Identification, Quantification, and Functional Aspects of Skeletal Muscle Protein-Carbonylation in Vivo during Acute Oxidative Stress Maria Fedorova,† Nadezhda Kuleva,‡ and Ralf Hoffmann*,† Institute of Bioanalytical Chemistry, Center for Biotechnology and Biomedicine, Faculty of Chemistry and Mineralogy, Leipzig University, Deutscher Platz 5, 04103 Leipzig, Germany, and Department of Biochemistry, Faculty of Biology and Soil Science, St. Petersburg State University, Universitetskaya nab. 7/9, 199034 St. Petersburg, Russia Received December 21, 2009
Reactive oxidative species (ROS) play important roles in cellular signaling but can also modify and often functionally inactivate other biomolecules. Thus, cells have developed effective enzymatic and nonenzymatic strategies to scavenge ROS. However, under oxidative stress, ROS production is able to overwhelm the scavenging systems, increasing the levels of functionally impaired proteins. A major class of irreversible oxidative modifications is carbonylation, which refers to reactive carbonyl-groups. In this investigation, we have studied the production and clearance rates for skeletal muscle proteins in a rat model of acute oxidative stress over a time period of 24 h using a gel-based proteomics approach. Optimized ELISA and Western blots with 10-fold improved sensitivities showed that the carbonylation level was stable at 4 nmol per mg protein 3 h following ROS induction. The carbonylation level then increased 3-fold over 6 h and then remained stable. In total, the oxidative stress changed the steady state levels of 20 proteins and resulted in the carbonylation of 38 skeletal muscle proteins. Carbonylation of these proteins followed diverse kinetics with some proteins being highly carbonylated very quickly, whereas others peaked in the 9 h sample or continued to increase up to 24 h after oxidative stress was induced. Keywords: Actin • enzyme-linked immunosorbent assay (ELISA) • myosin • reactive oxygen species (ROS) • skeletal muscle proteins
For many years, it was believed that reactive oxidative species (ROS), highly reactive molecules capable of causing modifications to proteins, lipids, and DNA, were only harmful to cellular structures. ROS can be produced by many different chemical reactions originating from cellular events, e.g., metal-catalyzed reactions, mitochondrial electron transport reactions, and neutrophil or macrophage activation during inflammation,1 or external events, e.g., UV light, X-rays, or pollutants. ROS as natural byproduct of the oxygen metabolism play important roles in cellular signaling,2,3 similar to reactive nitrogen species (RNS).1 Excessive production of ROS within the cells, caused, for example, in muscles by extensive exercises, or reduced capture capabilities of the antioxidative systems, can lead to oxidative stress.4 This deleterious condition is defined as an imbalance between ROS production and antioxidant defense systems. Proteins can, therefore, be severely damaged, as they capture 50% to 75% of all highly reactive species with functional groups in their side chains.5 These reactions, among the diverse ROS and different functional protein groups, as well as the consecu-
tive reactions among the resulting primary oxidation products, are not well understood, especially as the primary oxidation products can undergo consecutive reactions with other functional groups of the same protein or other molecules.6 One group of reactions is protein carbonylation, which refers to the production of highly reactive carbonyl groups in proteins that are reactive to hydrazine compounds.7 This irreversible modification includes aldehydes and ketones formed via different mechanisms: (i) direct oxidation of the polypeptide backbone leading to truncated peptides with N-terminal R-ketoacyl amino acid residues; (ii) oxidation of the side chains of lysine, arginine, proline, and threonine residues; (iii) reaction of histidine, cysteine, and lysine amino acid residues with aldehydes, e.g., produced by lipid peroxidation; and (iv) glycation (nonenzymatic glycosylation) of lysine residues forming Amadori and Heyns products. Protein carbonylation is used in many studies as a primary marker for oxidative stress. It can be easily quantified with 2,4-dinitrophenylhydrazine (DNPH), which forms 2,4-dinitrophenyl (DNP) derivatives with an absorption maximum at 370 nm. DNP-modified proteins can also be quantified with specific anti-DNP-antibodies.8–10
* Corresponding author. Tel: #49 (0) 341 9731330. Fax: #49 (0) 341 9731339. E-mail:
[email protected]. † Leipzig University. ‡ St. Petersburg State University.
Protein carbonylation is an irreversible oxidative modification that cannot be reversed by the enzymatic repair machinery of cells. Thus, the number of carbonyl groups present in proteins correlates well with the general protein damage caused
Introduction
2516 Journal of Proteome Research 2010, 9, 2516–2526 Published on Web 04/08/2010
10.1021/pr901182r
2010 American Chemical Society
Carbonylation of Muscle Proteins 6
by oxidative stress. In general, the carbonyl-group content increases significantly with age, when up to one-third of protein molecules can be modified.6,11 Carbonylation, therefore, appears to be a biomarker for aging,12,13 as well as for several diseases.7 Skeletal muscles are a good model for oxidative stress, as ROS are produced by many different cascades at elevated levels during muscle contractions, such as the mitochondrial electron transport chain,14 phospholipase A2,15 NAD(P)H oxidase, and metabolism of arachidonic acid by the lipoxygenase pathway.16 At the same time, muscle cells have very effective enzymatic systems to deal with ROS, such as superoxide dismutase, catalase, glutathione peroxidase, thioredoxin and thioredoxin reductase, and peroxiredoxin,17,18 which can prevent further damage. These enzymes are complimented by low-mass antioxidants, such as tocopherol (vitamin E), carotenes, and ubiquinol within cell membranes, and ascorbic acid (vitamin C), uric acid (2,6,8-trioxypurine), lipoic acid, and glutathione distributed in the myocyte.19 Significantly increased carbonylation of muscle proteins has been demonstrated in several studies: exercise-induced muscle carbonylation for different limb and respiratory muscles, for both short-term and chronic muscle activation; prolonged immobilization characterized by disuse muscle atrophy; and reduced expression of main cytoplasmic antioxidant proteins.20 Increased carbonylation levels were also found in skeletal muscles of patients with chronic obstructive pulmonary disease (COPD),21 obstructive sleep apnea,22 and sepsis.23 The functional aspects, however, remain unclear. Recently, an interesting proteomic study on muscle-type and age-related carbonyl modifications in rat skeletal muscles revealed how carbonylation affects protein functions.24 In this study, the influence of ROS on the expression level of proteins in skeletal hind leg muscle has been studied using a proteomics approach, using a recently established in vivo model of acute oxidative stress induced by X-ray irradiation.25,26 The carbonylation level of proteins identified by mass spectrometry was determined using ELISA and blotting techniques. Furthermore, the protein expression and protein carbonylation levels determined for different time points, provided specific kinetics for each protein that could be correlated to the protein functions in different cellular pathways.
Material and Methods Reagents. All reagents for electrophoresis, PMSF, protease inhibitor mix, trypsin, Tween-20, and BSA were obtained from Serva Electrophoresis GmbH (Heidelberg, Germany). All reagents for IEF were purchased from Bio-Rad Laboratories GmbH (Mu ¨nchen, Germany). Sodium and potassium salts were obtained from Sigma-Aldrich Chemie GmbH (Taufkirchen, Germany). CHAPS, DTT, SDS, glycerol, TCA, and ethanol were purchased from Carl Roth GmbH & Co. KG (Karlsruhe, Germany) and the other organic solvents from Biosolve VB (Valkenswaard, Germany). Urea and thiourea were obtained from Fluka Chemie GmbH (Buchs, CH) and Tris base from Calbiochem (Schwalbach, Germany). Animal Model.25,26 Male Wistar rats were housed in groups of five animals per cage at 22 ( 3 °C and exposed to 12 h of light per day. The animals were given free access to a granular basal standard diet and water ad libitum. Six month old rats (200-250 g body weight, 183 days (5 days old) were irradiated in a RUM-17 X-ray unit (200 kV, 13 mA, 20 cm focal distance; Mosrentgen, Moscow, USSR) at a dose rate of 1.75 Gy/min for
research articles a total dose of 5 Gy per animal. The rats were sacrificed after 3, 9, and 24 h by decapitation. Nonirradiated control rats (referred to as 0 h) had the same age and sex as the irradiated animals. For each of the four time points, a group of five animals was studied. Protein Preparation. Proteins were extracted from muscle tissue pooled from the five animals of each group using a high salt buffer (0.6 M KCl, 10 mM Na2CO3, and 40 mM NaHCO3; 20 min) and precipitated overnight with ice-cold acetone. The pellets were dissolved in buffer (7 M urea, 2 M thiourea, 2% CHAPS (w/v), 50 mM DTT, 8 mM PMSF, 0.16 µg/mL protease inhibitor mix, and 50 mM Tris-HCl, pH 7.5), and insoluble particles were removed by centrifugation (10,000g, 15 min, 4 °C). The protein concentration of the supernatant was determined by a Bradford assay.27 Two-Dimensional (2D) Gel Electrophoresis. All four samples were analyzed in triplicate in parallel in both dimensions, i.e., IEF and SDS-PAGE, using the Protean IEF Cell and ProteanPlus Dodeca Cell (Bio-Rad Laboratories GmbH, Munich Germany), respectively. Fifty micrograms of proteins were dissolved in rehydration buffer (330 µL; 7 M urea, 2 M thiourea, 2% CHAPS, 50 mM dithiothreitol (DTT), and 0.2% Bio-Lyte 3-10) and loaded onto IPG strips (17 cm, pH 3-10 NL). After active rehydration (50 V, 12 h), the proteins were focused in three steps: 250 V for 15 min, 4,000 V for 2 h, and finally 10,000 V for 50,000 V h. The strips were incubated in equilibration buffer (0.375 M Tris-HCl, pH 8.8, 6 M urea, 2% (w/v) SDS, and 20% (v/v) glycerol) in the presence of 2% (w/v) DTT for 15 min and then in fresh equilibration buffer containing 2.5% (w/v) iodoacetamide. The equilibrated strips containing S-alkylated proteins were washed with electrode buffer (25 mM Tris, pH 8.3, 192 mM glycine, and 0.1% (w/v) SDS) and loaded onto a 20 cm gel to separate the proteins by SDS-PAGE (12% T, 2.67% C, 1 mm thickness) in the second dimension (10 h, 80 V). SDS-PAGE was stopped when the dye marker migrated to the bottom of the gels, which were then stained with colloidal Coomassie Brilliant Blue G254. Gels were scanned on an Image scanner (GE Healthcare Europe GmbH, Munich, Germany). Scanned gel images were warped and matched by the Delta2D software tool (Decodon, Germany) and quantified by the TIGR Multiple Experimental Viewer algorithm included in Delta2D. The experimental setup was checked by the hierarchical sample clustering algorithm. Protein expression profiles were determined by template matching using the Pearson correlation. The relative expression values for control and irradiated samples were set to 0.5 and 1, and the significance (p-value) of the correction coefficient was set to 0.05. Western Blot Detection of Carbonylated Proteins.10 The samples with a protein content of 10 µg were diluted in rehydration buffer to a final volume of 125 µL and loaded onto IPG strips (7 cm, 3-10 NL), as described above for the 17 cm strips, except that the final focusing step was done at 4,000 V for 20,000 V h. The gel strips were derivatized in parallel with DNPH solution (1 mg/mL in 2 M aqueous HCl) (20 min, room temperature (RT)) on a rotary shaker. Strips were neutralized by washing for 15 min with 30% (v/v) glycerol in Tris-buffer (2 M). The following washing, reduction and alkylation steps, as well as SDS-PAGE (0.75 mm thickness), were performed as described above. The proteins were transferred to a PVDF membrane (SemiDry Trans-Blot SD; Bio-Rad) and blocked with blocking buffer (5% milk powder and 0.2% TPBS, overnight, 4 °C). The membrane was incubated with primary goat anti-DNP antibody Journal of Proteome Research • Vol. 9, No. 5, 2010 2517
research articles (Ab) (0.4 µg/mL; Sigma-Aldrich Chemie GmbH, Munich, Germany) dissolved in blocking buffer (1 h, RT). Detection was performed with peroxidase-conjugated donkey antigoat IgG Ab (0.16 µg/mL in blocking buffer; Dianova GmbH, Hamburg) and ECL Advance western Blotting Detection Kit (GE Healthcare Europe), following the supplier’s instructions. Membranes were developed on CL-XPosure Film (Perbio Science Deutschland GmbH, Bonn, Germany) with Kodak D-19 Film Developer and Kodak Fixer (Sigma-Aldrich). The blot images were analyzed and matched to the gels with the Delta 2D software package. All steps from 2-DE over blotting until film development were done in parallel for all time points of a given experiment to allow a more reliable quantification. Pathways and Functional Correlation Analysis of Carbonylated Proteins. Correlation of protein functions to different cellular pathways of carbonylated proteins were performed with MetaCore, version 5.4 (GeneGo Inc., St. Joseph, MI, USA). Quantification of Aldehydes and Ketones. Carbonyl groups present in all actin samples were quantified with DNPH by ELISA in parallel on a 96-well plate.9 The plate was coated with 100, 50, 25, or 12 ng protein per well in PBS buffer (8 mM Na2HPO4, 1 mM KH2PO4, and 150 mM NaCl, pH 7.5) and stored overnight at 4 °C. The plate was washed three times with 0.05% (v/v) Tween-20 in PBS (0.05% TPBS), and an aqueous DNPH solution (0.05 mM, pH 6.2, freshly prepared) was then added and the plate incubated in the dark (45 min, RT). The plate was then washed five times with a mixture of ethanol and PBS (1:1 by vol, 300 µL per well), three times with 0.05% TPBS (300 µL per well), blocked with 0.2% TPBS (300 µL per well, 1 h, RT), and washed once more with 0.05% TPBS (300 µL per well). One hundred microliters of goat anti-DNP antibody solution (Sigma-Aldrich; 0.4 µg/mL in 0.1% TPBS) was then added and the plate incubated (1 h, 37 °C). The plate was washed once more with 0.05% TPBS and incubated with peroxidaseconjugated donkey antigoat secondary antibody solution (Dianova; 0.16 µg/mL, 0.1% TPBS). The plate was then rewashed three times with 0.05% TPBS before 1-Step Ultra TMB-ELISA solution (Perbio Science; 100 µL per well) was added. The plate was then incubated in the dark (15 min, RT) and the reaction stopped with sulfuric acid (0.5 M, 100 µL per well). The absorption of each well was recorded at 450 nm with a SUNRISE Microplate reader (TECAN, Trading AG, Mannedorf, Switzerland). Carbonyl group content was calculated from an eight step dilution series of oxidized BSA containing 17.5 nmol/ mg carbonyl-groups (see below). Preparation of Standard Carbonylated BSA Solution for ELISA Calibration. BSA was oxidized by the Fenton reaction to produce aldehyde and keto groups, i.e., active carbonyl groups.28 A BSA solution (50 mg/mL in PBS, 10 µL) was incubated with water peroxide (10 mM) and ferric sulfate (1 mM) for 2 days (37 °C). The protein concentration was determined by a Bradford assay, and the content of active carbonyl groups was quantified with DNPH (see below). Oxidized BSA with a carbonyl-content of 17.5 nmol/mg was used as the standard for the direct ELISA. Protein Carbonylation Assay. Reactive carbonyl groups, i.e., aldehydes or ketones present in oxidized BSA samples, were derivatized with DNPH and quantified by recording the absorptions at 370 nm for the formed hydrazones.29 Briefly, the protein sample (1 mL, 0.7-1 mg/mL) was mixed with DNPH (1 mL, 10 mM) in hydrochloric acid (2 M). The control consisted of the same protein solution (1 mL) mixed with an equal 2518
Journal of Proteome Research • Vol. 9, No. 5, 2010
Fedorova et al. volume of hydrochloric acid (2 M). Samples were incubated (1 h, RT), and the proteins were precipitated with trichloroacetic acid (TCA, 20% (w/v) final concentration) on ice (30 min) and centrifuged (10 min, 21 000g). Pellets were washed three times with a mixture of ethanol and ethylacetate (1:1, v/v) to remove unreacted DNPH. The washed pellets were dissolved (1 mL, 6 M urea in phosphate buffer, pH 2.3), and the absorption was recorded at 370 nm against a blank. The content of carbonyl groups was determined with a molar extinction coefficient of 22,000 M-1 cm-1. MALDI Mass Spectrometry. Bands of interest were excised from the Coomassie stained gels, and an in-gel trypsin digest was performed.30 The tryptic peptides were dissolved in 60% aqueous acetonitrile containing 0.5% formic acid. An aliquot of this solution (0.5 µL) was mixed with an equal volume of R-cyano-4-hydroxy-cinnamic acid solution (4 mg/mL in 50% aqueous acetonitrile; Bruker Daltonics GmbH, Bremen, Germany), and 1 µL was spotted onto a MALDI target and airdried. The mass spectra were recorded on a MALDI-TOF-MS (4700 proteomic analyzer, Applied Biosystems GmbH, Darmstadt, Germany) operated in positive ion-reflector TOF (reTOF-) mode (acceleration voltage of 20 kV, 70% grid voltage, 1.277 ns delay, detector voltage of 2 kV) in the m/z range from 700 to 4,000 with a focus mass of 1,700. Typically, 40 subspectra with 50 laser shots each were accumulated at a fixed laser intensity of 5,000. The instrument was calibrated on the same plate using six calibration spots containing the 4700 Analyzer Calibration mixture (Applied Biosystems). Product ion spectra of selected precursor ions were acquired in reflector TOF/TOF mode (acceleration voltage of 8 kV, 70% grid voltage, 1.277 ns delay, detector voltage of 2.1 kV, and collision energy of 1 kV) by accumulation of 6,000 laser shots at a fixed laser intensity of 5,500. Mass spectra were analyzed with the Data Explorer software package (Version 4.6, Applied Biosystems). Proteins were identified with the MASCOT software package (matrixscience.com) in the Swiss-Prot database, allowing up to two missed cleavage sites and a mass tolerance of 100 ppm. For Mascot, the taxonomy setting was Rattus, and variable modifications were methionine oxidation and carbamidomethylation on cysteine.
Results Total Carbonyl Content. All male rats used in this study were sacrificed, either without irradiation (the control group) or 3, 9, and 24 h following irradiation. Skeletal muscles from the hind legs were prepared immediately after sacrifice and the proteins extracted under high salt conditions to obtain sarcoplasmic and myofibrillar proteins, including actin and myosin. The protein carbonylation level, universally quantified with DNPH, is generally accepted as a primary marker of oxidative stress. One drawback, however, is its limited sensitivity, requiring protein amounts in the µg-range, even in a recently introduced ELISA technique.9 Therefore, in this study, the protocol was modified by adding Tween-20 in PBS, instead of reduced BSA, as a blocking agent. This reduced the unspecific background significantly. Together with some other optimizations, the sensitivity was increased 10-fold, down to 100 ng of tissue proteins containing only 400 fmol reactive carbonyl groups. The calibration curve for oxidized BSA (17.5 pmol carbonyl per µg BSA) was linear from 50 pmol (2.8 µg of BSA) down to 100 fmol (6 ng of BSA) reactive carbonyl-groups per well (Figure 1, left panel), with a standard deviation of 7.68%. The skeletal muscle protein sample from the control animals contained about 4
Carbonylation of Muscle Proteins
research articles
Figure 1. Reactive carbonyl (CO) contents determined in oxidized BSA (left) and skeletal muscle protein preparations (right) by ELISA. Protein samples were diluted in PBS, coated on microtiter plates, and derivatized with DNPH in the wells. After blocking, the DNPHreactive carbonyl-content was probed with an anti-DNP antibody and visualized with a POD-conjugated secondary antibody. A serial dilution of BSA oxidized by the Fenton reaction was used for calibration. Skeletal proteins were prepared from nonirradiated male Wistar rats and rats 3, 9, or 24 h after X-ray irradiation.
Figure 2. Colloidal Coomassie stained 2-DE gels of skeletal muscle protein samples prepared from nonirradiated animals (control, panel A) and from animals 3 (B), 9 (C), and 24 h (D) post-irradiation. Fifty micrograms of protein was focused on each IPG strip (17 cm, 3-10 NL) and separated by SDS-PAGE in the second dimension (12% T, 2.67% C, 20 cm length, and 1 mm thickness). The shown gel images were representative for the triplicates analyzed in parallel.
nmol/mg DNPH-reactive carbonyl-groups, which did not increase during the first 3 h following irradiation. The carbonylcontent in the protein samples prepared 9 and 24 h following irradiation increased to around 14 and 20 nmol/mg, respectively (Figure 1, right panel). Protein Expression. Differences in the protein patterns and expression levels were studied by 2-DE for all four animal groups in order to identify the impact of ROS on the muscle proteins, including truncated or cross-linked protein versions. In general, the proteins were well resolved in all areas of the gel with no streaking, and the protein patterns appeared to be very similar between the four samples (Figure 2). The 440 most
intense spots were present at all four time points. The dominant spot in the acidic region, at an apparent molecular mass of approximately 40 kDa, contained actin. Matching all spots in the 2D-gels (triplicates), for all four time points, revealed a total of 22 spots containing 11 different proteins that had been down-regulated by a statistically significant degree following irradiation (Figure 3, left panel, and Table 1; details in Supporting Information, Table S1), relative to the control. In addition, 13 spots, corresponding to 9 proteins, had been upregulated to a statistically significant degree following irradiation, relative to the control (Figure 3, right panel, and Table 2; details in Supporting Information, Table S2). None of the spot Journal of Proteome Research • Vol. 9, No. 5, 2010 2519
research articles
Fedorova et al.
Figure 3. Two-dimensional gel images of the control (left) and 9 h (right) skeletal muscle preparations as shown in Figure 2 with spots of lower (22 spots marked on the left image) or higher intensities (13 spots marked on the right image). Gel images were matched and relatively quantified with the Delta2D software package as described in Materials and Methods. Table 1. Spot Intensity Ratios of Skeletal Muscle Proteins down-Regulated Following Irradiation, Relative to the Nonirradiated Controla ratio spot
proteins identified by MS
3 h/C
9 h/C
24 h/C
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22
heat shock cognate 71 kDa protein skeletal muscle-specific R-actinin 3 skeletal muscle-specific R-actinin 3 skeletal muscle-specific R-actinin 3 serum albumin precursor myosin binding protein H glycogen phosphorylase, muscle form glycogen phosphorylase, muscle form glycogen phosphorylase, muscle form glycogen phosphorylase, muscle form glycogen phosphorylase, muscle form aconitate hydratase (aconitase) aconitate hydratase (aconitase) collagen R-1 chain collagen R-1 chain collagen R-1 chain pyruvate kinase isozymes M1/M2 troponin T, fast skeletal muscle troponin LIM domain-binding protein 3 LIM domain-binding protein 3 LIM domain-binding protein 3 NADH-ubiquinone oxidoreductase 75 kDa subunit
0.49 0.24 0.20 0.19 0.54 0.36 0.45 0.41 0.27 0.29 0.18 0.30 0.24 0.06 0.10 0.16 0.44 0.35 0.78 0.86 0.88 0.30
0.29 0.13 0.26 0.29 0.29 0.42 0.24 0.29 0.25 0.41 0.30 0.47 0.44 0.05 0.02 0.12 0.76 0.62 0.80 0.67 0.70 0.10
0.75 0.47 0.50 0.59 0.61 0.40 0.66 0.56 0.33 0.35 0.32 0.68 0.57 0.09 0.09 0.13 0.63 0.53 0.56 0.53 0.41 1.22
a The presented ratios for the spot intensities of the 3 h, 9 h, and 24 h samples versus the control were calculated with the software package Delta2D based on three gel replications.
intensities returned to the control level within the studied time interval of 24 h. These 35 spots were cut from the gels, destained, digested with trypsin, and identified by MALDI-MS using peptide mass fingerprinting, together with at least three tandem mass spectra (Tables 1 and 2; details in Supporting Information, Tables S3 and S4). Most down-regulated proteins remained at the same low level over all three time points (Table 1). Only the LIM domain-binding protein 3 (spots 19 to 21) decreased further. The intensities of all other protein spots corresponding to skeletal muscle-specific R actinin 3 and aconitate hydratase partially recovered. The up-regulated protein spots showed different kinetics with only one (spot. 2520
Journal of Proteome Research • Vol. 9, No. 5, 2010
Nr. 9, Table 2 and Figure 3) being stable over all time points within the error range, whereas spots 2, 4, 5, 7, and 8 decreased, spot 11 increased, and spots 1, 3, 6, 10, 12, and 13 reached a maximum after 9 h (Table 2). In both cases, even spots containing the same protein followed different intensity kinetics for the different time points, e.g., glycogen phosphorylase (muscle form, Table 1; Supporting Information, Figure S-1, panel A) and the different myosin light chains (Table 2). In addition, some proteins (spots 3, 8, 10, 12, and 13, Table 2; Supporting Information, Figure S-1, panel B) were shifted to slightly higher molecular masses and more acidic pH values, indicating post-translational modifications.
research articles
Carbonylation of Muscle Proteins
Table 2. Spot Intensity Ratios of Skeletal Muscle Proteins Up-Regulated Following Irradiation, Relative to the Nonirradiated Controla ratio spot
1 2 3 4 5 6 7 8 9 10 11 12 13
proteins identified by MS
3 h/C
9 h/C
24 h/C
myosin light chain 3, skeletal muscle isoform myosin regulatory light chain 2, skeletal muscle isoform myosin regulatory light chain 2, skeletal muscle isoform myosin regulatory light chain 2, ventricular/cardiac muscle isoform troponin T, slow skeletal muscle troponin troponin T, slow skeletal muscle troponin myosin light polypeptide 3 myosin light polypeptide 3 probable C f U-editing enzyme APOBEC-2 Actin, R-skeletal muscle ATP synthase subunit R fructose-bisphosphate aldolase A fructose-bisphosphate aldolase A
2.73 1.97 3.52 3.00
7.14 1.93 3.79 2.65
2.53 1.46 2.33 1.30
2.35 1.78 2.28 4.33 1.54 1.20 1.10 1.42 1.84
1.86 1.94 1.68 2.97 1.24 2.67 1.44 1.91 4.90
1.63 1.36 1.87 2.35 1.45 1.85 1.75 1.28 2.19
a The presented ratios for the spot intensities of the 3 h, 9 h, and 24 h samples versus the control were calculated with the software package Delta2D based on three gel replications.
Protein Carbonylation. The standard approach used to detect carbonylated proteins following gel electrophoresis relies on electrotransfer onto a PVDF membrane followed by derivatization with DNPH. The obtained hydrazone derivatives are then detected with an anti-DNP-antibody, as used above in the ELISA, and visualized, e.g., with colorimetric or chemiluminescence techniques. In this way, carbonylated proteins can be detected with reasonable sensitivities for skeletal muscle proteins.8 Chemiluminescence detection, which typically improves the sensitivity of immunoblots at least 10-fold, did not provide a better sensitivity due to the high background level. Such high background levels also prevented the expected higher sensitivities of biotin-hydrazide and digoxigenin-3-0succinyl-ε-aminocapronic acid (DIG)-hydrazide (data not shown). Thus, we used a different derivatization strategy by labeling the proteins with DNPH within the IPG strips after the first dimension.10 The reaction time was optimized to prevent unspecific reactions with nonoxidized proteins and to obtain a high sensitivity for carbonylated species (data not shown). In this way, a total of 60 spots were visualized in the control samples on the PVDF membrane (Figure 4, panel A; Supporting Information Figure S2), in contrast to around 15 proteins in our previous report using standard conditions.26 Very similar spot intensities were obtained for the 3 h sample, confirming the ELISA results that the CO-level did not increase during this time period. Moreover, the spot pattern was very similar to the control, i.e., all detectable CO-containing proteins were present at comparable levels. This changed dramatically after 9 h (Figure 4, panel C), at which point the spot intensities had generally increased. This increase clearly indicated a higher carbonyl-content, as the loaded protein quantities were identical, on the basis of the Bradford assay and the spot intensities in the Coomassie stained 2D-gel. Several new spots had appeared after 9 h, especially in the upper mass range, and neutral to basic pH range. The spot intensities in the blot of the 24 h sample had increased by a slight degree, with the spot patterns being almost identical. Only in the high molecular mass range above 100 kDa and between pH 7 and 8 were a few more intense spots obvious. This result may signify the presence of cross-linked proteins, formed by the reaction of the carbonyl-groups with the side chains from other proteins. All blots (three for each time point) were warped and matched with Delta2D and then relatively quantified with the
same software tool (Supporting Information, Table S5). These blots were then further matched to the corresponding colloidal Coomassie-stained 2D-gels, in order to cut the subsequent spots and digest them for mass spectrometric identification. In total, proteins contained in 128 spots were analyzed, i.e., 23 in the control sample, 26 in the 3 h post-irradiation sample, 35 in the 9 h sample, and 43 in the 24 h sample (Figure 4; details in Supporting Information, Tables S6, S7, S8, and S9). Altogether 38 different proteins were identified (Table 3) on the basis of a PMF combined with at least three MS/MS spectra recorded on the MALDI-TOF/TOF-MS. The set of carbonylated proteins was analyzed for known functions of each protein with the MetaCore software tool and grouped by functional correlations, biological pathways and interaction analyses: 15 proteins belonging to muscle contraction; 15 to carbohydrate metabolism (glycolysis, gluconeogenesis and TCA cycle); and six to the ATP metabolic/catalytic processes (Table 3). Additionally, carbonic anhydrase 3, serum albumin precursor, collagen R-1 chain, and two heat shock proteins (heat shock cognate 71 kDa protein and 78 kDa glucose-regulated protein) were identified as being carbonylated (Table 3). The recorded MS/MS spectra permitted protein identification only, and could not provide information on carbonylation sites. Interestingly, some of the carbonylated proteins were down-regulated, e.g. glycogen phosphorylase, pyruvate kinase, R actinin and heat shock cognate 71 kDa protein, which may indicate a higher clearance rate.
Discussion Animal Model. The animal model used in this study was established three years ago in order to study the damaging effects of high ROS concentrations in vivo,25,26 monitored by three different early markers of oxidative stress: (i) lipid peroxide levels in muscle, liver, kidney, and plasma samples, (ii) DNPH-reactive carbonyl-groups, and (iii) dityrosines. In a recent study, we were able to further demonstrate that the disulfide patterns of muscle proteins are also severely altered26 (Supporting Information, Table S10). Many proteins were crosslinked by disulfide bridges, whereas others formed intramolecular disulfide bridges. Interestingly, different proteins followed different kinetics within the studied time period of 24 h. In this study, we have extended these earlier investigations in Journal of Proteome Research • Vol. 9, No. 5, 2010 2521
research articles
Fedorova et al.
Figure 4. Western-blots probed with anti-DNP antibodies to detect DNPH-reactive carbonyl groups in skeletal muscle protein preparations. The control (A), 3 h (B), 9 h (C), and 24 h (D) samples (10 µg protein each) were separated by 2-DE and semidry blotted on PVDF membranes. Following IEF, the carbonylated proteins were derivatized with DNPH in the strips (7 cm, 3-10 NL) for 20 min before the strips were transferred to SDS-PAGE. The PVDF membrane was first probed with a goat anti-DNP polyclonal serum and, after washing, incubated with donkey antigoat IgG Ab conjugated with POD. The membranes were stained with the ECL Advance Western blot detection kit. Marked spots are those that were matched to the colloidal Coomassie stain of the corresponding 2D-gels by the Delta2D software package and ultimately identified by tandem mass spectrometry following an in-gel trypsin digest.
order to identify carbonylated proteins reactive with DNPH. The hydrazine reagent reacts with aldehydes and ketones at similar rates31 and also with lactams, whereas hydryazides are selective for aldehydes. Thus, DNPH labels not only glutamic and aminoadipic semialdehydes formed from arginine, proline, and lysine residues but also 2-pyrrolidone (lactame formed from proline) and 2-amino-3-ketobutyric acid (ketone derived from threonine). A recently reported cross-reactivity of DNPH to sulfenic acids, produced by oxidation of cysteine residues,32 was prevented here by prior reduction with DTT. This reduces the unspecific background significantly, as sulfenic acids are formed continuously in cells and are thus present in all protein preparations at relatively high levels, and are further elevated by oxidative stress. Together with the 10-fold improved sensitivities of the optimized ELISA and Western blot protocols, we were able to specifically detect several carbonylated proteins even without induced oxidative stress. Moreover, the ELISA and Western-blot data were consistent over the studied time period. Here, the influence of a massive ROS burden induced in healthy male rats was studied for the expression profiles and the carbonylation levels of more than 400 skeletal muscle protein versions separated by 2-DE. Protein Expression. Altogether, 35 spots in the 2D-gels, corresponding to a total of 20 proteins, changed their relative intensities significantly, typically 2- to 5-fold, within the first 2522
Journal of Proteome Research • Vol. 9, No. 5, 2010
3 h. In particular, the up-regulated proteins showed a clear trend toward the original concentration. Intensity of most down-regulated protein spots also increased over time, although they typically stayed at a relatively low level, between 10% and 60% of the original spot intensity. None of the affected proteins returned to the original level within the 24 h period studied, and the effects appeared to be independent of protein function. Most of the identified proteins regulate muscle contractions. Among these, R-actinin (Figure 3, left panel, spots 2 to 4), myosin binding protein H (spot 6), fast skeletal muscle troponin T (spot 18), and LIM-domain binding proteins (spots 19 to 21) were present at lower quantities (down-regulated) (Table 1), whereas several isoforms of myosin light chains and slow skeletal muscle troponin T were detected at higher amounts (up-regulated) (Figure 3, right panel, spots 1 to 8, and Table 2). As the terms down- and up-regulated are commonly used in the literature to indicate that protein spots in 2D gels are present at lower or higher intensities, respectively, we use these terms here accordingly. It should be noted, however, that the quantitative changes most likely do not reflect only the protein translation but may also depend on higher or lower degradation rates due to oxidative modifications, cross-linking, and aggregation. Although it is not clear which parameter is more determinative, the slight changes between 3 and 24 h, with a tendency toward the control level, indicate that protein
research articles
Carbonylation of Muscle Proteins
Table 3. Carbonylated Skeletal Muscle Proteins Identified in Samples Prepared from Control and Irradiated Animals (3 h, 9 h, or 24 h Post-Irradiation) and Corresponding Spot Numbers on the Western Blotsa spot on WB C
3h
9h
Muscle Contraction 1 2
3
3 6
4,5 9, 11 13
8, 9 12, 15, 13 17, 18, 20
10, 14 7 23 22
19 10 25 1
24 h
3 1 1 2 5 8 6, 7 9 10, 14, 15 17 16 11 35 28, 29,32
5 7, 10 6 8, 9, 11, 12,16, 17, 23, 25 14, 15, 18, 20, 21 22 16, 19 43 36, 41 2
proteins identified by MS
myosin light chain 1, skeletal muscle isoform myosin light chain 3, skeletal muscle isoform myosin regulatory light chain 2, skeletal muscle isoform myosin light polypeptide 3 tropomyosin R-1 and β chains high molecular weight of tropomyosin actin, R-skeletal muscle high molecular weight aggregates of R actin myosin-4 myosin-binding protein C, fast-type skeletal muscle-specific R-actinin 3 desmin troponin I, fast skeletal muscle troponin T, fast skeletal muscle troponin T, slow skeletal muscle
Carbohydrate Metabolism: Glycolysis, Gluconeogenesis, and TCA Cycle pyruvate dehydrogenase complex 2 4 4 4 pyruvate dehydrogenase E1 component subunit β 14 dihydrolipoyllysine-residue acetyltransferase component of pyruvatedehydrogenase complex 32 dihydrolipoyl dehydrogenase 2-oxoglutarate dehydrogenase complex 12 11 12 24 dihydrolipoyllysine-residue succinyltransferase component of 2-oxoglutarate dehydrogenase complex 32 dihydrolipoyl dehydrogenase 26 R-enolase 18 22 22 34 -βenolase 15 19 30 glycogen phosphorylase, muscle form 17 20 31 pyruvate kinase isozymes M1/M2 21 24 30 39 glyceraldehyde-3-phosohate dehydrogenase 20 25, 27 38 fructose-bisphosphate aldolase A 31 40 L-lactate dehydrogenase A chain
maxb
3h 9 hc 9 hc 3h satc 9h Sat.c 9/24 hc 24 h 9h 24 hc 3 hc 9h 9h 9h
3h 3 hc 9h 9/24 h 9h 24 h 3h 24 h 9h 9h 24 h 9h
Energy Metabolism 6 19
23
Miscellaneous 26 21 16 16 8
21 9 23, 24 26 34 33 18 13
33 35 37
42 27 28, 29 13
ATP metabolic/catabolic processes ATP synthase subunit β (complex V) ATP synthase subunit R (complex V) NADH-ubiquinone oxidoreductase 75 kDa subunit (complex I) creatine kinase M-type creatine kinase, sarcomeric adenylate kinase isoenzyme 1
C 9-24 h 24 h 3-24 h 3-24 h 9h
carbonic anhydrase 3 serum albumin precursor collagen R-1 chain heat shock cognate 71 kDa protein 78 kDa glucose-regulated protein
3-24 h 24 h C/24 h 24 h Cc
a Note that the spots of each sample have been numbered from 1 upwards (Figure 4). The protein functions were correlated with the MetaCore analysis software. The maximal intensity was calculated from the Western blots (WB) after 2-DE with the software package Delta2D. b max denotes the sample that displayed the maximal intensity of this spot in the Western blot. c Denotes spots or areas which were saturated on Western blots and were thus not quantified with Delta2D but their content estimated manually. “sat” indicates that this spot was saturated in the blots from all samples, and thus, the maximum intensity could not be determined.
translation in these proteins is rather slow. This is also in agreement with the published half-lives of these proteins, which are typically several hours to days (Supporting Information, Table S11). Down-regulation of R-actinin and LIM-binding protein should mediate structural changes of the contractile muscle machinery, as they are both located in the Z-line of sarcomers that maintain the organization of the contractile machinery. However, such changes have not been reported for muscle pathologies, such as sarcopenia and aging. Down-regulation of fast-skeletal muscle troponin T and up-regulation of slowskeletal muscle troponin T within 3 h could indicate a rapid transition from fast to slow muscle fibers that was only partially reversed. Similar effects were also observed in recent proteomic studies on aging.33–35 Slow twitch muscle fibers are reportedly
more resistant to fatigue and are characterized by a more aerobic metabolism. Such a metabolic switch was also obvious from three key enzymes of the carbohydrate metabolism (Figure 5). Glycogen dephosphorylase (spots 7-11), pyruvate kinase M1 (spot 18), and aconitate hydratase (spots 12 and 13) were down-regulated at all time points relative to the control (Tab. 1). Among these enzymes, pyruvate kinase M1, which mediates one of the ratelimiting steps of muscle glycolysis (Figure 5), was drastically down-regulated and recovered slowly to approximately 70% (Table 2). This down-regulation appears to be a general effect for oxidative stress-related disorders. This phenomenon was also used as a marker for oxidative metabolism due to the inhibition of glycolysis.36 The intensities of all five spots containing glycogen phosphorylase showed similar time deJournal of Proteome Research • Vol. 9, No. 5, 2010 2523
research articles
Fedorova et al.
Figure 5. Carbohydrate metabolism pathway consisting of glycolysis, gluconeogenesis, TCA cycle, and pentose phosphate pathway to illustrate functional changes of enzymes down-regulated (V) or carbonylated (gray background) in skeletal muscles, as a consequence of acute oxidative stress induced by X-ray irradiation in vivo.
pendence. As this enzyme cleaves glycogen to glucose during extensive muscle contractions, its reduced level may also indicate a shift toward a more aerobic metabolism. The only up-regulated enzyme was aldolase A (Figure 3, right panel, spots 13 and 14, and Table 2), which catalyzes the cleavage of fructose-1,6-bisphosphate to glycerinaldehyde-3-phosphate and dihydoxyacetone phosphate (Figure 5). Both spots were shifted by the pH and molecular mass, which indicated negatively charged modifications, although it was not possible to identify a modification site by mass spectrometry in the digested protein spots. Protein Carbonylation. The carbonylation level increased over all time points in both ELISA and Western blots. Whereas the ELISA only provides a global picture for the carbonylation degree of a protein preparation, the high resolution of 2D in combination with the Western blot, allowed a reliable identification of carbonylated proteins. The relative quantification of all spots typically showed a continuous increase of the carbonylation level, which was mostly independent of the protein. This indicates that ROS were present in all parts of the skeletal muscle cells and could act on all proteins, independent of their sequence. In total, 20 proteins were 2524
Journal of Proteome Research • Vol. 9, No. 5, 2010
stained with anti-DNP-antibodies on the Western-blot of the control sample, whereas another 18 proteins appeared at later time points post-irradiation. The identified proteins fall mostly into three classes: muscle contraction, carbohydrate metabolism, and ATP-metabolic/catabolic processes (Table 3). Only carbonylation of desmin and the myosin light chains occurred quickly, detected already after 3 h. All other spots remained stable at this time point. The highest carbonylation degrees were detected at 9 or 24 h. The relatively low carbonylation level of myosin light chains is surprising, as we were able to show recently that myosin light chains MLE1, MRLC, and MYL3 contained a large degree of oxidized cysteine residues, i.e., intra- and intermolecular disulfide bridges, in the same animal model.26 The content of disulfides increased even over the studied 24 h time period. Although surprising at first, this finding might explain their low carbonylation degree. The high content of free cysteine residues may effectively capture ROS near myosin, thereby protecting it from further oxidations, such as carbonylation. The different troponin T isoforms showed the highest carbonylation degree at 9 h. Again, the fast-skeletal muscle troponin isoforms appeared to be more prone to carbonylation than the slow-skeletal isoforms, which is in good
research articles
Carbonylation of Muscle Proteins 37
agreement with a recent study on aging. This observation can be explained by the lower superoxide scavenging capacity of fast muscle fibers.38 Carbonylation of muscle-specific proteins has been demonstrated in a number of studies for pathological conditions, including ischemia reperfusion, diabetes, chronic obstructive pulmonary disease, sepsis, and aging, e.g., for actin,8,39 myosin light and heavy chains,40,41 desmin,42 and tropomyosin.43,44 The second major group of carbonylated proteins consisted of enzymes involved in carbohydrate metabolism (Figure 5), i.e., glycolysis, TCA-cycle, and glycogenesis. Among them were five glycolitic enzymes (enolase, pyruvate kinase, glyceraldehydes-3-phosphate dehydrogenase, and fructose 1,6-bisphosphate aldolase), all three subunits of the pyruvate dehydrogenase complex, and one enzyme complex of the TCA-cycle (two subunits of the 2-oxoglutarate dehydrogenase complex), lactate dehydrogenase, and glycogen phosphorylase. The glycolytic enzymes followed different carbonylation kinetics. The maximum carbonylation level had already been reached for three proteins 3 h after irradiation (β-enolase and subunits ODPB and ODP2 of pyruvate dehydrogenase) and for others after 9 h (pyruvate kinase, glyceraldehydes-3-phosphate dehydrogenase, and dihydrolypoil dehydrogenase). Only the carbonylation degree of aldolase increased continuously over all time points. Similar observations were reported for a model of sepsis. There, a significant increase in enolase carbonylation was already detected after 1 h, whereas aldolase carbonylation was clearly enhanced over a time period of 6 h following sepsis induction.45 Carbonylation is an irreversible oxidative modification that cannot be repaired by cellular enzymes. Proteins oxidized under mild oxidative stress can be degraded directly by the 20S proteasome without ubiquitination, whereas modestly oxidized proteins are degraded rapidly by proteases. Strong oxidative stress produces highly carbonylated proteins that cannot be degraded and tend to aggregate. Considering these aspects, the skeletal muscle cell is most likely able to keep the carbonylation degree at a low level for the first 3 h, before reduced proteasomal and protease activities result in elevated carbonylation levels (9 h) that are further increased by the formation of protein aggregates.
Conclusions Acute oxidative stress changes the protein expression patterns and carbonylation levels of skeletal muscle proteins within 3 h. Almost 10% of the protein spots detected by 2-DE (35 from around 440) had altered intensities following irradiation. Eleven proteins were detected at higher levels and 9 at lower levels, which recovered only partially within the studied time interval. Altogether, carbonylation affected 36 proteins, of which 20 were already carbonylated in the control samples at a low level. Functionally, most affected proteins are involved either in the regulation of muscle contractions or are part of carbohydrate metabolism. The modified or differentially expressed proteins, however, followed different kinetic pathways, indicating a complex regulation pattern that depends not only on the ROS levels but also on ROS scavenging, protein degradation, and protein translation.
Acknowledgment. A Ph.D.-stipend to Maria Fedorova provided by “Gottlieb Daimler- und Karl Benz-Stiftung” and financial support from the European Fond for Regional Structure Development (EFRE, European Union and Free State Saxony) are gratefully acknowledged. We thank
GeneGo Inc. for providing temporary free access to the MetaCore software tool and Dr. Christina Nielsen-Marsh for proofreading. We are grateful to Prof. V. M. Mikhelson (Institute of Cytology, St. Petersburg) for his help to irradiate the animals.
Supporting Information Available: Selected areas of the colloidal Coomassie stained 2D-gels for spots corresponding to muscle isoform of glycogen phosphorylase and fructose 1,6-bisphosphate aldolase from control animals and animals 3, 9, or 24 h after X-ray irradiation; original Western-blots probed with anti-DNP antibodies to detect DNPH-reactive carbonyl groups in skeletal muscle protein preparations; original Delta2D values, i.e., ratio, Student’s t-test, and relative standard deviation (RSD) for down-regulated proteins (refers to Table 1); original Delta2D values, i.e., ratio, Student’s t-test, and relative standard deviation (RSD) for up-regulated proteins (refers to Table 2; identification of proteins down-regulated by oxidative stress in spots 1 to 21 of the 2-DE presented in Figure 3 (left panel) by peptide mass fingerprinting (PMF) and tandem mass spectrometry (MS/MS); identification of proteins upregulated by oxidative stress in spots 1 to 14 of the 2-DE presented in Figure 3 (right panel) by peptide mass fingerprinting (PMF) and tandem mass spectrometry (MS/MS); original Delta2D values, i.e., ratio, Student’s t-test, and relative standard deviation (RSD) for relative quantification of carbonylated proteins; identification of carbonylated proteins in the control sample by peptide mass fingerprinting (PMF) and tandem mass spectrometry (MS/MS); identification of carbonylated proteins in the 3 h sample by peptide mass fingerprinting (PMF) and tandem mass spectrometry (MS/MS); identification of carbonylated proteins in the 9 h sample by peptide mass fingerprinting (PMF) and tandem mass spectrometry (MS/MS); identification of carbonylated proteins in the 24 h sample by peptide mass fingerprinting (PMF) and tandem mass spectrometry (MS/MS); t test for TBA assay as provided in ref 26; carbonylated skeletal muscle proteins identified in samples prepared from control and irradiated animals (3 h, 9 h, or 24 h post irradiation) and corresponding spot numbers on the Western blots. This material is available free of charge via the Internet at http://pubs.acs.org. References (1) Pfeilschifter, J.; Eberhardt, W.; Huwiler, A. Nitric oxide and mechanisms of redox signalling: matrix and matrix-metabolizing enzymes as prime nitric oxide targets. Eur. J. Pharmacol. 2001, 429 (1-3), 279–286. (2) D’Autreaux, B.; Toledano, M. B. ROS as Signalling Molecules: Mechanisms That Generate Specificity in ROS Homeostasis. Nature Rev. Mol. Cell Biol. 2007, 8 (10), 813–824. (3) Thannickal, V. J.; Fanburg, B. L. Reactive oxygen species in cell signaling. Am. J. Physiol. Lung Cell Mol. Physiol. 2000, 279 (6), L1005–L1028. (4) Moylan, J. S.; Reid, M. B. Oxidative stress, chronic disease, and muscle wasting. Muscle Nerve 2007, 35 (4), 411–429. (5) Chandel, N. S.; Budinger, G. R. The cellular basis for diverse responses to oxygen. Free Radical Biol. Med. 2007, 42 (2), 165– 174. (6) Stadtman, E. R.; Levine, R. L. Protein oxidation. Ann. N.Y. Acad. Sci. 2000, 899, 191–208. (7) Dalle-Donne, I.; Giustarini, D.; Colombo, R.; Rossi, R.; Milzani, A. Protein carbonylation in human diseases. Trends Mol. Med. 2003, 9 (4), 169–176. (8) Goto, S.; Nakamura, A.; Radak, Z.; Nakamoto, H.; Takahashi, R.; Yasuda, K.; Sakurai, Y.; Ishii, N. Carbonylated proteins in aging and exercise: immunoblot approaches. Mech. Ageing Dev. 1999, 107 (3), 245–253. (9) Alamdari, D. H.; Kostidou, E.; Paletas, K.; Sarigianni, M.; Konstas, A. G.; Karapiperidou, A.; Koliakos, G. High sensitivity enzyme-
Journal of Proteome Research • Vol. 9, No. 5, 2010 2525
research articles
(10)
(11)
(12)
(13)
(14)
(15)
(16)
(17)
(18)
(19)
(20) (21)
(22)
(23)
(24)
(25)
(26)
(27)
(28)
(29)
2526
linked immunosorbent assay (ELISA) method for measuring protein carbonyl in samples with low amounts of protein. Free Radical Biol. Med. 2005, 39 (10), 1362–1367. Conrad, C. C.; Choi, J.; Malakowsky, C. A.; Talent, J. M.; Dai, R.; Marshall, P.; Gracy, R. W. Identification of protein carbonyls after two-dimensional electrophoresis. Proteomics 2001, 1 (7), 829–834. Levine, R. L.; Wehr, N.; Williams, J. A.; Stadtman, E. R.; Shacter, E. Determination of carbonyl groups in oxidized proteins. Methods Mol. Biol. 2000, 99, 15–24. Kumagai, A.; Nakayashiki, N.; Aoki, Y. Analysis of age-related carbonylation of human vitreous humor proteins as a tool for forensic diagnosis. Legal Med. 2007, 9 (4), 175–180. Rabek, J. P.; Boylston III, W. H.; Papaconstantinou, J. Carbonylation of ER chaperone proteins in aged mouse liver. Biochem. Biophys. Res. Commun. 2003, 305 (3), 566–572. Vasilaki, A.; Mansouri, A.; Remmen, H.; van der Meulen, J. H.; Larkin, L.; Richardson, A. G.; McArdle, A.; Faulkner, J. A.; Jackson, M. J. Free radical generation by skeletal muscle of adult and old mice: effect of contractile activity. Aging Cell 2006, 5 (2), 109–117. Gong, M. C.; Arbogast, S.; Guo, Z.; Mathenia, J.; Su, W.; Reid, M. B. Calcium-independent phospholipase A2 modulates cytosolic oxidant activity and contractile function in murine skeletal muscle cells. J. Appl. Physiol. 2006, 100 (2), 399–405. Zuo, L.; Christofi, F. L.; Wright, V. P.; Bao, S.; Clanton, T. L. Lipoxygenase-dependent superoxide release in skeletal muscle. J. Appl. Physiol. 2004, 97 (2), 661–668. Criswell, D.; Powers, S.; Dodd, S.; Lawler, J.; Edwards, W.; Renshler, K.; Grinton, S. High intensity training-induced changes in skeletal muscle antioxidant enzyme activity. Med. Sci. Sports Exercise 1993, 25 (10), 1135–1140. Powers, S. K.; Criswell, D.; Lawler, J.; Ji, L. L.; Martin, D.; Herb, R. A.; Dudley, G. Influence of exercise and fiber type on antioxidant enzyme activity in rat skeletal muscle. Am. J. Physiol. 1994, 266 (2 Pt 2), R375–R380. Powers, S. K.; Jackson, M. J. Exercise-induced oxidative stress: cellular mechanisms and impact on muscle force production. Physiol. Rev. 2008, 88 (4), 1243–1276. Powers, S. K.; Kavazis, A. N.; McClung, J. M. Oxidative stress and disuse muscle atrophy. J. Appl. Physiol. 2007, 102 (6), 2389–2397. Wouters, E. F. Muscle wasting in chronic obstructive pulmonary disease: to bother and to measure. Am. J. Respir. Crit. Care Med. 2006, 173 (1), 4–5. Barreiro, E.; Nowinski, A.; Gea, J.; Sliwinski, P. Oxidative stress in the external intercostal muscles of patients with obstructive sleep apnoea. Thorax 2007, 62 (12), 1095–1101. Hasselgren, P. O.; Menconi, M. J.; Fareed, M. U.; Yang, H.; Wei, W.; Evenson, A. Novel aspects on the regulation of muscle wasting in sepsis. Int. J. Biochem. Cell Biol. 2005, 37 (10), 2156–2168. Feng, J.; Xie, H.; Meany, D. L.; Thompson, L. V.; Arriaga, E. A.; Griffin, T. J. Quantitative proteomic profiling of muscle typedependent and age-dependent protein carbonylation in rat skeletal muscle mitochondria. J. Gerontol., Ser. A 2008, 63 (11), 1137–1152. Kuleva, N. V.; Fedorova, M. A.; Frolov, A. A. Dynamics of rat hepatocyte histone glycoxidation after general x-ray irradiation. Tsitologyia 2006, 48 (6), 515–521. Fedorova, M.; Kuleva, N.; Hoffmann, R. Reversible and irreversible modifications of skeletal muscle proteins in a rat model of acute oxidative stress. Biochim. Biophys. Acta 2009, 1792, 1185–1193. Bradford, M. M. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal. Biochem. 1976, 72, 248–254. Domingues, M. R.; Domingues, P.; Reis, A.; Fonseca, C.; Amado, F. M.; Ferrer-Correia, A. J. Identification of oxidation products and free radicals of tryptophan by mass spectrometry. J. Am. Soc. Mass Spectrom. 2003, 14 (4), 406–16. Levine, R. L.; Garland, D.; Oliver, C. N.; Amici, A.; Climent, I.; Lenz, A. G.; Ahn, B. W.; Shaltiel, S.; Stadtman, E. R. Determination of
Journal of Proteome Research • Vol. 9, No. 5, 2010
Fedorova et al.
(30)
(31) (32)
(33)
(34) (35)
(36)
(37)
(38) (39) (40)
(41)
(42)
(43)
(44)
(45)
carbonyl content in oxidatively modified proteins. Methods Enzymol. 1990, 186, 464–78. Shevchenko, A.; Jensen, O. N.; Podtelejnikov, A. V.; Sagliocco, F.; Wilm, M.; Vorm, O.; Mortensen, P.; Boucherie, H.; Mann, M. Linking genome and proteome by mass spectrometry: large-scale identification of yeast proteins from two dimensional gels. Proc. Natl. Acad. Sci. U.S.A. 1996, 93 (25), 14440–14445. Avigad, G. Dansyl hydrazine as a fluorimetric reagent for thinlayer chromatographic analysis of reducing sugars. J. Chromatogr. 1977, 139 (2), 343–347. Dalle-Donne, I.; Carini, M.; Orioli, M.; Vistoli, G.; Regazzoni, L.; Colombo, G.; Rossi, R.; Milzani, A.; Aldini, G. Protein carbonylation: 2,4-dinitrophenylhydrazine reacts with both aldehydes/ketones and sulfenic acids. Free Radical Biol. Med. 2009, 46 (10), 1411– 1419. Baldwin, K. M.; Klinkerfuss, G. H.; Terjung, R. L.; Mole, P. A.; Holloszy, J. O. Respiratory capacity of white, red, and intermediate muscle: adaptative response to exercise. Am. J. Physiol. 1972, 222 (2), 373–378. Doran, P.; Donoghue, P.; O’Connell, K.; Gannon, J.; Ohlendieck, K. Proteomics of skeletal muscle aging. Proteomics 2009, 9 (4), 989– 1003. Lombardi, A.; Silvestri, E.; Cioffi, F.; Senese, R.; Lanni, A.; Goglia, F.; de Lange, P.; Moreno, M. Defining the transcriptomic and proteomic profiles of rat ageing skeletal muscle by the use of a cDNA array, 2D- and Blue native-PAGE approach. J. Proteomics 2009, 72 (4), 708–721. Doran, P.; O’Connell, K.; Gannon, J.; Kavanagh, M.; Ohlendieck, K. Opposite pathobiochemical fate of pyruvate kinase and adenylate kinase in aged rat skeletal muscle as revealed by proteomic DIGE analysis. Proteomics 2008, 8 (2), 364–77. Doran, P.; Donoghue, P.; O’Connell, K.; Gannon, J.; Ohlendieck, K. Proteomic profiling of pathological and aged skeletal muscle fibres by peptide mass fingerprinting (review). Int. J. Mol. Med. 2007, 19 (4), 547–564. Anderson, E. J.; Neufer, P. D. Type II skeletal myofibers possess unique properties that potentiate mitochondrial H(2)O(2) generation. Am. J. Physiol. Cell Physiol. 2006, 290 (3), C844–C851. Barreiro, E.; Gea, J.; Di Falco, M.; Kriazhev, L.; James, S.; Hussain, S. N. Protein carbonyl formation in the diaphragm. Am. J. Respir. Cell Mol. Biol. 2005, 32 (1), 9–17. Coirault, C.; Guellich, A.; Barbry, T.; Samuel, J. L.; Riou, B.; Lecarpentier, Y. Oxidative stress of myosin contributes to skeletal muscle dysfunction in rats with chronic heart failure. Am. J. Physiol. Heart Circ. Physiol. 2007, 292 (2), H1009–H1017. Yamada, T.; Mishima, T.; Sakamoto, M.; Sugiyama, M.; Matsunaga, S.; Wada, M. Myofibrillar protein oxidation and contractile dysfunction in hyperthyroid rat diaphragm. J. Appl. Physiol. 2007, 102 (5), 1850–1855. Luo, J.; Hill, B. G.; Gu, Y.; Cai, J.; Srivastava, S.; Bhatnagar, A.; Prabhu, S. D. Mechanisms of acrolein-induced myocardial dysfunction: implications for environmental and endogenous aldehyde exposure. Am. J. Physiol. Heart Circ. Physiol. 2007, 293 (6), H3673–H3684. Rao, V. S.; La Bonte, L. R.; Xu, Y.; Yang, Z.; French, B. A.; Guilford, W. H. Alterations to myofibrillar protein function in nonischemic regions of the heart early after myocardial infarction. Am. J. Physiol. Heart Circ. Physiol. 2007, 293 (1), H654–H659. Kjaersgard, I. V.; Norrelykke, M. R.; Baron, C. P.; Jessen, F. Identification of carbonylated protein in frozen rainbow trout (Oncorhynchus mykiss) fillets and development of protein oxidation during frozen storage. J. Agric. Food Chem. 2006, 54 (25), 9437– 9446. Barreiro, E.; Comtois, A. S.; Mohammed, S.; Lands, L. C.; Hussain, S. N. Role of heme oxygenases in sepsis-induced diaphragmatic contractile dysfunction and oxidative stress. Am. J. Physiol. Lung Cell Mol. Physiol. 2002, 283 (2), L476–L484.
PR901182R
