basal conditions (Fig. 4). It is highly likely that a significant number of mitochondria in the L and P fractions also contain PKC-, -, and -, indicating that the levels of these isozymes in the M fraction are underestimates of total cellular mitochondrial levels, possibly accounting for less than one-third of the total mitochondrial level of each isozyme. However, further study will be necessary to determine whether the actual percentages of individual PKC isozymes found in SSM differs from that observed in IFM.

Mitochondrial PKC- levels are increased following cardiac PC. In Fig. 5, adult Srague-Dawley rats were exposed to PC, and the myocardial RAR and RNAR for I/R injury were determined as in METHODS. These tissues were then isolated and fractionated as in Fig. 4. Because the percentages of individual PKC isozymes found in our M fractions ranged from only 3 to 18% of the total cellular levels of each isozymes, it was in some cases difficult to identify significant movements of isozymes into or out of mitochondria following PC. This is most apparent in data for the PKC- isozyme. For example, in one experiment, we observed what appeared to be a modest increase in mitochondrial PKC- in the RAR following PC; however, on average, this difference was small and variable, and in most experiments we found no significant PKC- levels inside mitochondria. Therefore, while translocation of this PKC isozyme to mitochondria was occasionally observed, it was not reproducible and could not be correlated with the induction of PC. However, when expressing PKC- levels as a ratio of PKC- in the RAR M fraction to that found in the RNAR M fraction, we did observe a 61 ± 8% enhancement following PC (Fig. 5B, P < 0.01). It is important to note that, since mitochondria are also present in the L and P fractions, it is likely that the total level of PKC- in mitochondria would be higher than our estimates based solely on the M fraction. We hypothesize that both the translocated PKC- and the prepositioned mitochondrial PKC- play roles in PC.

View larger version (29K): [in this window] [in a new window]   Fig. 5. PKC- (A), - (B), and - (C) isozymes are present in mitochondria following cardiac ischemic PC. Sprague-Dawley rats were anesthetized and subjected to brief left anterior descending coronary artery ligations to induce PC. The left ventricular RAR was determined, and tissue was isolated from the RAR and regions not at risk (RNAR) (METHODS). All other experimental details are as in Fig. 4. Results are presented in histograms as the mean + SE ratios of densitometric score obtained in the RAR divided by that obtained in the RNAR. Representative autoradiographs are shown, and histograms represent data from 8 (PKC-), 5 (PKC-) and 5 (PKC-) animals.

Activated mitochondrial PKC- increases following PC. Our results in Fig. 5B clearly established enhanced mitochondrial PKC- levels following PC. We, therefore, determined whether mitochondrial PKC- activation [monitored using PKC- phospho-serine 729 immunoreactivity (autophosphorylation)] in Western blots occurred. Figure 6A illustrates an increase in PKC- phospho-serine 729 immunoreactivity in the S, L, and M fractions isolated from the RAR following our in situ PC protocols. When S, L, and M fractions taken from RAR tissue were compared with corresponding RNAR tissue fractions, we observed increases of 2.5 ± 0.1-, 6.9 ± 1.3-, and 2.3 ± 0.2-fold, respectively (Fig. 6A). Of interest, the P fraction showed considerable PKC- phospho-serine 729 immunoreactivity in both the RAR and RNAR, implying substantial PKC- autophosphorylation in that fraction independent of PC. With the exception of the M fraction (which shows greater PKC- accumulation in mitochondria following PC; Fig. 5B), differences in PKC- phospho-serine 729 immunoreactivity could not be related to differences in mitochondrial PKC- isozyme levels. Neither could protein loading on gels explain differences in phospho-serine 729 immunoreactivity, because similar anti-PKC- immunoreactivity (nonphospho-antisera) in the S, L, and P fractions (Fig. 5), similar Ponceau S protein staining of blots (not shown), and near equivalent PKC- phospho-serine 729 immunoreactivity in the P fraction (Fig. 6A) were observed.

View larger version (18K): [in this window] [in a new window]   Fig. 6. Elevated PKC- autophosphorylation is detected in RAR tissue fractions following in situ cardiac PC. All conditions are as in Fig. 5, except phospho-PKC- serine 729 (A) and phospho-PKC- threonine 507 (B) immunoreactivity were monitored using antisera (METHODS). Note that increases in PKC- autophosphorylation (but not PKC- autophosphorylation) were observed in S, L, and M fractions. The P fraction showed considerable PKC- autophosphorylation in the RAR and RNAR.

Mitochondrial PKC isozyme distributions following I/R and PC + I/R. Consistent with previous studies, I30/R120 exposures caused a 58 ± 9% infarction in the myocardial RAR (Fig. 1), which correlated with a substantial loss in the total levels of many PKC isozymes (compare autoradiographs from Figs. 4 vs. 6) due to myocardial cell death (Fig. 1). However, despite a decrease in total mitochondria in tissue subjected to I30/R120, we were able to isolate viable mitochondria from all treatment groups using the Percoll/Optiprep gradients. By using 150 µg of protein in each (S, L, M, and P) fraction (even in I30/R120 alone groups), we could detect changes in mitochondrial PKC isozymes in the M fraction. We observed no changes in the distribution of the PKC- and - isozymes following I30/R120. However, the total levels of the PKC- isozyme declined substantially in cardiac tissue taken from the RAR after I30/R120, but there was no such decline for the PKC- isozyme (not shown). The PKC- isozyme was found to exist in S, L, M, and P fractions after I30/R120, with modest declines noted in the L, M, and P fractions in the RAR tissue vs. RNAR tissue. We observed PKC- present in the M fraction following I30/R120, consistent with it playing a mitochondrial role in myocardial damage, as previously reported (8, 36, 37). When comparing mitochondria isolated from the RAR with those from the RNAR, we found only a 23 ± 7% loss in PKC- levels in the M fraction isolated from the RAR after I30/R120 (P < 0.03, n = 6), with virtually no decline in the S, L, and P fractions (Fig. 7A). This PKC- decrease was considerably smaller than corresponding decreases in the PKC- (not shown), -, or - isozymes after I30/R120 (Fig. 7, B and C).

Of interest, the PKC- isozyme was virtually eliminated from the M fraction isolated from the RAR and RNAR for I/R injury following I30/R120 (Fig. 7B). This suggested substantial cardiac loss of PKC- and its cardioprotection in both the RAR and the RNAR. It also appeared, on average, that there was substantially more PKC- in the P fraction in the RNAR (P < 0.02, n = 5). This is consistent with greater infarctions in hearts lacking functioning PKC- (15, 53) and possibly changes in animals exposed to I30/R120, leading to general loss of mitochondrial PKC- throughout the ventricle. To our knowledge, PKC- levels have not been compared previously in gradient-purified mitochondria isolated from RAR vs. RNAR ventricles using this subcellular fractionation protocol following I30/R120. It is important to note that, in the same experiments, other mitochondrial PKC isozymes (e.g., PKC- and -) were downregulated less in the RNAR (Fig. 7, A and C) than PKC- was, arguing against a general "cell death/toxicity" type of response. Unlike PKC-, the mitochondrial levels of the PKC- isozyme were decreased predominantly in the RAR for I/R damage (Fig. 7C). In addition, and consistent with PKC- playing mitochondrial roles in I/R injury, we observed only 23 ± 7% decreases in PKC- levels in the RAR M fraction following prolonged I/R (Fig. 7A). Therefore, prepositioned PKC- may mediate mitochondrial damage in I/R in this model.

In our final analysis of mitochondrial PKC isozyme distributions, we monitored the S, L, M, and P fractions following PC + I30/R120 treatments. As expected, there was generally higher PKC isozyme levels in most cellular fractions compared with levels in the I30/R120-alone group, indicative of a general protective effect of PC (Figs. 7 vs. 8). As with all treatments, we observed virtually no mitochondrial distributions of the PKC- or - isozymes (not shown). While in general PKC- levels decreased in both the RAR and RNAR following PC + I/R, there was no significant difference in PKC- expression between these two regions. The ratio of PKC- found in the RAR to RNAR in the S, L, M, and P fractions following PC + I30/R120 exposures was 1.4 ± 0.3, 0.7 ± 0.3, 0.8 ± 0.3, and 0.8 ± 0.2, respectively.

View larger version (29K): [in this window] [in a new window]   Fig. 8. The PKC- (A), - (B), and - (C) isozymes are present in cardiac mitochondria in rats exposed to PC + I/R. All details are as in Fig. 6, except PC + I/R was the treatment.

Ischemic PC is associated with an PKC--COIV coimmunoprecipitation and preservation of CO activity. Mitochondria were isolated from I30/R120 and PC + I30/R120 groups. Equivalent amounts of mitochondrial protein were incubated on ice in the presence and absence of the detergent D--M and CO activity assayed as in METHODS. We observed <5% of the total CO activity in all treatment groups when mitochondria were assayed in the absence of D--M, which indicated intact mitochondrial membrane structure in the majority of our mitochondria (METHODS). In contrast, in mitochondria incubated in the presence of D--M, we observed 95% of the total CO activity in all treatment groups. In these assays, we found a 2.2 ± 0.1-fold (n = 4) increase in CO activity (when comparing RAR with the RNAR mitochondria) from I/R alone and PC + I/R animals. It is important to note that we assayed only mitochondria that could be collected from our Percoll/Optiprep gradients, and equivalent concentrations of mitochondrial protein were assayed in each group. We expected that, with an 50% infarct in the RAR of I/R-alone treatment groups (see Fig. 1), there would be less CO activity due to cell death in this zone. However, assaying only surviving mitochondria isolated from Percoll/Optiprep density gradients in this group corrects for cell death during I/R and, if anything, may underestimate total mitochondrial damage to CO. Our results indicate that PC administered before prolonged I/R exposures improves CO activity in mitochondria taken from the RAR for I/R damage (Fig. 9A). This suggests that recovery of enhanced CO activity may play a role in PC protection, and elevated CO activities are clearly observed after PC + I30/R120 compared with I30/R120-alone treatment groups. Therefore, by facilitating CO activity, PC administered before I30/R120 should improve the recovery of ATP production and diminish ROS production from the ETC following I30/R120.

View larger version (20K): [in this window] [in a new window]   Fig. 9. Ischemic PC enhances ventricular cytochrome-c oxidase activity and induces an PKC--COIV coimmunoprecipitation. Sprague-Dawley rats were exposed to I/R and PC + I/R, as described in Fig. 1. Left ventricular tissue from the RAR was homogenized, as described in METHODS, and mitochondria were isolated from Percoll/Optiprep gradients (41, 42). Mitochondria were then assayed spectrophotometrically for cytochrome-c oxidase activity (A) and PKC- antisera-mediated coimmunoprecipitation of COIV (B), as in METHODS.

	DISCUSSION

TOP ABSTRACT METHODS RESULTS DISCUSSION REFERENCES

 
Our present laboratory focus is to better define the mitochondrial mechanisms of individual PKC isozymes in cardiac PC and I/R injury. We have previously determined that, during hypoxic PC, the PKC- isozyme interacts with the COIV to enhance CO activity in neonatal cardiac myocytes (42). In this study, we present evidence for the existence of this (PKC--COIV) relationship in a rat coronary ligation model of PC. In the progression of our work, we wished to clearly define the mitochondrial distributions of individual PKC isozymes in this model system, first, to determine whether activated PKC- was present in our gradient-purified adult rat heart mitochondria during PC and, second, to aid in determining which PKC isozymes to focus on in our future analyses. To our knowledge, there has never been a comprehensive mitochondrial survey of all PKC isozymes in an in situ rat coronary ligation model that involved purification of mitochondria using a Percoll/Optiprep density gradient and the subcellular fractionation protocols we employed. Based on our previous studies, we hypothesized that PKC- enhances CO activity to preserve and facilitate ATP production as a result of PC. Consistent with this hypothesis, it has been reported that recovery of high-energy phosphates was improved when PC was administered before prolonged I/R (20). This also should increase electron flux through the ETC, allowing reduced electron leak and ROS production from complexes I and III. We believe this is a previously uncharacterized mechanism of PC-induced cardioprotection by PKC-. However, there clearly are many other mechanisms, such as inhibition of the opening of the mitochondrial permeability transition pore (1), opening of the mitochondrial ATP-sensitive K⁺ channel (10), and others (2, 3, 26, 45, 47, 49, 55, 59).

Our results indicate that only 3% of the total PKC- is found in the M fraction under basal conditions. Figure 5 suggests this level increases by 60% following PC. This would be more than sufficient to modulate mitochondrial responses in PC, particularly if mitochondrial PKC- is localized to a specific submitochondrial domain, such as the inner mitochondrial membrane, as our PKC--COIV co-IP data suggest. It is also important to note that, if we had quantified the mitochondrial PKC- in L and P fractions, the total cellular mitochondrial pool of PKC- would likely be considerably higher.

As with PKC- KO mice (15, 53), PKC- KO mice have also been previously reported to be unable to precondition their hearts in response to a brief I/R exposure (34). In addition, the PKC- isozyme has previously been reported to translocate to cardiac myocyte mitochondria, which correlated with PC in Langendorf rat heart preparations (58). PKC- has also been implicated in cardiac I/R injury and apoptosis by many studies (6, 8, 20, 21, 36). We observed PKC- to be present in mitochondria before and following PC (Figs. 4 and 5), but there was no change in the mitochondrial level of PKC- following PC alone, however, and PKC- also existed in mitochondria following I30/R120 (Figs. 7 and 8). Our data, therefore, cannot rule out a possible role for PKC- in PC in this model. However, since PKC- is also present in mitochondria during I30/R120 (Fig. 7), it could play roles in cardiac damage, as previously proposed (8, 36, 37). The reasons for these discrepancies in the literature are currently unknown, but could involve differences in model systems or experimental protocols. For example, Wang et al. (58) used immunofluorescence staining methods in tissue harvested from Langendorf-perfused hearts to localize PKC- to mitochondria, while we used a more biochemical approach with purified mitochondria (Figs. 4–8). We also used an in situ model of PC and I/R injury in which the hemodynamics, metabolism, and physiology are likely to differ from Langendorf preparations. Mayr et al. (33, 34) conducted ischemia studies with hearts isolated from PKC- KO mice, and there could also be differences between these genetically modified mice and our in situ rat coronary ligation model. An interesting study by Hahn and coworkers (16) may shed some light on the conflicting reports of the roles of PKC- in PC and I/R injury. In their study, they demonstrated dose-dependent effects of overexpressing a PKC- translocation inhibitor in transgenic mice. At low levels of expression, the inhibitor resulted in increased resistance to I/R injury; at medium-level expression, the resistance was abolished; and, at high-level expression, the PKC- inhibitor was lethal, leading to excessive hypertrophy and myocardial structural and functional anomalies. It is, therefore, possible that a determining factor in whether PKC- mediates cardiac damage or protection may be related to where, when, and to what extent it becomes activated in the PC and I/R injury spectrum. This may mean that PKC- has multiple potential targets in these responses, and selective inhibition of certain responses contributes protection, whereas "global inhibition" of all PKC- in cells may result in accelerated cardiac injury and or death following I/R challenge.

We observed PKC- in the cell-soluble fraction, and, while its total levels were reduced after I/R, we found no consistent accumulation of PKC- in mitochondria under any of our experimental conditions. Our data, therefore, suggest that PKC- does not play significant mitochondrial roles in cardiac PC or I/R. The role of the PKC- isozyme in cardiac mitochondria is at present poorly defined. Phillipson et al. (46) determined that inhibition of PKC- in myocardium correlated with cardioprotection, but this effect was not mitochondrial in origin and was associated with decreased ROS production in polymorphonuclear leukocytes and enhanced nitric oxide release from vascular endothelial cells. As we are interested in PKC modulation of complex IV and other ETC complexes, the PKC- isozyme may be an excellent candidate for modulation of these complexes. However, PKC- was found in mitochondria under basal conditions, and there was no change in its distributions following PC (Fig. 5). It is possible that prelocalized mitochondrial PKC- may become activated to play roles in cardiac protection or cell death, but future studies will be required to assess its roles in these processes. These studies confirm that PKC isozymes exist in mitochondria under conditions of PC and I/R injury in this model. Which mitochondrial functions are mediated by each PKC isozyme is only beginning to be understood (40). Our studies suggest that prior localization of the PKC-, -, and - isozymes to cardiac mitochondria may allow these enzymes to be positioned to play roles in cardiac PC and, in the case of PKC-, possibly I/R damage. In addition, significant PKC- accumulation in mitochondria was observed following PC (Fig. 5, n = 5, P < 0.01), consistent with a role for mitochondrial PKC- in the regulation of CO in PC.

Significance to ischemic PC and I/R injury. If PC is administered before the I30/R120 period, the myocardium is protected, the COIV-PKC- co-IP is enhanced twofold, and CO activity is similarly enhanced. These events should be beneficial at reperfusion, as they should favor enhanced ATP production and better contribute to a more rapid return of ETC and oxidative phosphorylation functions. They should also reduce electron leak and the consequent damaging production of ROS from ETC complexes I and III. Our study is the first to demonstrate this response in an in situ model of PC.

	ACKNOWLEDGMENTS

 
This research was funded by National Heart, Lung, and Blood Institute Grants R01 HL076805 to J. A. Johnson and R01 HL070215 to R. W. Caldwell.

	FOOTNOTES

 

Address for reprint requests and other correspondence: J. A. Johnson, Dept. of Pharmacology & Toxicology, Medical College of Georgia, Augusta, GA 30912-2300 (e-mail: jjohnson{at}mail.mcg.edu)

The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

^* D. Guo and T. Nguyen contributed equally to this work.

	REFERENCES

TOP ABSTRACT METHODS RESULTS DISCUSSION REFERENCES

 

1. Baines CP, Song C, Zheng Y, Wang G, Zhang J, Wang O, Guo Y, Bolli R, Cardwell EM, Ping P. Protein kinase C interacts with and inhibits the permeability transition pore in cardiac mitochondria. Circ Res 92: 873–880, 2003.[Abstract/Free Full Text]
2. Baines CP, Zhang J, Wang GW, Zheng YT, Xiu JX, Cardwell EM, Bolli R, Ping P. Mitochondrial PKCepsilon and MAPK form modules in the murine heart: enhanced mitochondrial PKCepsilon-MAPK interactions and differential MAPK activation in PKCepsilon-induced cardioprotection. Circ Res 90: 390–397, 2002.[Abstract/Free Full Text]
3. Balafonova Z, Bolli R, Zhang J, Zheng Y, Pass JM, Bhatnagar A, Tang XL, Wang O, Cardwell E, Ping P. Nitric oxide (NO) induces nitration of protein kinase C epsilon, facilitating PKCEpsilon translocation via enhanced PKCEpsilon-RACK2 interactions. A novel mechanism of NO-triggered activation of PKCEpsilon. J Biol Chem 277: 15021–15027, 2002.[Abstract/Free Full Text]
4. Barbosa V, Sievers RE, Zaugg CE, Wolfe CL. Preconditioning ischemia time determines the degree of glycogen depletion and infarct size reduction in rat hearts. Am Heart J 131: 224–230, 1996.[CrossRef][ISI][Medline]
5. Brew EC, Mitchell MB, Rehring TF, Gamboni-