The identities of proteins that show disuse-related changes in the content of oxidative modification are unknown. Furthermore, it is unknown whether the global accumulation of oxidized proteins is greater in aged animals with muscle disuse. The purposes of this study are 1) to identify the exact proteins that show disuse-related changes in oxidation levels and 2) to test the hypothesis that the global accumulation of oxidized proteins with muscle disuse would be greater in aged animals. Adult and old rats were randomized into four groups: weight bearing and 3, 7, or 14 days of hindlimb unloading. Soleus muscles were harvested to investigate the protein oxidation with unloading. Slot blot, SDS-PAGE, and Western blot analyses were used to detect the accumulation of 4-hydroxy-2-nonenol (HNE)- and nitrotyrosine (NT)-modified proteins. Matrix-assisted laser desorption ionization-time of flight and tandem mass spectroscopy were used to identify modified proteins. We found that global HNE- and NT-modified proteins accumulated significantly with aging but not with muscle unloading. Two HNE and NT target proteins, four-and-a-half LIM protein 1 (FHL1) and carbonic anhydrase III (CAIII), showed changes in the oxidation levels with muscle unloading. The changes in the oxidation levels happened to adult rats but not old rats. However, old rats had higher baseline levels of HNE-modified FHL1. In summary, the data suggest that the muscle unloading-related changes of protein oxidation are more significant in specific proteins and that the changes are age related.
- muscle disuse
- hindlimb unloading
muscle disuse due to immobilization, bed rest, spaceflight, unloading, and mechanical ventilation results in loss of muscle mass (atrophy). Studies show that the disuse-related muscle atrophy is associated with increased oxidative stress (20, 38, 39). Increased oxidative stress results in protein oxidation and the oxidized proteins are subsequently degraded by the ubiquitin-proteasome system. Support for the increased protein oxidation and degradation with muscle disuse is found in investigations using adult animals. These studies report a global accumulation of oxidized proteins (21, 43–45, 56) and ubiquitinated proteins (18, 51) using multiple models of muscle disuse. However, despite knowing the global accumulation of oxidized proteins with muscle disuse, the identity of the proteins that show disuse-related changes in the content of oxidative modification is currently unknown.
Aging is associated with the accumulation of oxidative stress-induced damage in many tissues. Studies demonstrate that older organisms have greater protein oxidation compared with younger counterparts (5, 26). The accumulation of oxidized proteins with aging has been associated with increased oxidant generation (2) and decreased capacity for elimination of oxidized proteins (1, 10, 16). Organisms respond to the age-related increase of oxidative stress by a compensatory increase of antioxidant capacity (6, 12–14, 25, 30). However, the ability of aged organisms to adapt to additional stressors appears to be compromised (23, 25). For example, Leeuwenburgh et al. (25) reported that antioxidant capacity of muscles increases with exercise training in young rats but not in old rats. Our previous study found that whereas the levels of antioxidant glutathione (GSH) remained stable with muscle unloading in adult rats, the levels in old rats decreased dramatically(6). The age-dependent difference in response of the antioxidant system to stimuli raises the possibility that the accumulation of oxidized proteins with muscle disuse would be greater in the aged animals.
The aims of the present study are 1) to investigate whether the global accumulation of oxidized proteins is greater in the aged animals with muscle disuse and 2) to identify proteins that show changes in the oxidation levels with muscle disuse. On the basis of the age-related changes of oxidant generation and antioxidant adaptation, as well as proteolytic capacity, we hypothesized that the global accumulation of oxidized proteins with muscle disuse would be greater in the aged animals.
Animals and Hindlimb Unloading
Forty-nine male Fischer 344 rats ages 13 (100% strain survival; n = 24) and 26 mo (25% strain survival; n = 25) were purchased from the Minneapolis Veterans Affairs Aged Rodent Colony that was maintained by the University of Minnesota. The rats were randomized into four experimental groups: normal weight bearing (control) and hindlimb unloading (HU) for 3, 7, and 14 days. The durations of HU (3, 7, and 14 days) were chosen because there is evidence of age-dependent changes in the antioxidant capacities of the soleus with 7 and 14 days of unloading (6). In addition, muscles of adult rats following 7 days of cast immobilization and 14 days of HU have increased oxidized proteins (21, 43, 44).
The HU intervention was achieved by attaching the tail of the rat to a swivel mounted at the top of the cage. The height of the suspension was adjusted to prevent the hindlimbs from contacting the floor. This arrangement permits animals to move around with their forelimbs while the hindlimbs are unloaded (57). All the animals were housed in a research animal facility and checked daily for any abnormal response to tail suspension. The protocol of this study was approved by the University of Minnesota Institutional Animal Care and Use Committee.
Overall Experimental Strategy to Determine Oxidized Proteins
To determine whether the age of the rat influences the accumulation of oxidized proteins in unloaded muscle and to identify the modified proteins, we selected the soleus muscle. The soleus muscle is composed of predominantly type I fibers that are affected significantly by unloading and show age-related changes (6, 15, 48). The muscle proteins within the soleus were separated experimentally into two fractions, soluble and myofibrillar protein fractions (49). Subsequent individual protein separation (SDS-PAGE), extent of oxidized proteins (Western blotting), and protein identification (mass spectroscopy) are facilitated when the muscles proteins are subfractionated (49).
The global or total accumulation of oxidized proteins in the unloaded muscles was evaluated by Western blot analysis using two experimental approaches, slot blot (total proteins) and SDS-PAGE (proteins separated by molecular weight). 4-Hydroxy-2-nonenol (HNE) and nitrotyrosine (NT) were chosen as two specific markers of protein oxidation and represent two different forms of oxidation. HNE is formed from lipid peroxidation and can modify cysteines, lysines, and histidines (9). NT is the product of tyrosine nitration by peroxynitrite (31). Both of these modifications have been shown to render proteins dysfunctional (7, 17, 52). Mass spectroscopy was used to identify individual proteins that show changes in the content of HNE and NT modifications with muscle unloading.
Muscle tissue was prepared as described (49). Briefly, the rats were anesthetized with pentobarbital sodium (35 mg/kg body wt) after the intervention. Soleus muscles were harvested, weighed, and immediately frozen in liquid nitrogen. The frozen soleus muscles were stored in a −80°C freezer until processing. Muscles were separated into soluble and myofibrillar fractions that contained mainly the cytosolic and myofibrillar proteins, respectively (32). Specifically, a small piece of frozen soleus muscle was pulverized with a mortar and pestle. The pulverized tissue was then homogenized with a glass homogenizer (Kontes Duall) in buffer containing 20 mM imidazole, 2 mM EDTA, and 0.25 mM PMSF (pH 7.4). The supernatant that contained the extracted proteins was collected after centrifugation at 12,000 g for 30 min at 4°C. Buffer containing 2% 3-[(3-chloamidoprophyl)dimethylammonio]-2-hydroxy-1-propanesulfonate and 4 M urea was added into the collected supernatant to ensure complete protein solubilization and prevent aggregation. This fraction of homogenate is called the soluble fraction (49). The pellet was homogenized in buffer containing 10 mM tris(2-carboxyethyl)phosphine and 10% trifluoroacetic acid. The supernatant that contained the extracted proteins was collected after centrifugation at 12,000 g for 30 min at 4°C. This fraction of homogenate is called the myofibrillar fraction. The homogenates were stored in a −80°C freezer. Protein concentrations were determined using the bicinchoninic acid protein assay kit (Pierce) with bovine serum albumin as the standard.
Evaluation of the Global Accumulation of Oxidized Proteins
The global accumulation of HNE- and NT-modified proteins in the soluble and myofibrillar fractions of soleus muscles were evaluated by slot blot analysis and SDS-PAGE followed by Western blot analysis. In slot blots, an equal amount (soluble fraction, 5 μg for HNE and 2.5 μg for NT; myofibrillar fractions, 10 μg for HNE and 5 μg for NT) of protein was absorbed to polyvinylidene difluoride (PVDF) membranes using a Bio-Dot SF microfiltration apparatus (Bio-Rad, Hercules, CA) and following the manufacturer's instructions. These protein loads were experimentally determined to be in the linear range of responses for each antibody and fraction.
In SDS-PAGE, an equal amount (soluble fraction, 9 μg for HNE and 20 μg for NT; myofibrillar fractions, 15 μg for HNE and 18 μg for NT) of protein was loaded onto 5 or 12% SDS-polyacrylamide gels and separated by electrophoresis. The separated proteins were transferred from gels to PVDF membranes using a Transblot SD semidry transfer cell (Bio-Rad) at 800 mA for 35 min. To control equal loading, each sample was loaded into two gels and the two gels were run in parallel. Whereas one gel was used for Western blotting, the other gel was silver stained using a Silver Stain Plus kit (Bio-Rad) to confirm the equal loading of each sample.
In both slot blot and SDS-PAGE, a standard sample (internal control) was loaded and transferred on each blot. The intensity of immune responses of all samples was normalized to the intensity of the standard sample, thus permitting the comparison of samples across multiple blots.
Western blotting for HNE and NT.
The protein bound membranes were incubated overnight at 4°C with polyclonal HNE antibody (Alpha Diagnostics, San Antonio, TX; 1:2,500) or polyclonal NT antibody (Cayman Chemical, Ann Arbor, MI; 1:2,000). Membranes probed with HNE antibody and NT antibody were then incubated with biotinylated goat anti-rabbit secondary antibody (Bio-Rad; 1:3,000) at room temperature for 1 h. After the incubation with secondary antibodies, membranes probed with either HNE or NT antibody were incubated with a streptavidin signal amplification solution (Sigma, St. Louis, MO; 1:3,000) and biotinylated alkaline phosphatase (Bio-Rad; 1:3,000) at room temperature for 1.5 h. Finally, substrate 5-bromo-4-chlor-3′-iodolylphosphate p-toluidine/nitro blue tetrazolium chloride (BCIP-NBT) was used to visualize the immunoreaction on membranes. Membranes were imaged using a GS-800 calibrated densitometer (Bio-Rad), and intensity (slot area and whole lane for slot blot and SDS-PAGE, respectively) was quantified by densitometry (optical density) using Sigma Scan Pro (Systat, Point Richmond, CA) (11, 49).
Identification of Proteins
The overall strategy to identify proteins is the same as described previously (11, 49). Briefly, we matched the individual bands on Western blots that showed HU-related changes in intensity of immune reaction with the bands on the silver-stained gels. After the match, corresponding protein bands on silver-stained gels were excised for mass spectroscopy. In mass spectroscopy, initial protein identification was done by measuring peptide mass using matrix-assisted laser desorption ionization-time of flight mass spectrometry (MALDI-TOF MS) and obtaining an initial identification using MASCOT (http://www.matrixscience.com) and the NCBInr (National Center for Biotechnology Information) rattus database. Peptide tandem mass spectrometry (MS/MS) was used to confirm the initial protein identification. If needed, Western blotting with protein antibody was used to reconfirm the identified protein from mass spectroscopy.
The cysteine residues of the excised protein were reduced and alkylated by 10 mM DTT and 55 mM iodoacetamide. Trypsin (0.00125%) was added to digest the protein to peptides. To extract the peptides, we added 25 mM ammonium bicarbonate, acetonitrile, and 5% formic acid. The extracted peptides were then rehydrated, concentrated, and desalted using Millipore C18 Zip Tips following the protocol of the manufacturer. MALDI-TOF MS (QSTAR XL; Applied Biosystems, Foster City, CA) was used to obtain peptide mass fingerprints with α-cyano-4-hydroxy-trans-cinnamic acid as the matrix. To obtain the initial protein identification, we submitted the peaks of peptides to MASCOT and matched to the NCBInr rattus database using the peptide tolerance of 100 ppm. Positive protein identification was based on the significant MOWSE (molecular weight search) score. Confirmation of the initial protein identification was done by sequencing peptide mass with peptide MS/MS. A peptide tolerance of 100 ppm was used in the searching (11, 49).
Western blotting for FHL1.
To reconfirm the protein identity, four-and-a-half LIM protein 1 (FHL1) monoclonal antibody (Abnova, Taipei, Taiwan; 1:1,000), goat anti-mouse alkaline phosphatase-conjugated secondary antibody (1:3,000), BCIP-NBT, and a GS-800 calibrated densitometer were used. The procedures were the same as described in Western blotting for HNE and NT.
Data are means ± SE. Two-way ANOVA was used to determine the effect of aging and HU on the total amount of HNE and NT modification of soluble and myofibrillar proteins in soleus muscles. The Tukey-Kramer multiple comparison test was used as a post hoc test when the main effect reached significance. One-way ANOVA was used to determine the effect of HU on the relative content of individual proteins and levels of modification of individual proteins in each age group. The Tukey-Kramer multiple comparison test was used as a post hoc test. An independent t-test was used to compare the baseline difference between adult and old rats. Significant difference was considered achieved when P < 0.05.
Age-Independent Effects of HU on Total Protein Oxidation
To investigate whether there are age-related differences in protein oxidative modifications with muscle unloading, we analyzed soleus muscle proteins (soluble and myofibrillar fractions) for the content of HNE and NT using a slot blot immunoassay. Figure 1 shows a summary of the densitometry of the immune reactions for HNE and NT. There was no change in the content of HNE- and NT-modified proteins with HU in either soluble or myofibrillar fractions. The results were further confirmed by measuring the density of HNE and NT immunoreaction in whole lanes (SDS-PAGE) on a Western blot (data not shown).
Soleus muscles from old animals have more HNE-modified proteins in both the soluble (P = 0.002; (Fig. 1A) and myofibrillar fractions (P = 0.017; Fig. 1C). In addition, more NT-modified proteins were found in the soluble fraction of muscles from aged animals (P = 0.003; Fig. 1B).
Protein-Specific Changes in Oxidation With HU: Identification of Proteins
Although no change was detected in the total content of HNE- and NT-modified proteins with HU, three individual immunoreactive bands on Western blots showed significant densitometric changes with HU (described below). As shown in Table 1, mass spectrometry analysis identified band 1 as carbonic anhydrase III (CAIII), and bands 2 and 3 were identified as FHL1. Figure 2 A shows the peptide mass fingerprints from band 2. The mass of eight peptides matched to FHL1. The sequence of the peptide at 1,180 m/z corresponds with the sequence in FHL1, which further confirms the protein identification (Fig. 2B). Similar results of mass spectroscopy were found for band 3. Because two closely migrating bands (bands 2 and 3) were identified as FHL1 with mass spectrometry, we confirmed the result by performing Western blotting with FHL1 antibody. Figure 2C shows that the protein bands modified by HNE and NT correspond with the immunoreaction for FHL1.
CAIII and unloading.
Figure 3 shows the representative immunoblots as well as a summary of the optical density of HNE-modified CAIII in soleus muscles from adult and old rats with different days of HU. The amount of HNE-modified CAIII in soleus muscles changed through the time course of HU in adult rats (ANOVA, P = 0.034) but not in old rats (ANOVA, P = 0.088).
FHL1 and unloading.
Figure 4 shows the representative immunoblots and densitometric analysis of HNE- and NT-modified FHL1 from adult and old rats. In adult rats, the content of HNE- and NT-modified FHL1 in soleus muscles changed during the time course of HU (Fig. 4, A and B). The Tukey-Kramer multiple comparison test indicated that the content of NT-modified FHL1 in muscles from rats with 7 days of HU was eight times higher than the content in muscles from rats with weight bearing or 14 days of HU. In old animals, no significant changes were detected with HU in the content of HNE- and NT-modified FHL1 (Fig. 4, C and D).
To clarify whether the changes in the content of HNE- and NT-modified CAIII and FHL1 with HU are associated with changes in the relative content of individual proteins, we determined the optical density of silver-stained CAIII and FHL1. No changes in the relative content of CAIII and FHL1 with HU were found in either adult or old rats (data not shown).
Age-Related Changes of Oxidative Modification in CAIII and FHL1
To further understand whether the oxidative modification of CAIII and FHL1 are age related, we compared the content of HNE- and NT-modified CAIII and FHL1 between adult and old rats with weight bearing (control). Except for HNE-modified FHL1, the content of NT-modified FHL1 and HNE- and NT- modified CAIII did not show age-related changes (data not shown). The content of HNE-modified FHL1 was five times greater in muscles from old rats than from adult rats (P = 0.04; Fig. 5).
Overview of Main Findings
This study had two major objectives: 1) to investigate whether aging influences the global accumulation of oxidized proteins in the unloaded soleus muscles and 2) to identify proteins that show disuse-related changes in the content of oxidative modification. The findings, analyzed using two-way ANOVA, reveal that the age of the rat does not play a role in the global accumulation of HNE- and NT-modified proteins with unloading. Using mass spectrometry, we identified two proteins, FHL1 and CAIII, which showed changes in the oxidation levels with muscle unloading. The changes in the oxidation levels happened to adult rats but not to old rats. The old rats had baseline levels of HNE-modified FHL1 five times higher than that of adult rats.
Description of the Experimental Model
To investigate both age- and disuse-related protein oxidation in skeletal muscles, we chose Fischer 344 rats, because this rat strain is a well-established rodent model for mammalian aging (27). Many animal models of disuse (limb immobilization, hindlimb suspension, mechanical ventilation, and cordotomy) have been developed to mimic various conditions that lead to disuse-induced muscle atrophy in humans. This study selected hindlimb suspension as the disuse model because it has been reported as the corresponding best model for mimicking conditions of bed rest and spaceflight in humans (38). Hindlimb suspension unloads the muscles of the lower limbs but permits free movement of the lower limbs and maintains the neural input of the hindlimb muscles.
Rat Age Does Not Influence the Global Accumulation of Oxidized Proteins With Unloading
We hypothesized that there would be an age-dependent global accumulation of oxidized proteins with muscle unloading because 1) oxidant generation in skeletal muscles from aged animals is greater compared with young animals (2), 2) proteasome functions decrease with aging (1, 10, 16), and 3) our previous study (6) showed that antioxidant responses to muscle unloading are age related. Specifically, the GSH levels in adult rats are maintained stable, whereas the levels are decreased dramatically in old rats with muscle unloading (6). In the current study, we found no global accumulation of oxidized proteins with muscle unloading. This finding suggests that the balance of the oxidized protein generation and removal is maintained even though the antioxidant system in old rats responds differently to unloading than adult rats.
Aging-induced muscle atrophy and weakness have been associated with oxidative stress (6, 36). The results of this study show that the overall HNE- and NT-modified proteins in soleus, type I muscles, accumulate with aging (Fig. 1). Previous research showed a greater content of protein carbonyls in type II muscles of older rats, monkeys, and humans. Increased levels of HNE- and NT-modified proteins have been reported in type II muscles of older rats and monkeys (4, 37, 55). Collectively, the results suggest that increased protein oxidation with aging occurs in both type I and type II muscles.
In the current study, we found that the global accumulation of HNE- and NT-modified proteins did not occur with muscle unloading in both adult and old rats. One interpretation of this finding is that the HNE- and NT-modified proteins are efficiently removed from the unloaded muscles. Indeed, previous studies have shown that the ubiquitin-proteasome system that degrades oxidized proteins is upregulated with muscle unloading (8, 18, 19, 47, 54). For instance, there is increased mRNA of ubiquitin (8, 18, 47), mRNA of specific ubiquitin-conjugating enzymes (E2) (18, 47), and mRNA of proteasome subunits (8, 18, 19), and an increased level of ubiquitin conjugation (54) with muscle unloading. Most likely, the upregulated ubiquitin-proteasome system efficiently degrades the HNE- and NT-specific modified proteins of the unloaded soleus muscles.
An alternative interpretation of the finding is that lipid peroxidation and subsequent generation of HNE do not increase with muscle unloading. Evidence for this is reported in studies using thiobarbituric acid, a specific marker of lipid peroxidation. The levels of thiobarbituric acid-reactive substances in soleus muscles were unchanged with 7 days of cast immobilization in rats (22), with 14 days of HU in rats (21), and with 21 days of cast immobilization in rabbits (28). These results, along with our finding of unchanged HNE-modified proteins with muscle unloading, suggest that the reported accumulation of protein oxidative-modification may not originate from lipid peroxidation.
Global accumulation of oxidized proteins determined by protein carbonylation has been reported in muscles from disuse experimental models (21, 45, 56). Protein carbonylation is a generic marker of protein oxidation and occurs through two major mechanisms, metal-catalyzed oxidation and reaction of nucleophilic amino acid side chains with lipid oxidation products such as HNE. In the former mechanism, metals such as copper and iron catalyze the formation of highly reactive, short-lived hydroxyl radicals that modify nearby amino acids, such as proline, arginine, lysine, and threonine (46). In the latter mechanism, lipid peroxidation leads to the generation of aldehyde-containing byproducts, which covalently modify nucleophilic amino acid side chains in proteins, such as cysteine, histidine, and lysine (46). Thus carbonylation reflects the overall status of oxidized proteins, whereas proteins with HNE adducts would be a subpopulation of the carbonylated proteins. Therefore, our results suggest that HNE does not contribute to the overall increase of carbonyls previously reported.
In contrast to the HU model of disuse, cast immobilization does result in an increase of HNE- and NT-modified proteins in muscles. Studies by Selsby and colleagues (43, 44) showed that HNE- and NT-modified proteins increased 22–33% and 35–50%, respectively, after 7 and 8 days of limb immobilization. This finding would seem to suggest that mechanisms of protein oxidation may be model specific.
Oxidation of FHL1 and CAIII Changes With Muscle Unloading in Adult But Not Old Rats
Although several studies have reported global accumulation of oxidized protein with muscle disuse, the identification of the proteins has not been provided. The novel finding in the current study is that FHL1 and CAIII, identified using mass spectrometry, showed changes in the levels of oxidative modification in the unloaded soleus muscles. Moreover, the changes were seen in adult rats but not in old rats.
Protein FHL1, consisting of four complete LIM domains and a NH2-terminal half LIM domain, is one member of the LIM-only protein family. The LIM domain is a cysteine-rich double zinc finger protein-binding motif and is involved in protein-protein interaction. FHL1 is highly expressed in skeletal muscles, especially in type I and type IIa muscle fibers (24, 29, 35), and is found predominantly at I band of mature skeletal muscles (33). The functions of FHL1 are not fully characterized, yet this protein appears to be a regulator of myogenesis and muscle growth (33–35, 42). For instance, the mRNA expression of FHL1 has been found associated with the muscle size of rats (29) and levels of myogenic factors in C2C12 skeletal muscle cells (35). FHL1 is thought to play a role in the assembly of the sarcomere by interacting with myosin-binding protein C (33).
The abundance of cysteine residues in FHL1 makes this protein susceptible to HNE modifications. Although old rats have greater basal levels of FHL1 oxidation, we found that HU caused an increase of oxidation in adult rats only. The influence of FHL1 oxidation on muscle function remains unclear. However, alterations of cysteine by missense mutation of FHL1 gene (and other mutations) result in protein conformation disruption, atrophy, decreased expression, and myopathy in humans (40, 53). Collectively, these studies suggest that alterations of amino acids of FHL1 affect stability of the protein and may cause muscle dysfunctions. Further studies are needed to investigate the effect of oxidation on the stability of this key skeletal muscle protein.
The other protein that shows changes of the oxidation levels with muscle unloading is CAIII. CAIII, being the most abundant soluble protein in liver (5%), slow-twitch muscles (8%), and adipocytes (25%), functions as a carbonic anhydrase, esterase, and phosphatase (3). Recent studies indicate that CAIII also functions as an antioxidant through thiol oxidation (3, 41, 58). Studies found that the content of CAIII dramatically decreases with aging (3), and cells that overexpress CAIII have a lower basal level of ROS and greater resistance to oxidative stress (41).
CAIII contains several cysteine, histidine, and lysine residues that make it a target of HNE. Although the oxidative modification of CAIII has only mild inhibition of enzyme activities (50), the covalent binding of CAIII by oxidants allows it to play a role in regulating the redox balance of a cell. Interestingly, we found the level of HNE-modified CAIII changes with muscle unloading in adult rats but not in old rats. This finding suggests that CAIII in adult rats responds to the stimuli, whereas the old rats do not, supporting the previous finding (6, 23, 25) that the ability of aged organisms to adapt to additional stressors appears to be compromised.
The global HNE- and NT-modified proteins accumulate significantly with aging but not with disuse. However, two HNE and NT target proteins, FHL1 and CAIII, show changes in the oxidation levels with muscle unloading. The changes in the oxidation levels happen to adult rats but not to old rats.
This project was partially funded by National Institute on Aging Grants AG-17768 and AG-21626 (to L. V. Thompson).
We thank Janice Shoeman, Sheng Zhong, Nicole Fugere, and David Durand for technical assistance.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2008 the American Physiological Society