Proteomic Analysis of Pharmacological Preconditioning
http://www.100md.com
D. Kent Arrell, Steven T. Elliott, Lesle
参见附件。
the Department of Physiology (D.K.A., S.T.E., J.E.V.E.), Queen’s University, Kingston, Ontario, Canada
Departments of Medicine (S.T.E., Y.G., M.M., E.M., J.E.V.E.), Biological Chemistry (L.A.K., Y.H.K., P.L.P., J.E.V.E.), Pediatrics (J.R., A.M.M.)
Biomedical Engineering (J.E.V.E.), The Johns Hopkins University, Baltimore, Md. Present address for D.K.A.: Division of Cardiovascular Diseases, Departments of Medicine, Molecular Pharmacology and Experimental Therapeutics, Mayo Clinic, Rochester, Minn.
Abstract
Ischemic preconditioning is characterized by resistance to ischemia reperfusion injury in response to previous short ischemic episodes, a protective effect that can be mimicked pharmacologically. The underlying mechanism of protection remains controversial and requires greater understanding before it can be fully exploited therapeutically. To investigate the overall effect of preconditioning on the myocardial proteome, isolated rabbit ventricular myocytes were treated with drugs known to induce preconditioning, adenosine or diazoxide (each at 100 μmol/L for 60 minutes). Their protein profiles were then compared with vehicle-treated controls (n=4 animals per treatment) using a multitiered 2D gel electrophoresis approach. Of 28 significantly altered protein spots, 19 nonredundant proteins were identified (5 spots remained unidentified). The majority of these proteins are involved in mitochondrial energetics, including subunits of tricarboxylic acid cycle enzymes and oxidative phosphorylation complexes. These changes were not indiscriminate, with only a small number of enzymes or complex subunits altered, indicating a very specific and targeted affect of these 2 preconditioning mimetics. Among the changes were shifts in the extent of posttranslational modification of 4 proteins. One of these, the adenosine-induced phosphorylation of the ATP synthase subunit, was fully characterized with the identification of 5 novel phosphorylation sites. This proteomics approach provides an overall assessment of the cellular response to pharmacological treatment with adenosine and diazoxide and identifies a distinct subset of enzymes and protein complex subunit that may underlie the preconditioned phenotype.
Key Words: ATP synthase phosphorylation preconditioning proteomics
Introduction
Ischemic preconditioning (PC) is an endogenous cytoprotective process whereby cells become profoundly resistant to ischemia/reperfusion injury following brief episodes of conditioning ischemia. The initial report of PC described a 76% reduction in cell death in PC compared with control hearts.1 Protection was found to be transient, lasting 1 to 3 hours following the conditioning trigger,1,2 and is now known to be biphasic, resuming after 24 hours to protect for a second phase of 3 to 4 additional days.3,4 First described in the heart, PC is a ubiquitous phenomenon that has now been demonstrated in every mammalian tissue type and species examined to date,5 including humans.6,7
PC can also be initiated pharmacologically. Like ischemic PC, pharmacological PC triggers biphasic protection and inhibitors of pharmacological PC block ischemic PC,5 indicating commonality in their underlying mechanistic properties. A broad range of PC mimetic agents have been identified, encompassing such diverse functions as G protein-coupled receptor (GPCR) agonists (eg, adenosine), potassium channel openers (eg, diazoxide), sodium/hydrogen exchange inhibitors, and volatile anesthetics.8 Preservation of the preconditioned phenotype across this array of triggers suggests that they share a unifying molecular mechanism and, thus, common effects at the level of the proteome.
The proteome of preconditioned myocardium has not been extensively characterized. To address this, we undertook a broad-based proteomic analysis9 of an established isolated rabbit ventricular myocyte model of PC.10,11 We compared the proteomes of myocytes following 60-minute exposures to PC doses of either adenosine or diazoxide to that of vehicle-treated control myocytes. Simultaneous examination of 2 distinct modes of pharmacological PC allowed us to investigate possibilities of congruence or mutual exclusivity of the proteome alterations caused by each treatment. Here we describe extensive changes induced by both adenosine and diazoxide to the myocardial proteome, including alterations to a distinct subset of enzymes and subunits involved in mitochondrial energetics. This includes the characterization of PC-induced posttranslational modification (PTM) to a mitochondrial protein, the ATP synthase subunit.
Materials and Methods
Isolation and PC of Rabbit Ventricular Myocytes
New Zealand White rabbits (1 to 2 kg) were used in compliance with the Animals for Research Act (Province of Ontario), the Canadian Council on Animal Care, and the Guide for the Care and Use of Laboratory Animals.12 Rabbit ventricular myocyte isolation and pharmacological PC were performed as described previously,10,11 and cells were stored at –80°C. Cells were isolated from 1 rabbit heart and split into 3 aliquots; each aliquot was either treated with adenosine, diazoxide, or buffer (n=4). The use of a homogeneous cell population in the isolated myocyte model of PC eliminated confounding effects possible with multiple cell types present in the heart.
Subproteome Fractionation
Protein extracts were fractionated into HEPES and trifluoroacetic acid (TFA) fractions according to the IN Sequence extraction as described previously,13,14 aliquoted and stored at –80°C. This extraction protocol is robust, reproducible, and minimizes the introduction of artificial PTMs.
Two-Dimensional Gel Electrophoresis
Two-dimensional gel electrophoresis (2-DE) was performed in the first dimension using pH 3 to 10, pH 4 to 7, pH 4.7 to 5.9, or pH 6 to 9 linear gradients (Bio-Rad) and, in the second dimension, either by 10% or 12.5% SDS-PAGE using a Tris-glycine (pH 8.3) running buffer13 or by 10% bis-Tris SDS-PAGE with 2-(n-morpholino) ethanesulfonic acid (MES) running buffer.15 Gels were silver stained according to the protocol of Shevchenko et al16 and quantified (within the linear range of silver stain). Statistically significant protein changes were determined using 2-tailed Student’s t tests of the normalized spot intensities between groups using a probability value of <0.05. Other gels were transferred to nitrocellulose for immunoblotting. More extensive descriptions of 2-DE, image analysis, quantification, and immunoblotting are in the online data supplement, available at http://circres.ahajournals.org.
Protein Identification by Mass Spectrometry
Tryptic digests of excised protein spots were analyzed by matrix-assisted laser desorption ionization time of flight (MALDI TOF), mass spectrometry (MS), or electrospray ionization (ESI) tandem MS (MS/MS) (Q-TOF), as described previously.13,17 For details, see the online data supplement.
Sequencing of Rabbit ATP Synthase Subunit
Primers were designed based on the known ATP synthase subunit sequences and used to amplify overlapping fragments of the coding region from rabbit heart cDNA. The amplified DNA fragments were sequenced at the Johns Hopkins DNA Sequencing Facility. Alignment of ATP synthase subunit sequences was performed using ClustalW (version 1.83).
Phosphopeptide Analysis by Immobilized Metal Affinity Chromatography
Phosphopeptides were enriched with an immobilized metal affinity chromatography (IMAC) column.18,19 All reported phosphopeptide amino acid sequences were manually confirmed. For details see the online data supplement.
Phosphorylated Residue Mapping
The F1 moiety of the rat mitochondrial ATP synthase20 was depicted by a ribbon diagram using the Pymol viewer program. Serines and threonines are represented by a sphere model depicting van der Waals radii.
Results
To facilitate quantitative detection and to maximize the likelihood of identifying subtle changes to the proteome, ventricular myocytes from multiple rabbbits were subfractionated into 2 sequential subproteomes, a cytosolic-enriched extract and a myofilament-enriched extract, using an established protocol.13,14 Each extract was resolved by 2-DE using 11 different combinations of protein load, the isoelectric focusing (IEF) pH gradients, and the percentage of acrylamide used for SDS PAGE (Figure 1). A total of 28 protein spots were consistently and significantly altered. From these spots, 19 nonredundant proteins were identified (5 altered spots were not identified), based on MS analysis by peptide mass fingerprinting and/or MS/MS (Figure 2; Table II in the online data supplement). Diazoxide induced modifications to 15 identified proteins (and all 5 unidentified spots), whereas 13 identified proteins were affected by adenosine (and 3 of the 5 unidentified spots) (Figure 2). A total of 9 identified protein changes were common to both treatments. The majority of the proteins changed in abundance but 4 differed with respect to the presence and distribution of PTMs. Even with this extensive 2-DE analysis of multiple subproteomes, this is not comprehensive list of protein changes; rather, it is an overview of the cellular response with treatment of diazoxide and adenosine. Therefore, to put the protein changes in proper context, it is beneficial to identify those proteins unaffected by these drugs. MS analysis was performed on every well-resolved 2-DE spot, yielding the identities of 135 protein spots (supplemental Table I and Figure I) that encompassed a number of functional categories. 2-DE is limited in the observation of membrane proteins, as they are very hydrophobic and have difficulty entering the first dimension or focusing; thus we do not observe the membrane spanning subunits of the oxidative phosphorylation (OxPhos) complexes or other proteins.
Proteins Altered by PC
Mitochondrial energetic proteins comprised the majority of identified changes (10/19), including 3 constituents of the tricarboxylic acid (TCA) cycle (Figure 2A) and 7 oxidative phosphorylation (OxPhos) subunits (Figure 2B). The modulation of these energetics proteins was not indiscriminant; only distinct subsets of proteins were altered. For instance, the TCA cycle enzymes isocitrate dehydrogenase subunit and ADP-specific succinyl-CoA ligase chain were observed to increase with diazoxide or both treatments, respectively. However, other several other enzymes in this cycle (citrate synthase, fumarate hydratase, and 2-oxoisovalerate dehydrogenase subunit [Figure 2F]) were observed and remained unchanged with either treatment. Interestingly, subunits within the same enzyme complex were not collectively modulated. For example, the 23-kDa (diazoxide), 24-kDa (diazoxide and adenosine), and 30-kDa (diazoxide and adenosine) subunits of NADH ubiquinone oxidoreductase (complex I) increased in abundance, whereas the 13- and 49-kDa subunits were unaffected (Figure 2F).
Mitochondrial changes were not limited to only proteins involved in energetics. Three chaperones also were affected: metaxin 2, prohibitin 1, and heat shock protein 27 (HSP27) (Figure 2C). Although these chaperone proteins are involved with similar functions, they are modulated very differently in PC, with prohibitin (adenosine) and HSP27 (diazoxide) increasing and metaxin decreasing (both drugs).
The remainder of identified changes occurred to 6 proteins with a variety of homeostatic functions, including stress-response elements, regulation of NO or reactive oxygen species (ROS), Ca2+ mobilization, and muscle contraction (Figure 2D). Other proteins of similar function or subcellular localization were often observed and unchanged. Of note, myosin light chain (MLC)-1 was altered with a PTM by both drug treatments; yet its partner, MLC-2, was not changed, nor were other myofilament proteins observed (cardiac troponin T, cardiac actin, and tropomyosin).
Characterization of PC-Induced PTMs
MLC-1 is among the proteins with an observed PTM. Previously, our laboratory reported, using this same model of PC, that MLC-1 undergoes a novel diphosphorylation on treatment with adenosine,13 and we have now demonstrated that this also occurs with diazoxide (Figure 2D). Three additional proteins underwent PTM with drug treatments: the pyruvate dehydrogenase (PDH) E3-binding protein, ADP-ribosyl hydrolase, and the ATP synthase subunit (Figure 2). Unfortunately, the modifications to either the E3-binding protein or ADP-ribosyl hydrolase (Figure 2A and 2D, respectively) could not be determined because of insufficient protein quantity or peptide coverage even when spots were pooled from many 2-DE gels. However, E3-binding protein has a known PTM, a reductive acetylation of a lipoylated lysine residue, at K97.21
The modification of ATP synthase subunit, a protein subunit of ATP synthase (complex V) is of particular interest, as it is located in the inner mitochondrial membrane and neither it, nor any other inner mitochondrial proteins have previously been observed to have PC induced PTMs. The number of spots assigned to ATP synthase subunit increased from 2 to 6 when 2-DE resolution was improved by switching from the traditional pH 4 to 7 to a pH 4.7 to 5.9 gel (compare Figure 2B with Figure 3A, respectively). The 6 spots observed with the narrower pH gradient (pH 4.7 to 5.9) were identified by MS as the ATP synthase subunit, and this was confirmed by 2D immunoblot analysis (Figure 3A). The pI shift observed is consistent with phosphorylation, as reported recently for ATP synthase in diabetic human skeletal muscle,22 but -subunit phosphorylation has not previously been shown to occur in cardiac muscle. To investigate the possibility of phosphorylation, tryptic peptides of these 6 spots obtained from 2-D gels resolved for 3 separate rabbits were enriched for phosphopeptides by IMAC and analyzed by MS/MS for phosphorylation-site identification.
As the amino acid sequence of rabbit ATP synthase has not previously been determined, it was essential to clone it from rabbit cardiac muscle allowing the accurate matching of the resulting MS/MS spectra and mapping the positions of the phosphorylated amino acid residues. The first 478 amino acids were obtained (supplemental Figure II), corresponding to all but the last 50 amino acids of all known mammalian sequences (human, bovine, rat, mouse). This protein is highly conserved, differing from the human sequence by only 13 amino acids (>97% identity). Analysis of MS/MS spectra revealed that ATP synthase is phosphorylated, at up to 5 amino acid residues in the rabbit heart (Figure 3B and supplemental Figures III and IV). Phosphorylated serine (S) and threonine (T) residues were identified at the following positions in the rabbit sequence (followed by the corresponding position in the mature rat protein in italics): S106 (S56), T107 (T57), T262 (T212), or S263 (S213), T312 (T262), and T368 (T318). Each of these sites is conserved across many mammalian species (supplemental Figure II).
The locations of these novel phosphorylated residues were mapped onto the x-ray crystal 3D structure of rat F1 ATP synthase (96% identical to rabbit sequence) to determine their accessibility in the oligomeric state of the F1 complex. Based on this model,20 2 of the phosphorylated amino acid residues, T312 (T262) and T368 (T318), are not accessible in the fully assembled F1 catalytic complex (of which 1 , 3 , and 3 subunits constitute the core), indicating that these sites must be phosphorylated before assembly (supplemental Figure V). The remaining 3 residues, S106 (S56), T107 (T57), and T262/S263 (T212/S213), are accessible and therefore have the potential to be dynamically regulated in the fully assembled state during adenosine treatment. S106 (S56) is located in a loop structure on top of the enzyme within binding distance of the subunit (at E84 in the mature rat protein). The T262/S263 (T212/S213) sites are located on the side of the enzyme in a loop structure that interacts with another region of the subunit (at N207 in the rat mature protein).
Discussion
This study was undertaken to assess the cellular response to pharmacological treatment of adenosine and diazoxide, both of which result in a PC phenotype. Although the protein changes encompassed a wide number of functional groups, it was the mitochondrial proteins (10/19 identified changes) that dominated the proteome changes (Figure 4). It is not surprising that the mitochondria is involved in PC, because several proposed effectors of PC have been either proteins residing, or processes occurring, within mitochondria.8,23–25 Signaling events including various kinases (protein kinase C [PKC] isoforms, phosphatidylinositol 3-kinase [PI3K], protein kinase G [PKG], mitogen-activated protein kinase [MAPK], and others) are involved in transmitting the PC signal from the cell surface to the cytosol and mitochondria26–30 and are activated within minutes of exposure to a PC stimulus.31 However, less is known about the ultimate protein targets of these activated kinases. There is strong evidence supporting a role for several candidate protein complexes in the mitochondria specifically, the mitochondrial permeability transition pore protein complex made up of voltage-dependent anion channel (VDAC), hexokinase II, and adenine nucleotide translocase (ANT) which has been shown to bind PKC.32 As well, the succinate dehydrogenase protein complex, which displays functional properties that are similar to the mitochondrial KATP channel, which on opening (initiated by diazoxide) induces PC.33 What is not clear is the breadth of the PC targets and how these differ among PC agents. It is striking that the protein changes observed in this broad overview of the myocyte proteome are a defined subset of proteins. Within the mitochondria, the changes are not sweeping, with only a few enzymes targeted, and, in fact, specific subunits of these protein complexes modulated, by the PC stimuli. The specificity of this response indicates that these changes are not random events that may arise for nonspecific NO effects or oxidation but, rather, are targeted effects. The adenosine or diazoxide induced changes were observed in both high abundance proteins (eg, ATP synthase subunit or MLC-1) as well as in low-abundance proteins (eg, ADP-ribosyl hydrolase or the pyruvate dehydrogenase E3-binding protein), again supporting a targeted action by these drugs. Listed in the online data supplement (supplemental Table II) are the proteins previously observed to be connected to PC (9/19) or ischemia injury (8/19). The majority of the protein changes described in this study have not been previously connected to PC or ischemia. (See the online data supplement for further discussion of protein functions and possible implications.)
There are 3 protein changes to PDH or TCA cycle subunits (Figure 2) as well as alterations to some, but not all, subunits of the OxPhos complexes I and V, with a single change in complex III (Figure 2). Complex I and II have primarily been shown to be inhibited by a number of PC agents.34–38 Metabolic effects were proposed in the initial report of PC1 and further substantiated as a slowing of metabolism.39,40 On reperfusion, energetic processes are also preserved by PC, ATP levels recover more rapidly,41,42 mitochondrial respiration is better preserved,43,44 and postischemic recovery is improved through enhanced coupling between energy production and utilization.45 Several OxPhos complexes have been mechanistically linked to PC and this study links PC to individual subunits abundance changes or PTMs. These data can lead to more in-depth mechanistic and functional studies of how and why these particular subunits are modulated in PC.
The OxPhos pathway is the primary source of ROS formation in higher organisms,46 specifically at complexes I and III,47 and both of these complexes contained altered subunits (Figure 2B). Slowing of electron transport by partial inhibition of either complex I or III decreases electron transfer efficiency and thereby promotes ROS formation, a mechanism by which other PC mimetics have been proposed to protect.48,49 It is important to note that the OxPhos complexes contain large numbers of integral membrane proteins not easily amenable to 2-DE, so it is possible that not every change was detected by this method.
OxPhos complex V, or F1Fo ATP synthase, has also been mechanistically linked with PC. In an isolated rat heart model, the protective effect of ischemia or diazoxide PC (at 20 μmol/L) was associated with lowered ATP synthase activity.50 PC doses of diazoxide (ranging from 20 to 50 μmol/L) also reduce ATP synthase activity in isolated liver51 and heart mitochondria.38,50 Three of the 4 ATP synthase subunits we identified (d, , and ) were altered by PC (Figure 2), making them strong candidates for further investigation into defining the mechanism of the inhibitory effect of PC on ATP synthase. It is not clear how increases in the d and subunits might influence ATP synthase activity. The d subunit is part of the Fo H+ channel that links electron transport to ATP production, whereas the subunit forms part of the stator that stabilizes the ATP catalysis portion of F1.52
The extensive phosphorylation of the subunit is the first report of phosphorylation of this critical enzyme in cardiac tissue and connects a PC stimulus to a PTM of a protein within the mitochondria. Interesting, this subunit has been linked to PC of neuronal cell cultures. PC of cortical neuronal culture with erythropoietin,53 heat stress,54 and MK80154 lead to decreases in 1 protein spot of the ATP synthase subunit. (See the online data supplement, Table II.) These studies only analyzed the changing protein spots and did not provide data for the unchanged spot identifications; thus these findings could indicate a reduced concentration or change in PTM status. This finding also renews the emphasis of ATP production as either an initiator or effector of PC, as initially speculated in previous PC research.40–42
To better understand the significance of the phosphorylated residues, they were modeled onto the 3D x-ray crystallographic structure of the rat F1 catalytic unit20 (supplemental Figure V). The clustering and symmetry of the phosphorylated residues of the subunit is striking, with all of the residues aligning within each subunit. All 3 of the accessible phosphorylated residues are located in loop structures that form structural interactions with both the subunit (S106/T107 [S56/T57]) and within the subunit (T262 and S263 [T212 and S213]) and have the potential to regulate function when modulated by adenosine treatment. Experiments to test the functional consequences of these phosphorylations are important to understanding the true function in PC and are ongoing. This finding implies that endogenous mitochondrial kinase or phosphatase activities are influenced during the 60 minutes adenosine treatment and could have larger implications for PC models and signaling to the mitochondria.
Three proteins in addition to ATP synthase were found to have PC-induced PTMs. As reported previously, adenosine increases the proportion of phosphorylated MLC-1,13 and diazoxide has now been shown to confer the same effect (Figure 2D). ADP-ribosyl hydrolase, including a PTM form, is increased by adenosine and diazoxide. Not previously known to be modified, ADP-ribosyl hydrolase has a broad array of functions. It is among many enzymes involved in NAD+ metabolism and also competitively inhibits OxPhos complex I,55 both of which could influence electron transport during PC. Further analysis is required to uncover its role in PC and to identify the PTM and its potential effect on enzyme activity. As well, diazoxide-induced increase in the PTM form of PDH E3-binding protein (Figure 2A) could lead to reduced PDH activity via the well-documented phosphorylation of the E1 subunit.21 As described in greater detail in the online data supplement, a chain of events arising from E3-binding protein modification would decrease acetyl coenzyme A (CoA) production, necessary for maintenance of metabolic flux through the TCA cycle.
Implications for PC
Proteome changes induced by 2 different modes of PC in the same model show a distinctive pattern of affected proteins consistent with specific perturbation of mitochondrial metabolism through changes to a selected number of mitochondrial protein complexes. Interestingly, specific mitochondrial subunits of the targeted protein complexes were altered, whereas other remained unaffected, indicating complexity and specificity of the drug response. This may be a general phenomenon. As well, the demonstration that inner mitochondrial protein, ATP synthase subunit, can be phosphorylated within 60 minutes of treatment with adenosine is intriguing with respect to how kinase/phosphatase activity is regulated within this segregated compartment of the cell. Although mitochondria is central to PC, it is does not have an exclusive role. Our study suggest that, at least, these PC agents have a wider cellular affect but equally selective response to other subproteomes, which suggests these are not random events (Figure 4). This study provides an overall assessment of the cellular response adenosine and diazoxide treatment and identifies several protein targets that are both unexpected and novel.
Acknowledgments
We thank Gilles Lajoie (University of Western Ontario, Siebens-Drake Research Institute) for MS/MS contributions to this study.
Sources of Funding
D.K.A. was the recipient of a doctoral research award from the Heart and Stroke Foundation of Ontario and a postdoctoral fellowship from the Heart and Stroke Foundation of Canada. J.E.V.E. acknowledges the following funding sources: the Canadian Institute of Health Research, the Canadian Heart and Stroke Foundation, the National Heart Lung Blood Institute Proteomic Initiative (contract NO-HV-28120), and the Donald P. Amos Family Foundation. P.L.P. acknowledges funding from the following sources: National Institute of Health CA-O1O951 and P01 HL081427.
Disclosures
None.
Footnotes
This manuscript was sent to Joseph Loscalzo, Consulting Editor, for review by expert referees, editorial decision, and final disposition.
These authors contributed equally to this work.
Original received February 24, 2006; resubmission received August 11, 2006; accepted August 22, 2006.
References
Murry CE, Jennings RB, Reimer KA. Preconditioning with ischemia: a delay of lethal cell injury in ischemic myocardium. Circulation. 1986; 74: 1124–1136.
Murry CE, Richard VJ, Jennings RB, Reimer KA. Myocardial protection is lost before contractile function recovers from ischemic preconditioning. Am J Physiol. 1991; 260: H796–H804. [Order article via Infotrieve]
Kuzuya T, Hoshida S, Yamashita N, Fuji H, Oe H, Hori M, Kamada T, Tada M. Delayed effects of sublethal ischemia on the acquisition of tolerance to ischemia. Circ Res. 1993; 72: 1293–1299.
Marber MS, Latchman DS, Walker JM, Yellon DM. Cardiac stress protein elevation 24 hours after brief ischemia or heat stress is associated with resistance to myocardial infarction. Circulation. 1993; 88: 1264–1272.
Kloner RA, Jennings RB. Consequences of brief ischemia: stunning, preconditioning, and their clinical implications: part 2. Circulation. 2001; 104: 3158–3167.
Yellon DM, Alkhulaifi AM, Pugsley WB. Preconditioning the human myocardium. Lancet. 1993; 342: 276–277. [Order article via Infotrieve]
Ikonomidis JS, Tumiati LC, Weisel RD, Mickle DA, Li RK. Preconditioning human ventricular cardiomyocytes with brief periods of simulated ischaemia. Cardiovasc Res. 1994; 28: 1285–1291. [Order article via Infotrieve]
Yellon DM, Downey JM. Preconditioning the myocardium: from cellular physiology to clinical cardiology. Physiol Rev. 2003; 83: 1113–1151.
Graham DR, Elliott ST, Van Eyk JE. Broad-based proteomic strategies: a practical guide to proteomics and functional screening. J Physiol. 2005; 563: 1–9.
Liu Y, Sato T, O’Rourke B, Marban E. Mitochondrial ATP-dependent potassium channels: novel effectors of cardioprotection Circulation. 1998; 97: 2463–2469.
Sato T, Sasaki N, O’Rourke B, Marban E. Adenosine primes the opening of mitochondrial ATP-sensitive potassium channels: a key step in ischemic preconditioning Circulation. 2000; 102: 800–805.
National Research Council. Institute of Laboratory Animal Resources Commission on Life Sciences. Guide for the Care and Use of Laboratory Animals. Wash DC: National Academy Press; 1996.
Arrell DK, Neverova I, Fraser H, Marban E, Van Eyk JE. Proteomic analysis of pharmacologically preconditioned cardiomyocytes reveals novel phosphorylation of myosin light chain 1. Circ Res. 2001; 89: 480–487.
Neverova I, Van Eyk JE. Application of reversed phase high performance liquid chromatography for subproteomic analysis of cardiac muscle. Proteomics. 2002; 2: 22–31. [Order article via Infotrieve]
Graham DR, Garnham CP, Fu Q, Robbins J, Van Eyk JE. Improvements in two-dimensional gel electrophoresis by utilizing a low cost "in-house" neutral pH sodium dodecyl sulfate-polyacrylamide gel electrophoresis system. Proteomics. 2005; 5: 2309–2314. [Order article via Infotrieve]
Shevchenko A, Wilm M, Vorm O, Mann M. Mass spectrometric sequencing of proteins silver-stained polyacrylamide gels. Anal Chem. 1996; 68: 850–858. [Order article via Infotrieve]
Dobransky T, Brewer D, Lajoie G, Rylett RJ. Phosphorylation of 69-kDa choline acetyltransferase at threonine 456 in response to amyloid-beta peptide 1–42. J Biol Chem. 2003; 278: 5883–5893.
Ficarro SB, McCleland ML, Stukenberg PT, Burke DJ, Ross MM, Shabanowitz J, Hunt DF, White FM. Phosphoproteome analysis by mass spectrometry and its application to Saccharomyces cerevisiae. Nat Biotechnol. 2002; 20: 301–305. [Order article via Infotrieve]
Zhang C, Williams EH, Guo Y, Lum L, Beachy PA. Extensive phosphorylation of Smoothened in Hedgehog pathway activation. Proc Natl Acad Sci U S A. 2004; 101: 17900–17907.
Bianchet MA, Hullihen J, Pedersen PL, Amzel LM. The 2.8-A structure of rat liver F1-ATPase: configuration of a critical intermediate in ATP synthesis/hydrolysis. Proc Natl Acad Sci U S A. 1998; 95: 11065–11070.
Roche TE, Hiromasa Y, Turkan A, Gong X, Peng T, Yan X, Kasten SA, Bao H, Dong J. Essential roles of lipoyl domains in the activated function and control of pyruvate dehydrogenase kinases and phosphatase isoform 1. Eur J Biochem. 2003; 270: 1050–1056.
Hojlund K, Wrzesinski K, Larsen PM, Fey SJ, Roepstorff P, Handberg A, Dela F, Vinten J, McCormack JG, Reynet C, Beck-Nielsen H. Proteome analysis reveals phosphorylation of ATP synthase beta -subunit in human skeletal muscle and proteins with potential roles in type 2 diabetes. J Biol Chem. 2003; 278: 10436–10442.
Opie LH, Sack MN. Metabolic plasticity and the promotion of cardiac protection in ischemia and ischemic preconditioning. J Mol Cell Cardiol. 2002; 34: 1077–1089. [Order article via Infotrieve]
Minners J, McLeod CJ, Sack MN. Mitochondrial plasticity in classical ischemic preconditioning-moving beyond the mitochondrial KATP channel. Cardiovasc Res. 2003; 59: 1–6. [Order article via Infotrieve]
Marin-Garcia J, Goldenthal MJ. Mitochondria play a critical role in cardioprotection. J Card Fail. 2004; 10: 55–66. [Order article via Infotrieve]
Vondriska TM, Klein JB, Ping P. Use of functional proteomics to investigate PKC epsilon-mediated cardioprotection: the signaling module hypothesis. Am J Physiol Heart Circ Physiol. 2001; 280: H1434–H1441.
Robinet A, Hoizey G, Millart H. PI 3-kinase, protein kinase C, and protein kinase A are involved in the trigger phase of beta1-adrenergic preconditioning. Cardiovasc Res. 2005; 66: 530–542. [Order article via Infotrieve]
Cuong DV, Kim N, Youm JB, Joo H, Warda M, Lee JW, Park WS, Kim T, Kang S, Kim H, Han J. Nitric oxide-cGMP-protein kinase G signaling pathway induces anoxic preconditioning through activation of ATP-sensitive K+ channels in rat hearts. Am J Physiol Heart Circ Physiol. 2006; 290: H1808–H1817.
Xuan YT, Guo Y, Zhu Y, Wang OL, Rokosh G, Messing RO, Bolli R. Role of the protein kinase C-epsilon-Raf-1-MEK-1/2-p44/42 MAPK signaling cascade in the activation of signal transducers and activators of transcription 1 and 3 and induction of cyclooxygenase-2 after ischemic preconditioning. Circulation. 2005; 112: 1971–1978.
Ballard-Croft C, Kristo G, Yoshimura Y, Reid E, Keith BJ, Mentzer RM Jr, Lasley RD. Acute adenosine preconditioning is mediated by p38 MAPK activation in discrete subcellular compartments. Am J Physiol Heart Circ Physiol. 2005; 288: H1359–H1366.
Solenkova NV, Solodushko V, Cohen MV, Downey JM. Endogenous adenosine protects preconditioned heart during early minutes of reperfusion by activating Akt. Am J Physiol Heart Circ Physiol. 2006; 290: H441–H449.
Baines CP, Song CX, Zheng YT, Wang GW, Zhang J, Wang OL, Guo Y, Bolli R, Cardwell EM, Ping P. Protein kinase Cepsilon interacts with and inhibits the permeability transition pore in cardiac mitochondria. Circ Res. 2003; 92: 873–880.
Ardehali H, Chen Z, Ko Y, Mejia-Alvarez R, Marban E. Multiprotein complex containing succinate dehydrogenase confers mitochondrial ATP-sensitive K+ channel activity. Proc Natl Acad Sci U S A. 2004; 101: 11880–11885.
Berman MC, Kewley CF, Kench JE. Contribution of inhibition of NADH-dehydrogenase to the cardiotoxic effects of halothane. J Mol Cell Cardiol. 1974; 6: 39–47. [Order article via Infotrieve]
Hanley PJ, Mickel M, Loffler M, Brandt U, Daut J. K(ATP) channel-independent targets of diazoxide and 5-hydroxydecanoate in the heart. J Physiol. 2002; 542: 735–741.
Hanley PJ, Ray J, Brandt U, Daut J. Halothane, isoflurane and sevoflurane inhibit NADH: ubiquinone oxidoreductase (complex I) of cardiac mitochondria. J Physiol. 2002; 544: 687–693.
Schafer G, Wegener C, Portenhauser R, Bojanovski D. Diazoxide, an inhibitor of succinate oxidation. Biochem Pharmacol. 1969; 18: 2678–2681. [Order article via Infotrieve]
Dzeja PP, Bast P, Ozcan C, Valverde A, Holmuhamedov EL, Van Wylen DG, Terzic A. Targeting nucleotide-requiring enzymes: implications for diazoxide-induced cardioprotection. Am J Physiol Heart Circ Physiol. 2003; 284: H1048–H1056.
Murry CE, Richard VJ, Reimer KA, Jennings RB. Ischemic preconditioning slows energy metabolism and delays ultrastructural damage during a sustained ischemic episode. Circ Res. 1990; 66: 913–931.
Jennings RB, Sebbag L, Schwartz LM, Crago MS, Reimer KA. Metabolism of preconditioned myocardium: effect of loss and reinstatement of cardioprotection. J Mol Cell Cardiol. 2001; 33: 1571–1588. [Order article via Infotrieve]
Miyamae M, Fujiwara H, Kida M, Yokota R, Tanaka M, Katsuragawa M, Hasegawa K, Ohura M, Koga K, Yabuuchi Y, Sasayama S. Preconditioning improves energy metabolism during reperfusion but does not attenuate myocardial stunning in porcine hearts. Circulation. 1993; 88: 223–234.
Fryer RM, Eells JT, Hsu AK, Henry MM, Gross GJ. Ischemic preconditioning in rats: role of mitochondrial K(ATP) channel in preservation of mitochondrial function. Am J Physiol Heart Circ Physiol. 2000; 278: H305–H312.
Yabe K, Nasa Y, Sato M, Iijima R, Takeo S. Preconditioning preserves mitochondrial function and glycolytic flux during an early period of reperfusion in perfused rat hearts. Cardiovasc Res. 1997; 33: 677–685. [Order article via Infotrieve]
Laclau MN, Boudina S, Thambo JB, Tariosse L, Gouverneur G, Bonoron-Adele S, Saks VA, Garlid KD, Dos Santos P. Cardioprotection by ischemic preconditioning preserves mitochondrial function and functional coupling between adenine nucleotide translocase and creatine kinase. J Mol Cell Cardiol. 2001; 33: 947–956. [Order article via Infotrieve]
Pucar D, Dzeja PP, Bast P, Juranic N, Macura S, Terzic A. Cellular energetics in the preconditioned state: protective role for phosphotransfer reactions captured by 18O-assisted 31P NMR. J Biol Chem. 2001; 276: 44812–44819.
Chance B, Sies H, Boveris A. Hydroperoxide metabolism in mammalian organs. Physiol Rev. 1979; 59: 527–605.
Cadenas E, Boveris A, Ragan CI, Stoppani AO. Production of superoxide radicals and hydrogen peroxide by NADH-ubiquinone reductase and ubiquinol-cytochrome c reductase from beef-heart mitochondria. Arch Biochem Biophys. 1977; 180: 248–257. [Order article via Infotrieve]
Mullenheim J, Ebel D, Frassdorf J, Preckel B, Thamer V, Schlack W. Isoflurane preconditions myocardium against infarction via release of free radicals. Anesthesiology. 2002; 96: 934–940. [Order article via Infotrieve]
Novalija E, Varadarajan SG, Camara AK, An J, Chen Q, Riess ML, Hogg N, Stowe DF. Anesthetic preconditioning: triggering role of reactive oxygen and nitrogen species in isolated hearts. Am J Physiol Heart Circ Physiol. 2002; 283: H44–H52.
Ala-Rami A, Ylitalo KV, Hassinen IE. Ischaemic preconditioning and a mitochondrial KATP channel opener both produce cardioprotection accompanied by F1F0-ATPase inhibition in early ischaemia. Basic Res Cardiol. 2003; 98: 250–258. [Order article via Infotrieve]
Schafer G, Portenhauser R, Trolp R. Inhibition of mitochondrial metabolism by the diabetogenic thiadiazine diazoxide. I. Action on succinate dehydrogenase and TCA-cycle oxidations. Biochem Pharmacol. 1971; 20: 1271–1280. [Order article via Infotrieve]
Boyer PD. The ATP synthase—a splendid molecular machine. Annu Rev Biochem. 1997; 66: 717–749. [Order article via Infotrieve]
Meloni BP, Tilbrook PA, Boulos S, Arthur PG, Knuckey NW. Erythropoietin preconditioning in neuronal cultures: signaling, protection from in vitro ischemia, and proteomic analysis. J Neurosci Res. 2006; 83: 584–593. [Order article via Infotrieve]
Meloni BP, Van Dyk D, Cole R, Knuckey NW. Proteome analysis of cortical neuronal cultures following cycloheximide, heat stress and MK801 preconditioning. Proteomics. 2005; 5: 4743–4753. [Order article via Infotrieve]
Zharova TV, Vinogradov AD. A competitive inhibition of the mitochondrial NADH-ubiquinone oxidoreductase (complex I) by ADP-ribose. Biochim Biophys Acta. 1997; 1320: 256–264. [Order article via Infotrieve]
the Department of Physiology (D.K.A., S.T.E., J.E.V.E.), Queen’s University, Kingston, Ontario, Canada
Departments of Medicine (S.T.E., Y.G., M.M., E.M., J.E.V.E.), Biological Chemistry (L.A.K., Y.H.K., P.L.P., J.E.V.E.), Pediatrics (J.R., A.M.M.)
Biomedical Engineering (J.E.V.E.), The Johns Hopkins University, Baltimore, Md. Present address for D.K.A.: Division of Cardiovascular Diseases, Departments of Medicine, Molecular Pharmacology and Experimental Therapeutics, Mayo Clinic, Rochester, Minn.
Abstract
Ischemic preconditioning is characterized by resistance to ischemia reperfusion injury in response to previous short ischemic episodes, a protective effect that can be mimicked pharmacologically. The underlying mechanism of protection remains controversial and requires greater understanding before it can be fully exploited therapeutically. To investigate the overall effect of preconditioning on the myocardial proteome, isolated rabbit ventricular myocytes were treated with drugs known to induce preconditioning, adenosine or diazoxide (each at 100 μmol/L for 60 minutes). Their protein profiles were then compared with vehicle-treated controls (n=4 animals per treatment) using a multitiered 2D gel electrophoresis approach. Of 28 significantly altered protein spots, 19 nonredundant proteins were identified (5 spots remained unidentified). The majority of these proteins are involved in mitochondrial energetics, including subunits of tricarboxylic acid cycle enzymes and oxidative phosphorylation complexes. These changes were not indiscriminate, with only a small number of enzymes or complex subunits altered, indicating a very specific and targeted affect of these 2 preconditioning mimetics. Among the changes were shifts in the extent of posttranslational modification of 4 proteins. One of these, the adenosine-induced phosphorylation of the ATP synthase subunit, was fully characterized with the identification of 5 novel phosphorylation sites. This proteomics approach provides an overall assessment of the cellular response to pharmacological treatment with adenosine and diazoxide and identifies a distinct subset of enzymes and protein complex subunit that may underlie the preconditioned phenotype.
Key Words: ATP synthase phosphorylation preconditioning proteomics
Introduction
Ischemic preconditioning (PC) is an endogenous cytoprotective process whereby cells become profoundly resistant to ischemia/reperfusion injury following brief episodes of conditioning ischemia. The initial report of PC described a 76% reduction in cell death in PC compared with control hearts.1 Protection was found to be transient, lasting 1 to 3 hours following the conditioning trigger,1,2 and is now known to be biphasic, resuming after 24 hours to protect for a second phase of 3 to 4 additional days.3,4 First described in the heart, PC is a ubiquitous phenomenon that has now been demonstrated in every mammalian tissue type and species examined to date,5 including humans.6,7
PC can also be initiated pharmacologically. Like ischemic PC, pharmacological PC triggers biphasic protection and inhibitors of pharmacological PC block ischemic PC,5 indicating commonality in their underlying mechanistic properties. A broad range of PC mimetic agents have been identified, encompassing such diverse functions as G protein-coupled receptor (GPCR) agonists (eg, adenosine), potassium channel openers (eg, diazoxide), sodium/hydrogen exchange inhibitors, and volatile anesthetics.8 Preservation of the preconditioned phenotype across this array of triggers suggests that they share a unifying molecular mechanism and, thus, common effects at the level of the proteome.
The proteome of preconditioned myocardium has not been extensively characterized. To address this, we undertook a broad-based proteomic analysis9 of an established isolated rabbit ventricular myocyte model of PC.10,11 We compared the proteomes of myocytes following 60-minute exposures to PC doses of either adenosine or diazoxide to that of vehicle-treated control myocytes. Simultaneous examination of 2 distinct modes of pharmacological PC allowed us to investigate possibilities of congruence or mutual exclusivity of the proteome alterations caused by each treatment. Here we describe extensive changes induced by both adenosine and diazoxide to the myocardial proteome, including alterations to a distinct subset of enzymes and subunits involved in mitochondrial energetics. This includes the characterization of PC-induced posttranslational modification (PTM) to a mitochondrial protein, the ATP synthase subunit.
Materials and Methods
Isolation and PC of Rabbit Ventricular Myocytes
New Zealand White rabbits (1 to 2 kg) were used in compliance with the Animals for Research Act (Province of Ontario), the Canadian Council on Animal Care, and the Guide for the Care and Use of Laboratory Animals.12 Rabbit ventricular myocyte isolation and pharmacological PC were performed as described previously,10,11 and cells were stored at –80°C. Cells were isolated from 1 rabbit heart and split into 3 aliquots; each aliquot was either treated with adenosine, diazoxide, or buffer (n=4). The use of a homogeneous cell population in the isolated myocyte model of PC eliminated confounding effects possible with multiple cell types present in the heart.
Subproteome Fractionation
Protein extracts were fractionated into HEPES and trifluoroacetic acid (TFA) fractions according to the IN Sequence extraction as described previously,13,14 aliquoted and stored at –80°C. This extraction protocol is robust, reproducible, and minimizes the introduction of artificial PTMs.
Two-Dimensional Gel Electrophoresis
Two-dimensional gel electrophoresis (2-DE) was performed in the first dimension using pH 3 to 10, pH 4 to 7, pH 4.7 to 5.9, or pH 6 to 9 linear gradients (Bio-Rad) and, in the second dimension, either by 10% or 12.5% SDS-PAGE using a Tris-glycine (pH 8.3) running buffer13 or by 10% bis-Tris SDS-PAGE with 2-(n-morpholino) ethanesulfonic acid (MES) running buffer.15 Gels were silver stained according to the protocol of Shevchenko et al16 and quantified (within the linear range of silver stain). Statistically significant protein changes were determined using 2-tailed Student’s t tests of the normalized spot intensities between groups using a probability value of <0.05. Other gels were transferred to nitrocellulose for immunoblotting. More extensive descriptions of 2-DE, image analysis, quantification, and immunoblotting are in the online data supplement, available at http://circres.ahajournals.org.
Protein Identification by Mass Spectrometry
Tryptic digests of excised protein spots were analyzed by matrix-assisted laser desorption ionization time of flight (MALDI TOF), mass spectrometry (MS), or electrospray ionization (ESI) tandem MS (MS/MS) (Q-TOF), as described previously.13,17 For details, see the online data supplement.
Sequencing of Rabbit ATP Synthase Subunit
Primers were designed based on the known ATP synthase subunit sequences and used to amplify overlapping fragments of the coding region from rabbit heart cDNA. The amplified DNA fragments were sequenced at the Johns Hopkins DNA Sequencing Facility. Alignment of ATP synthase subunit sequences was performed using ClustalW (version 1.83).
Phosphopeptide Analysis by Immobilized Metal Affinity Chromatography
Phosphopeptides were enriched with an immobilized metal affinity chromatography (IMAC) column.18,19 All reported phosphopeptide amino acid sequences were manually confirmed. For details see the online data supplement.
Phosphorylated Residue Mapping
The F1 moiety of the rat mitochondrial ATP synthase20 was depicted by a ribbon diagram using the Pymol viewer program. Serines and threonines are represented by a sphere model depicting van der Waals radii.
Results
To facilitate quantitative detection and to maximize the likelihood of identifying subtle changes to the proteome, ventricular myocytes from multiple rabbbits were subfractionated into 2 sequential subproteomes, a cytosolic-enriched extract and a myofilament-enriched extract, using an established protocol.13,14 Each extract was resolved by 2-DE using 11 different combinations of protein load, the isoelectric focusing (IEF) pH gradients, and the percentage of acrylamide used for SDS PAGE (Figure 1). A total of 28 protein spots were consistently and significantly altered. From these spots, 19 nonredundant proteins were identified (5 altered spots were not identified), based on MS analysis by peptide mass fingerprinting and/or MS/MS (Figure 2; Table II in the online data supplement). Diazoxide induced modifications to 15 identified proteins (and all 5 unidentified spots), whereas 13 identified proteins were affected by adenosine (and 3 of the 5 unidentified spots) (Figure 2). A total of 9 identified protein changes were common to both treatments. The majority of the proteins changed in abundance but 4 differed with respect to the presence and distribution of PTMs. Even with this extensive 2-DE analysis of multiple subproteomes, this is not comprehensive list of protein changes; rather, it is an overview of the cellular response with treatment of diazoxide and adenosine. Therefore, to put the protein changes in proper context, it is beneficial to identify those proteins unaffected by these drugs. MS analysis was performed on every well-resolved 2-DE spot, yielding the identities of 135 protein spots (supplemental Table I and Figure I) that encompassed a number of functional categories. 2-DE is limited in the observation of membrane proteins, as they are very hydrophobic and have difficulty entering the first dimension or focusing; thus we do not observe the membrane spanning subunits of the oxidative phosphorylation (OxPhos) complexes or other proteins.
Proteins Altered by PC
Mitochondrial energetic proteins comprised the majority of identified changes (10/19), including 3 constituents of the tricarboxylic acid (TCA) cycle (Figure 2A) and 7 oxidative phosphorylation (OxPhos) subunits (Figure 2B). The modulation of these energetics proteins was not indiscriminant; only distinct subsets of proteins were altered. For instance, the TCA cycle enzymes isocitrate dehydrogenase subunit and ADP-specific succinyl-CoA ligase chain were observed to increase with diazoxide or both treatments, respectively. However, other several other enzymes in this cycle (citrate synthase, fumarate hydratase, and 2-oxoisovalerate dehydrogenase subunit [Figure 2F]) were observed and remained unchanged with either treatment. Interestingly, subunits within the same enzyme complex were not collectively modulated. For example, the 23-kDa (diazoxide), 24-kDa (diazoxide and adenosine), and 30-kDa (diazoxide and adenosine) subunits of NADH ubiquinone oxidoreductase (complex I) increased in abundance, whereas the 13- and 49-kDa subunits were unaffected (Figure 2F).
Mitochondrial changes were not limited to only proteins involved in energetics. Three chaperones also were affected: metaxin 2, prohibitin 1, and heat shock protein 27 (HSP27) (Figure 2C). Although these chaperone proteins are involved with similar functions, they are modulated very differently in PC, with prohibitin (adenosine) and HSP27 (diazoxide) increasing and metaxin decreasing (both drugs).
The remainder of identified changes occurred to 6 proteins with a variety of homeostatic functions, including stress-response elements, regulation of NO or reactive oxygen species (ROS), Ca2+ mobilization, and muscle contraction (Figure 2D). Other proteins of similar function or subcellular localization were often observed and unchanged. Of note, myosin light chain (MLC)-1 was altered with a PTM by both drug treatments; yet its partner, MLC-2, was not changed, nor were other myofilament proteins observed (cardiac troponin T, cardiac actin, and tropomyosin).
Characterization of PC-Induced PTMs
MLC-1 is among the proteins with an observed PTM. Previously, our laboratory reported, using this same model of PC, that MLC-1 undergoes a novel diphosphorylation on treatment with adenosine,13 and we have now demonstrated that this also occurs with diazoxide (Figure 2D). Three additional proteins underwent PTM with drug treatments: the pyruvate dehydrogenase (PDH) E3-binding protein, ADP-ribosyl hydrolase, and the ATP synthase subunit (Figure 2). Unfortunately, the modifications to either the E3-binding protein or ADP-ribosyl hydrolase (Figure 2A and 2D, respectively) could not be determined because of insufficient protein quantity or peptide coverage even when spots were pooled from many 2-DE gels. However, E3-binding protein has a known PTM, a reductive acetylation of a lipoylated lysine residue, at K97.21
The modification of ATP synthase subunit, a protein subunit of ATP synthase (complex V) is of particular interest, as it is located in the inner mitochondrial membrane and neither it, nor any other inner mitochondrial proteins have previously been observed to have PC induced PTMs. The number of spots assigned to ATP synthase subunit increased from 2 to 6 when 2-DE resolution was improved by switching from the traditional pH 4 to 7 to a pH 4.7 to 5.9 gel (compare Figure 2B with Figure 3A, respectively). The 6 spots observed with the narrower pH gradient (pH 4.7 to 5.9) were identified by MS as the ATP synthase subunit, and this was confirmed by 2D immunoblot analysis (Figure 3A). The pI shift observed is consistent with phosphorylation, as reported recently for ATP synthase in diabetic human skeletal muscle,22 but -subunit phosphorylation has not previously been shown to occur in cardiac muscle. To investigate the possibility of phosphorylation, tryptic peptides of these 6 spots obtained from 2-D gels resolved for 3 separate rabbits were enriched for phosphopeptides by IMAC and analyzed by MS/MS for phosphorylation-site identification.
As the amino acid sequence of rabbit ATP synthase has not previously been determined, it was essential to clone it from rabbit cardiac muscle allowing the accurate matching of the resulting MS/MS spectra and mapping the positions of the phosphorylated amino acid residues. The first 478 amino acids were obtained (supplemental Figure II), corresponding to all but the last 50 amino acids of all known mammalian sequences (human, bovine, rat, mouse). This protein is highly conserved, differing from the human sequence by only 13 amino acids (>97% identity). Analysis of MS/MS spectra revealed that ATP synthase is phosphorylated, at up to 5 amino acid residues in the rabbit heart (Figure 3B and supplemental Figures III and IV). Phosphorylated serine (S) and threonine (T) residues were identified at the following positions in the rabbit sequence (followed by the corresponding position in the mature rat protein in italics): S106 (S56), T107 (T57), T262 (T212), or S263 (S213), T312 (T262), and T368 (T318). Each of these sites is conserved across many mammalian species (supplemental Figure II).
The locations of these novel phosphorylated residues were mapped onto the x-ray crystal 3D structure of rat F1 ATP synthase (96% identical to rabbit sequence) to determine their accessibility in the oligomeric state of the F1 complex. Based on this model,20 2 of the phosphorylated amino acid residues, T312 (T262) and T368 (T318), are not accessible in the fully assembled F1 catalytic complex (of which 1 , 3 , and 3 subunits constitute the core), indicating that these sites must be phosphorylated before assembly (supplemental Figure V). The remaining 3 residues, S106 (S56), T107 (T57), and T262/S263 (T212/S213), are accessible and therefore have the potential to be dynamically regulated in the fully assembled state during adenosine treatment. S106 (S56) is located in a loop structure on top of the enzyme within binding distance of the subunit (at E84 in the mature rat protein). The T262/S263 (T212/S213) sites are located on the side of the enzyme in a loop structure that interacts with another region of the subunit (at N207 in the rat mature protein).
Discussion
This study was undertaken to assess the cellular response to pharmacological treatment of adenosine and diazoxide, both of which result in a PC phenotype. Although the protein changes encompassed a wide number of functional groups, it was the mitochondrial proteins (10/19 identified changes) that dominated the proteome changes (Figure 4). It is not surprising that the mitochondria is involved in PC, because several proposed effectors of PC have been either proteins residing, or processes occurring, within mitochondria.8,23–25 Signaling events including various kinases (protein kinase C [PKC] isoforms, phosphatidylinositol 3-kinase [PI3K], protein kinase G [PKG], mitogen-activated protein kinase [MAPK], and others) are involved in transmitting the PC signal from the cell surface to the cytosol and mitochondria26–30 and are activated within minutes of exposure to a PC stimulus.31 However, less is known about the ultimate protein targets of these activated kinases. There is strong evidence supporting a role for several candidate protein complexes in the mitochondria specifically, the mitochondrial permeability transition pore protein complex made up of voltage-dependent anion channel (VDAC), hexokinase II, and adenine nucleotide translocase (ANT) which has been shown to bind PKC.32 As well, the succinate dehydrogenase protein complex, which displays functional properties that are similar to the mitochondrial KATP channel, which on opening (initiated by diazoxide) induces PC.33 What is not clear is the breadth of the PC targets and how these differ among PC agents. It is striking that the protein changes observed in this broad overview of the myocyte proteome are a defined subset of proteins. Within the mitochondria, the changes are not sweeping, with only a few enzymes targeted, and, in fact, specific subunits of these protein complexes modulated, by the PC stimuli. The specificity of this response indicates that these changes are not random events that may arise for nonspecific NO effects or oxidation but, rather, are targeted effects. The adenosine or diazoxide induced changes were observed in both high abundance proteins (eg, ATP synthase subunit or MLC-1) as well as in low-abundance proteins (eg, ADP-ribosyl hydrolase or the pyruvate dehydrogenase E3-binding protein), again supporting a targeted action by these drugs. Listed in the online data supplement (supplemental Table II) are the proteins previously observed to be connected to PC (9/19) or ischemia injury (8/19). The majority of the protein changes described in this study have not been previously connected to PC or ischemia. (See the online data supplement for further discussion of protein functions and possible implications.)
There are 3 protein changes to PDH or TCA cycle subunits (Figure 2) as well as alterations to some, but not all, subunits of the OxPhos complexes I and V, with a single change in complex III (Figure 2). Complex I and II have primarily been shown to be inhibited by a number of PC agents.34–38 Metabolic effects were proposed in the initial report of PC1 and further substantiated as a slowing of metabolism.39,40 On reperfusion, energetic processes are also preserved by PC, ATP levels recover more rapidly,41,42 mitochondrial respiration is better preserved,43,44 and postischemic recovery is improved through enhanced coupling between energy production and utilization.45 Several OxPhos complexes have been mechanistically linked to PC and this study links PC to individual subunits abundance changes or PTMs. These data can lead to more in-depth mechanistic and functional studies of how and why these particular subunits are modulated in PC.
The OxPhos pathway is the primary source of ROS formation in higher organisms,46 specifically at complexes I and III,47 and both of these complexes contained altered subunits (Figure 2B). Slowing of electron transport by partial inhibition of either complex I or III decreases electron transfer efficiency and thereby promotes ROS formation, a mechanism by which other PC mimetics have been proposed to protect.48,49 It is important to note that the OxPhos complexes contain large numbers of integral membrane proteins not easily amenable to 2-DE, so it is possible that not every change was detected by this method.
OxPhos complex V, or F1Fo ATP synthase, has also been mechanistically linked with PC. In an isolated rat heart model, the protective effect of ischemia or diazoxide PC (at 20 μmol/L) was associated with lowered ATP synthase activity.50 PC doses of diazoxide (ranging from 20 to 50 μmol/L) also reduce ATP synthase activity in isolated liver51 and heart mitochondria.38,50 Three of the 4 ATP synthase subunits we identified (d, , and ) were altered by PC (Figure 2), making them strong candidates for further investigation into defining the mechanism of the inhibitory effect of PC on ATP synthase. It is not clear how increases in the d and subunits might influence ATP synthase activity. The d subunit is part of the Fo H+ channel that links electron transport to ATP production, whereas the subunit forms part of the stator that stabilizes the ATP catalysis portion of F1.52
The extensive phosphorylation of the subunit is the first report of phosphorylation of this critical enzyme in cardiac tissue and connects a PC stimulus to a PTM of a protein within the mitochondria. Interesting, this subunit has been linked to PC of neuronal cell cultures. PC of cortical neuronal culture with erythropoietin,53 heat stress,54 and MK80154 lead to decreases in 1 protein spot of the ATP synthase subunit. (See the online data supplement, Table II.) These studies only analyzed the changing protein spots and did not provide data for the unchanged spot identifications; thus these findings could indicate a reduced concentration or change in PTM status. This finding also renews the emphasis of ATP production as either an initiator or effector of PC, as initially speculated in previous PC research.40–42
To better understand the significance of the phosphorylated residues, they were modeled onto the 3D x-ray crystallographic structure of the rat F1 catalytic unit20 (supplemental Figure V). The clustering and symmetry of the phosphorylated residues of the subunit is striking, with all of the residues aligning within each subunit. All 3 of the accessible phosphorylated residues are located in loop structures that form structural interactions with both the subunit (S106/T107 [S56/T57]) and within the subunit (T262 and S263 [T212 and S213]) and have the potential to regulate function when modulated by adenosine treatment. Experiments to test the functional consequences of these phosphorylations are important to understanding the true function in PC and are ongoing. This finding implies that endogenous mitochondrial kinase or phosphatase activities are influenced during the 60 minutes adenosine treatment and could have larger implications for PC models and signaling to the mitochondria.
Three proteins in addition to ATP synthase were found to have PC-induced PTMs. As reported previously, adenosine increases the proportion of phosphorylated MLC-1,13 and diazoxide has now been shown to confer the same effect (Figure 2D). ADP-ribosyl hydrolase, including a PTM form, is increased by adenosine and diazoxide. Not previously known to be modified, ADP-ribosyl hydrolase has a broad array of functions. It is among many enzymes involved in NAD+ metabolism and also competitively inhibits OxPhos complex I,55 both of which could influence electron transport during PC. Further analysis is required to uncover its role in PC and to identify the PTM and its potential effect on enzyme activity. As well, diazoxide-induced increase in the PTM form of PDH E3-binding protein (Figure 2A) could lead to reduced PDH activity via the well-documented phosphorylation of the E1 subunit.21 As described in greater detail in the online data supplement, a chain of events arising from E3-binding protein modification would decrease acetyl coenzyme A (CoA) production, necessary for maintenance of metabolic flux through the TCA cycle.
Implications for PC
Proteome changes induced by 2 different modes of PC in the same model show a distinctive pattern of affected proteins consistent with specific perturbation of mitochondrial metabolism through changes to a selected number of mitochondrial protein complexes. Interestingly, specific mitochondrial subunits of the targeted protein complexes were altered, whereas other remained unaffected, indicating complexity and specificity of the drug response. This may be a general phenomenon. As well, the demonstration that inner mitochondrial protein, ATP synthase subunit, can be phosphorylated within 60 minutes of treatment with adenosine is intriguing with respect to how kinase/phosphatase activity is regulated within this segregated compartment of the cell. Although mitochondria is central to PC, it is does not have an exclusive role. Our study suggest that, at least, these PC agents have a wider cellular affect but equally selective response to other subproteomes, which suggests these are not random events (Figure 4). This study provides an overall assessment of the cellular response adenosine and diazoxide treatment and identifies several protein targets that are both unexpected and novel.
Acknowledgments
We thank Gilles Lajoie (University of Western Ontario, Siebens-Drake Research Institute) for MS/MS contributions to this study.
Sources of Funding
D.K.A. was the recipient of a doctoral research award from the Heart and Stroke Foundation of Ontario and a postdoctoral fellowship from the Heart and Stroke Foundation of Canada. J.E.V.E. acknowledges the following funding sources: the Canadian Institute of Health Research, the Canadian Heart and Stroke Foundation, the National Heart Lung Blood Institute Proteomic Initiative (contract NO-HV-28120), and the Donald P. Amos Family Foundation. P.L.P. acknowledges funding from the following sources: National Institute of Health CA-O1O951 and P01 HL081427.
Disclosures
None.
Footnotes
This manuscript was sent to Joseph Loscalzo, Consulting Editor, for review by expert referees, editorial decision, and final disposition.
These authors contributed equally to this work.
Original received February 24, 2006; resubmission received August 11, 2006; accepted August 22, 2006.
References
Murry CE, Jennings RB, Reimer KA. Preconditioning with ischemia: a delay of lethal cell injury in ischemic myocardium. Circulation. 1986; 74: 1124–1136.
Murry CE, Richard VJ, Jennings RB, Reimer KA. Myocardial protection is lost before contractile function recovers from ischemic preconditioning. Am J Physiol. 1991; 260: H796–H804. [Order article via Infotrieve]
Kuzuya T, Hoshida S, Yamashita N, Fuji H, Oe H, Hori M, Kamada T, Tada M. Delayed effects of sublethal ischemia on the acquisition of tolerance to ischemia. Circ Res. 1993; 72: 1293–1299.
Marber MS, Latchman DS, Walker JM, Yellon DM. Cardiac stress protein elevation 24 hours after brief ischemia or heat stress is associated with resistance to myocardial infarction. Circulation. 1993; 88: 1264–1272.
Kloner RA, Jennings RB. Consequences of brief ischemia: stunning, preconditioning, and their clinical implications: part 2. Circulation. 2001; 104: 3158–3167.
Yellon DM, Alkhulaifi AM, Pugsley WB. Preconditioning the human myocardium. Lancet. 1993; 342: 276–277. [Order article via Infotrieve]
Ikonomidis JS, Tumiati LC, Weisel RD, Mickle DA, Li RK. Preconditioning human ventricular cardiomyocytes with brief periods of simulated ischaemia. Cardiovasc Res. 1994; 28: 1285–1291. [Order article via Infotrieve]
Yellon DM, Downey JM. Preconditioning the myocardium: from cellular physiology to clinical cardiology. Physiol Rev. 2003; 83: 1113–1151.
Graham DR, Elliott ST, Van Eyk JE. Broad-based proteomic strategies: a practical guide to proteomics and functional screening. J Physiol. 2005; 563: 1–9.
Liu Y, Sato T, O’Rourke B, Marban E. Mitochondrial ATP-dependent potassium channels: novel effectors of cardioprotection Circulation. 1998; 97: 2463–2469.
Sato T, Sasaki N, O’Rourke B, Marban E. Adenosine primes the opening of mitochondrial ATP-sensitive potassium channels: a key step in ischemic preconditioning Circulation. 2000; 102: 800–805.
National Research Council. Institute of Laboratory Animal Resources Commission on Life Sciences. Guide for the Care and Use of Laboratory Animals. Wash DC: National Academy Press; 1996.
Arrell DK, Neverova I, Fraser H, Marban E, Van Eyk JE. Proteomic analysis of pharmacologically preconditioned cardiomyocytes reveals novel phosphorylation of myosin light chain 1. Circ Res. 2001; 89: 480–487.
Neverova I, Van Eyk JE. Application of reversed phase high performance liquid chromatography for subproteomic analysis of cardiac muscle. Proteomics. 2002; 2: 22–31. [Order article via Infotrieve]
Graham DR, Garnham CP, Fu Q, Robbins J, Van Eyk JE. Improvements in two-dimensional gel electrophoresis by utilizing a low cost "in-house" neutral pH sodium dodecyl sulfate-polyacrylamide gel electrophoresis system. Proteomics. 2005; 5: 2309–2314. [Order article via Infotrieve]
Shevchenko A, Wilm M, Vorm O, Mann M. Mass spectrometric sequencing of proteins silver-stained polyacrylamide gels. Anal Chem. 1996; 68: 850–858. [Order article via Infotrieve]
Dobransky T, Brewer D, Lajoie G, Rylett RJ. Phosphorylation of 69-kDa choline acetyltransferase at threonine 456 in response to amyloid-beta peptide 1–42. J Biol Chem. 2003; 278: 5883–5893.
Ficarro SB, McCleland ML, Stukenberg PT, Burke DJ, Ross MM, Shabanowitz J, Hunt DF, White FM. Phosphoproteome analysis by mass spectrometry and its application to Saccharomyces cerevisiae. Nat Biotechnol. 2002; 20: 301–305. [Order article via Infotrieve]
Zhang C, Williams EH, Guo Y, Lum L, Beachy PA. Extensive phosphorylation of Smoothened in Hedgehog pathway activation. Proc Natl Acad Sci U S A. 2004; 101: 17900–17907.
Bianchet MA, Hullihen J, Pedersen PL, Amzel LM. The 2.8-A structure of rat liver F1-ATPase: configuration of a critical intermediate in ATP synthesis/hydrolysis. Proc Natl Acad Sci U S A. 1998; 95: 11065–11070.
Roche TE, Hiromasa Y, Turkan A, Gong X, Peng T, Yan X, Kasten SA, Bao H, Dong J. Essential roles of lipoyl domains in the activated function and control of pyruvate dehydrogenase kinases and phosphatase isoform 1. Eur J Biochem. 2003; 270: 1050–1056.
Hojlund K, Wrzesinski K, Larsen PM, Fey SJ, Roepstorff P, Handberg A, Dela F, Vinten J, McCormack JG, Reynet C, Beck-Nielsen H. Proteome analysis reveals phosphorylation of ATP synthase beta -subunit in human skeletal muscle and proteins with potential roles in type 2 diabetes. J Biol Chem. 2003; 278: 10436–10442.
Opie LH, Sack MN. Metabolic plasticity and the promotion of cardiac protection in ischemia and ischemic preconditioning. J Mol Cell Cardiol. 2002; 34: 1077–1089. [Order article via Infotrieve]
Minners J, McLeod CJ, Sack MN. Mitochondrial plasticity in classical ischemic preconditioning-moving beyond the mitochondrial KATP channel. Cardiovasc Res. 2003; 59: 1–6. [Order article via Infotrieve]
Marin-Garcia J, Goldenthal MJ. Mitochondria play a critical role in cardioprotection. J Card Fail. 2004; 10: 55–66. [Order article via Infotrieve]
Vondriska TM, Klein JB, Ping P. Use of functional proteomics to investigate PKC epsilon-mediated cardioprotection: the signaling module hypothesis. Am J Physiol Heart Circ Physiol. 2001; 280: H1434–H1441.
Robinet A, Hoizey G, Millart H. PI 3-kinase, protein kinase C, and protein kinase A are involved in the trigger phase of beta1-adrenergic preconditioning. Cardiovasc Res. 2005; 66: 530–542. [Order article via Infotrieve]
Cuong DV, Kim N, Youm JB, Joo H, Warda M, Lee JW, Park WS, Kim T, Kang S, Kim H, Han J. Nitric oxide-cGMP-protein kinase G signaling pathway induces anoxic preconditioning through activation of ATP-sensitive K+ channels in rat hearts. Am J Physiol Heart Circ Physiol. 2006; 290: H1808–H1817.
Xuan YT, Guo Y, Zhu Y, Wang OL, Rokosh G, Messing RO, Bolli R. Role of the protein kinase C-epsilon-Raf-1-MEK-1/2-p44/42 MAPK signaling cascade in the activation of signal transducers and activators of transcription 1 and 3 and induction of cyclooxygenase-2 after ischemic preconditioning. Circulation. 2005; 112: 1971–1978.
Ballard-Croft C, Kristo G, Yoshimura Y, Reid E, Keith BJ, Mentzer RM Jr, Lasley RD. Acute adenosine preconditioning is mediated by p38 MAPK activation in discrete subcellular compartments. Am J Physiol Heart Circ Physiol. 2005; 288: H1359–H1366.
Solenkova NV, Solodushko V, Cohen MV, Downey JM. Endogenous adenosine protects preconditioned heart during early minutes of reperfusion by activating Akt. Am J Physiol Heart Circ Physiol. 2006; 290: H441–H449.
Baines CP, Song CX, Zheng YT, Wang GW, Zhang J, Wang OL, Guo Y, Bolli R, Cardwell EM, Ping P. Protein kinase Cepsilon interacts with and inhibits the permeability transition pore in cardiac mitochondria. Circ Res. 2003; 92: 873–880.
Ardehali H, Chen Z, Ko Y, Mejia-Alvarez R, Marban E. Multiprotein complex containing succinate dehydrogenase confers mitochondrial ATP-sensitive K+ channel activity. Proc Natl Acad Sci U S A. 2004; 101: 11880–11885.
Berman MC, Kewley CF, Kench JE. Contribution of inhibition of NADH-dehydrogenase to the cardiotoxic effects of halothane. J Mol Cell Cardiol. 1974; 6: 39–47. [Order article via Infotrieve]
Hanley PJ, Mickel M, Loffler M, Brandt U, Daut J. K(ATP) channel-independent targets of diazoxide and 5-hydroxydecanoate in the heart. J Physiol. 2002; 542: 735–741.
Hanley PJ, Ray J, Brandt U, Daut J. Halothane, isoflurane and sevoflurane inhibit NADH: ubiquinone oxidoreductase (complex I) of cardiac mitochondria. J Physiol. 2002; 544: 687–693.
Schafer G, Wegener C, Portenhauser R, Bojanovski D. Diazoxide, an inhibitor of succinate oxidation. Biochem Pharmacol. 1969; 18: 2678–2681. [Order article via Infotrieve]
Dzeja PP, Bast P, Ozcan C, Valverde A, Holmuhamedov EL, Van Wylen DG, Terzic A. Targeting nucleotide-requiring enzymes: implications for diazoxide-induced cardioprotection. Am J Physiol Heart Circ Physiol. 2003; 284: H1048–H1056.
Murry CE, Richard VJ, Reimer KA, Jennings RB. Ischemic preconditioning slows energy metabolism and delays ultrastructural damage during a sustained ischemic episode. Circ Res. 1990; 66: 913–931.
Jennings RB, Sebbag L, Schwartz LM, Crago MS, Reimer KA. Metabolism of preconditioned myocardium: effect of loss and reinstatement of cardioprotection. J Mol Cell Cardiol. 2001; 33: 1571–1588. [Order article via Infotrieve]
Miyamae M, Fujiwara H, Kida M, Yokota R, Tanaka M, Katsuragawa M, Hasegawa K, Ohura M, Koga K, Yabuuchi Y, Sasayama S. Preconditioning improves energy metabolism during reperfusion but does not attenuate myocardial stunning in porcine hearts. Circulation. 1993; 88: 223–234.
Fryer RM, Eells JT, Hsu AK, Henry MM, Gross GJ. Ischemic preconditioning in rats: role of mitochondrial K(ATP) channel in preservation of mitochondrial function. Am J Physiol Heart Circ Physiol. 2000; 278: H305–H312.
Yabe K, Nasa Y, Sato M, Iijima R, Takeo S. Preconditioning preserves mitochondrial function and glycolytic flux during an early period of reperfusion in perfused rat hearts. Cardiovasc Res. 1997; 33: 677–685. [Order article via Infotrieve]
Laclau MN, Boudina S, Thambo JB, Tariosse L, Gouverneur G, Bonoron-Adele S, Saks VA, Garlid KD, Dos Santos P. Cardioprotection by ischemic preconditioning preserves mitochondrial function and functional coupling between adenine nucleotide translocase and creatine kinase. J Mol Cell Cardiol. 2001; 33: 947–956. [Order article via Infotrieve]
Pucar D, Dzeja PP, Bast P, Juranic N, Macura S, Terzic A. Cellular energetics in the preconditioned state: protective role for phosphotransfer reactions captured by 18O-assisted 31P NMR. J Biol Chem. 2001; 276: 44812–44819.
Chance B, Sies H, Boveris A. Hydroperoxide metabolism in mammalian organs. Physiol Rev. 1979; 59: 527–605.
Cadenas E, Boveris A, Ragan CI, Stoppani AO. Production of superoxide radicals and hydrogen peroxide by NADH-ubiquinone reductase and ubiquinol-cytochrome c reductase from beef-heart mitochondria. Arch Biochem Biophys. 1977; 180: 248–257. [Order article via Infotrieve]
Mullenheim J, Ebel D, Frassdorf J, Preckel B, Thamer V, Schlack W. Isoflurane preconditions myocardium against infarction via release of free radicals. Anesthesiology. 2002; 96: 934–940. [Order article via Infotrieve]
Novalija E, Varadarajan SG, Camara AK, An J, Chen Q, Riess ML, Hogg N, Stowe DF. Anesthetic preconditioning: triggering role of reactive oxygen and nitrogen species in isolated hearts. Am J Physiol Heart Circ Physiol. 2002; 283: H44–H52.
Ala-Rami A, Ylitalo KV, Hassinen IE. Ischaemic preconditioning and a mitochondrial KATP channel opener both produce cardioprotection accompanied by F1F0-ATPase inhibition in early ischaemia. Basic Res Cardiol. 2003; 98: 250–258. [Order article via Infotrieve]
Schafer G, Portenhauser R, Trolp R. Inhibition of mitochondrial metabolism by the diabetogenic thiadiazine diazoxide. I. Action on succinate dehydrogenase and TCA-cycle oxidations. Biochem Pharmacol. 1971; 20: 1271–1280. [Order article via Infotrieve]
Boyer PD. The ATP synthase—a splendid molecular machine. Annu Rev Biochem. 1997; 66: 717–749. [Order article via Infotrieve]
Meloni BP, Tilbrook PA, Boulos S, Arthur PG, Knuckey NW. Erythropoietin preconditioning in neuronal cultures: signaling, protection from in vitro ischemia, and proteomic analysis. J Neurosci Res. 2006; 83: 584–593. [Order article via Infotrieve]
Meloni BP, Van Dyk D, Cole R, Knuckey NW. Proteome analysis of cortical neuronal cultures following cycloheximide, heat stress and MK801 preconditioning. Proteomics. 2005; 5: 4743–4753. [Order article via Infotrieve]
Zharova TV, Vinogradov AD. A competitive inhibition of the mitochondrial NADH-ubiquinone oxidoreductase (complex I) by ADP-ribose. Biochim Biophys Acta. 1997; 1320: 256–264. [Order article via Infotrieve]
您现在查看是摘要介绍页，详见ORG附件。