Published ahead of print on June 19, 2003, doi:10.1165/rcmb.2003-0138OC
© 2003 American Thoracic Society DOI: 10.1165/rcmb.2003-0138OC Nitric Oxide Synthases and Protein Oxidation in the Quadriceps Femoris of Patients with Chronic Obstructive Pulmonary DiseaseCritical Care and Respiratory Divisions, Royal Victoria Hospital and Meakins-Christie Laboratories, McGill University, Montreal, Quebec, Canada; and Muscle Research Unit, Departments of Respiratory Medicine and Pathology, IMIM-Hospital del Mar, Universitat Pompeu Fabra & Universitat Autònoma, Barcelona, Catalonia, Spain Address correspondence to: Dr. S. Hussain, Room L3.05, 687 Pine Ave. West, Montreal, PQ, H3A 1A1 Canada. E-mail: sabah.hussain{at}muhc.mcgill.ca
Skeletal muscle dysfunction contributes to poor exercise performance in patients with chronic obstructive pulmonary disease (COPD). Increased oxygen radicals and nitric oxide (NO) have been proposed as mechanisms. In this study, we assessed the levels of protein oxidation (carbonyl formation), lipid peroxidation (4-hydroxy-2-nonenal formation), catalase and Mn-superoxide dismutase (Mn-SOD) expressions, nitric oxide synthases (NOSs), and protein tyrosine nitration in quadriceps muscles of 12 patients with patients with COPD and 6 control subjects. Lipid peroxidation was elevated in muscles of patients with patients with COPD as compared with control subjects, but protein oxidation was not. Muscle Mn-SOD but not catalase protein expression was significantly higher (200%) in patients with patients with COPDas compared with control subjects. Expression of neuronal NOS and endothelial NOS isoforms did not differ between control subjects and patients with COPD, whereas no inducible NOS protein expression was detected in limb muscles of the two groups of subjects. In patients with COPD, neuronal NOS expression correlated negatively with the degree of the airway obstruction (%FEV1 predicted). 3-Nitrotyrosine levels were significantly elevated in muscles of patients with COPDas compared with control subjects, and correlated positively with nNOS protein levels. These results indicate the development of both oxidative and nitrosative stresses in the quadriceps of patients with COPD, suggesting their involvement in muscle dysfunction.
Abbreviations: body mass index, BMI • chronic obstructive pulmonary disease, COPD • carbon monoxide trasfer, DLCO • endothelial nitric oxide synthase, eNOS • 4-Hydroxy-2-nonenal, HNE • high-performance liquid chromatography, HPLC • horseradish peroxidase, HRP • inducible nitric oxide synthase, iNOS • manganese superoxide dismutase, Mn-SOD • myosin heavy chain, MyHC • neuronal nitric oxide synthase, nNOS • nitric oxide, NO • nitric oxide synthase, NOS • optical density, OD • reactive oxygen species, ROS • maximal oxygen uptake,
Limb muscle dysfunction has been recognized in the last decade as an important factor in reduced exercise capacity in patients with chronic obstructive pulmonary disease (COPD). Muscle dysfunction is frequently associated with weight loss, which, in turn, contributes to poor quality of life as well as reduced survival (1). Muscle dysfunction is characterized by reduced muscle strength and endurance in addition to muscle wasting. Moreover, limb muscles of patients with COPD develop distinct structural alterations including attenuation of the proportion of type I fibers, reduction in myoglobin levels, and lower number of capillaries per unit surface area compared with control subjects (for review see Ref. 1). These structural changes lead to reduced oxygen delivery within skeletal muscles. In addition, there is evidence that metabolic alterations also occur in limb muscles of patients with COPD and that these alterations are manifested as reduction in the activity of oxidative enzymes, with no significant changes in glycolytic enzymes (2). The etiology of limb muscle dysfunction in patients with COPD is still under investigation; however, many factors have been implicated, including factors related to comorbid conditions such as deconditioning, electrolyte imbalance (alterations in phosphate, calcium, manganese, and chloride metabolism), heart failure, and detraining (1). Furthermore, hypoxia, hypercapnia, poor nutritional status, and a possible increase in inflammatory cytokine levels have also been proposed to contribute to limb muscle dysfunction (3). Finally, oxidative stress has also been suggested to play a role in limb muscle dysfunction in patients with COPD and to contribute to metabolic and contractile dysfunctions of these muscles (4). This proposal is based on the findings that higher levels of lipid peroxidation and oxidized glutathione are present following exercise in arterial blood of patients with COPD as compared with control subjects (5), and that total glutathione is reduced in leg muscles of patients with COPD as compared with control subjects (6). However, no measurements of the effects of reactive oxygen species (ROS) on muscle proteins or lipids have been reported thus far. Therefore, the first objective of this study was to test our hypothesis that oxidative stress develops within peripheral muscles of patients with COPD with normal nutritional status, and that this phenomenon is attributed in part to downregulation of the expression of important antioxidant enzymes such as catalase (responsible for removal of hydrogen peroxide) and Mn-superoxide disumutase (Mn-SOD)(involved in dismutation of superoxide anions). To achieve this objective, we measured protein oxidation (carbonyl formation), lipid peroxidation (4-hyrdroxynonenal protein adduct formation), and protein expression of catalase and Mn-SOD in vastus lateralis muscles of patients with COPD. Over the past several years, there has been increasing evidence that in addition to ROS, muscle redox status is strongly influenced by nitric oxide (NO). NO is produced inside skeletal muscle fibers by nitric oxide synthases (NOSs). The main source of NO synthesis in normal muscle fibers is the neuronal (nNOS) isoform, which is localized in close proximity to the sarcolemma (7). NO is also produced by the endothelial (eNOS) isoform, which is localized inside skeletal muscle mitochondria and in endothelial cells of blood vessels (8). Although the inducible (iNOS) isoform is not usually present in normal skeletal muscle fibers, this enzyme is induced in severe sepsis and endotoxin shock. There is increasing evidence that NO plays a major role in regulating muscle glucose metabolism, Ca++ release from the sarcoplasmic reticulum, blood flow, and the defense against oxidative stress (for review, see Ref. 9). Excessive NO production inside skeletal muscle fibers, however, exerts deleterious effects on contractile function and sarcolemmal integrity (9). The iNOS isoform has also been shown to be involved in the formation of peroxynitrite, which in turn targets proteins and lipids and leads to inactivation of various enzymes, including those involved in the defense against oxidative stress (10). Despite the importance of NO in the regulation of muscle function, no information is yet available regarding changes in the expression of constitutive NOS isoforms (nNOS and eNOS) in the limb muscles of patients with COPD, nor is there any data regarding the presence of the iNOS isoform and peroxynitrite inside these muscles. The second objective of this study was, therefore, to test our hypothesis that NO production is enhanced and leads to peroxynitrite formation in leg muscles of patients with COPD, and that constitutive and inducible NOS isoforms are responsible for this elevation in muscle NO production.
Patient Characteristics Twelve patients with COPD and six control individuals with normal pulmonary functions were selected. The COPD diagnosis was established based on a clinical history compatible with chronic bronchitis and/or emphysema, a long history of cigarette smoking, and pulmonary function testing revealing fixed airflow obstruction (FEV1/FVC ratio < 70% and FEV1 < 75% predicted). All subjects underwent thoracotomy for a localized lung neoplasm. Exclusion criteria included chronic respiratory failure, positive bronchodilator test, bronchial asthma, coronary disease, undernourishment (body mass index [BMI] < 20 kg/m2), chronic metabolic diseases, orthopedic diseases, suspected paraneoplastic or myopathic syndromes, previous abdominal or thoracic surgery, and/or treatment with drugs known to alter muscle structure and/ or function.
Study Design
Nutritional Assessment
Physiologic Studies Peripheral muscle function This was assessed through the classical handgrip (HG) maneuver, performed by the nondominant hand, using a Collins dynamometer (Herrera, Barcelona, Spain). At least three maneuvers were performed in each case, and the highest value was chosen. Predicted values of Mathiowetz and colleagues (13) were used.
General exercise capacity.
This was assessed using a standardized incremental exercise test performed on a cycloergometer (Monark-Crescent 864; Varberg, Sweden). The test consisted of a 3-min warm-up period at 25 watts, followed by 25-watt increments at 2-min intervals while the subject was pedaling at a constant frequency of 60 rpm. Breathing pattern, electrocardiogram, and transcutaneous oxygen saturation were continuously recorded. The exercise was stopped when at least three of the following criteria were reached: (i) a plateau of oxygen uptake (
Biological Muscle Studies
Immunoblotting.
Frozen muscle samples were homogenized in a buffer containing tris-maleate 10 mM, EGTA 3 mM, sucrose 275 mM, DTT 0.1 mM, leupeptin 2 µg/ml, PMSF 100 µg/ml, aprotinin 2 µg/ml, and pepstatin A 1 mg/100 ml (pH 7.2). Samples were then centrifuged at 1,000 x g for 10 min. The pellet was discarded and the supernatant was designated as a crude homogenate. Total muscle protein level in each sample was determined with the Bradford technique (BioRad Inc., Hercules, CA) Crude muscle homogenates (20 µg per sample) were separated by electrophoresis, transferred to polyvinylidene difluoride membranes, blocked with nonfat milk, and incubated overnight with selective monoclonal antibodies to NOS isoforms (Transduction Laboratories Inc., Lexington, KY), monoclonal anti–3-nitrotyrosine antibody (Cayman Chemical Inc., Ann Arbor, MI), polyclonal anti-Mn SOD (StressGen, Victoria, BC, Canada) and polyclonal anti-catalase (Calbiochem Corp., San Diego, CA) antibodies. Lysates obtained from rat cerebellum, endothelial cells, and cytokine-activated macrophages were used as positive controls for nNOS, eNOS, and iNOS protein expression, respectively. Specificity of anti–3-nitrotyrosine antibody was evaluated by preincubation of this antibody with either 10 mM of nitrotyrosine or 10-fold excess of peroxynitrite-tyrosine nitrated bovine serum albumin (generously provided by Dr. Ischiropoulos, University of Pennsylvania). 4-Hydroxy-2-nonenal (HNE) is an Protein oxidation. We assessed protein oxidation by measuring protein carbonyls (Oxyblot kit; Intergen Inc., Purchase, NY). Protein carbonyls are sensitive indices of oxidative injury (16). Carbonyls were detected in muscle protein samples (15 µg per sample) using the immunoblotting procedure described by Taillé and coworkers (17). To evaluate the selectivity of carbonyl measurements, muscle protein samples also underwent protein carbonyl detection procedure without the derivatization step (negative controls). Total carbonyls in each muscle sample were calculated by adding ODs of individual positive protein bands. Immunohistochemistry. Muscle samples were immersed in subsequent baths of different degrees of alcohol, formol, and xylol, to be finally embedded in paraffin. Slides were then fixed in amino propyl-triethoxilane and acetone, and dried by heat (60°C). Three-micrometer muscle paraffin-embedded sections were obtained using a microtome. All sections were deparaffinazed, and incubated with citric acid solution in a pressure cooker (antigen retrieval protocol). Slides were then blocked in 6% H2O2, incubated for 1 h at 37°C in a humid chamber with monoclonal anti-3-nitrotyrosine (1/10 dilution), monoclonal anti-MyHC-I (clone MHC, Biogenesis Inc., England, 1/25 dilution), monoclonal anti–MyHC-II (clone MY-32, 1/150 dilution; Sigma) or polyclonal anti-HNE (1/100 dilution) antibodies. After several washes in phosphate-buffered saline, slides were incubated for 1 h with biotinoylated secondary antibodies followed by HRP-conjugated streptavidin and diaminobenzidine (Dako Corporation, Carpinteria, CA) as a substrate. Negative control slides were exposed only to secondary antibodies. Slides were counterstained with hematoxylin, dehydrated, and mounted for conventional microscopy. Capillary density measurement. Paraffin-embedded sections were deparaffinized, rehydrated, washed, and incubated with a monoclonal anti-CD34 antibody (Biomeda Inc., Hayward, CA). Slides were then blocked with H2O2, probed with HRP-conjugated anti-mouse antibody, and incubated with a mixture of diaminobenzidine and chromogen solution (Dako). Capillary density was quantified as the number of capillaries per muscle fiber.
Statistical Analysis
Table 1 indicates characteristics of control subjects and patients with COPD. No differences in age, weight, BMI, nutritional status, and arterial blood gases were observed between control subjects and patients with COPD. FEV1, FVC, and FEV1/FVC were significantly lower in patients with COPD as compared with control subjects. In addition, both peripheral muscle strength and exercise capacity were mildly reduced in patients with COPD. No nutritional abnormalities were detected in patients with COPD or control subjects. No differences were observed in patients with COPD and control subjects with respect to quadriceps fiber type distributions, fiber sizes, or capillary densities.
In control subjects, protein oxidation was limited to two main 40- and 30-kD protein bands (large arrows in Figure 1). Weaker protein bands were also detected at 38 and 58 kD. The intensity of these oxidized protein bands varied considerably in muscle samples of patients with COPD, with the highest intensities detected in two patients with COPD (FEV1 of 49% and 43% predicted). Furthermore, an additional oxidized protein band with apparent mass of 9 kD was detectable in muscle samples of several patients with COPD (small arrow in Figure 1). Mean values of total carbonyl contents in patients with COPD were not significantly different from those detected in the muscles of control subjects.
In addition to protein oxidation, we evaluated whether lipid peroxidation is elevated in limb muscles of patients with COPD by measuring the level of HNE protein adducts (Figure 2A). The anti-HNE antibody detected several positive protein bands with apparent mass of 55, 52, 48, 44, 40, 35, 32, and 30 kD (Figure 2A). In most samples, the 30-kD protein band is the most intense among all protein bands. The intensities of the majority of HNE-positive bands and total HNE OD were significantly greater in muscles of patients with COPD compared with control subjects (Figure 2C, P < 0.05). Positive HNE staining was localized in close proximity to the sarcolemma in muscle fibers of patients with COPD and control subjects (Figure 2D). Few muscle fibers also showed positive cytosolic HNE staining (middle panel of Figure 2D). Removal of the primary anti-HNE antibody completely eliminated positive HNE staining (right panel of Figure 2D).
Figure 3 illustrates the changes in catalase and Mn-SOD protein expressions in vastus lateralis muscles of control subjects and patients with COPD. No differences could be detected in the expression of catalase between control subjects and patients with COPD; however, catalase ODs showed a strong tendency for positive correlation with FEV1 in patients with COPD (Figure 3C). Mn-SOD expression in patients with COPD averaged  200% of that detected in control subjects (P < 0.05).
NOS protein expressions in vastus lateralis muscle samples are shown in Figure 4. The anti-nNOS antibody detected a prominent band at 165 kD in muscles of control subjects. Likewise, the same band was also detected in the muscles of patients with COPD. Although no differences in muscle nNOS expression were detected between control subjects and COPD as a group (Figure 4B), patients with COPD with FEV1 < 50% predicted showed a strong tendency to have higher expression of nNOS compared with control subjects. When only patients with COPD were considered, a significant negative correlation between nNOS intensity and relative FEV1 values (%predicted) was detected (Figure 4C). The anti-eNOS antibody detected a positive protein band of 130 kD in vastus lateralis of control subjects (Figure 4A). Mean eNOS OD in muscles of patients with COPD was not different from that of control subjects (Figure 4B). Neither nNOS nor eNOS abundance correlated with muscle capillary density. We were unable to detect iNOS protein expression in vastus lateralis muscles of normal subjects and patients with COPD upon probing muscle lysates with a selective anti-iNOS antibody (Figure 4A).
Figure 5 illustrates representative examples of protein tyrosine nitration in crude muscle homogenates of control subjects and patients with COPD. A selective 3-nitrotyrosine antibody detected five tyrosine-nitrated protein bands with apparent molecular masses of 216, 46, 42, 36, and 30 kD. The intensities of these protein bands, particularly those of 216 and 36 kD, were stronger in the muscles of patients with severe COPD as compared with control subjects. Mean values of total muscle 3-nitrotyrosine OD for all patients with COPD rose to 200% of those of control subjects (Figure 5B). This increase was particularly evident in muscle samples obtained from patients with severe COPD. Total muscle 3-nitrotyrosine OD in all muscle samples correlated positively and significantly with nNOS OD (r = 0.60, P < 0.05) and negatively with FEV1 (% predicted, r = -0.652, P < 0.05). No such correlation was detected between the muscle tyrosine nitration and eNOS OD. Tyrosine nitration did not correlate with muscle capillary density. Positive tyrosine-nitrated protein bands were undetected when anti–3-nitrotyrosine antibody was preincubated with either pure nitrotyrosine or tyrosine-nitrated bovine serum albumin, confirming the specificity of this antibody. Clear cyotsolic and membrane-associated positive 3-nitrotyrosine staining were evident inside skeletal muscle sections of patients with COPD as well as control subjects (Figure 5C). No clear staining was detectable in negative control sections (Figure 5C).
The main findings of this study are that in the vastus lateralis muscle of patients with COPD as compared with subjects with normal lung function: (i) lipid peroxidation but not protein oxidation was significantly elevated; (ii) the expression of Mn-SOD protein was significantly higher; (iii) the nNOS isoform expression showed a strong tendency to increase in patients with severe COPD; (iv) protein tyrosine nitration averaged 200% of that of control subjects, and was detectable despite the absence of abundant iNOS protein expression.
Oxidative Stress Little is known about changes in antioxidant defenses inside skeletal muscles of patients with COPD. Only recently has it been shown that local glutathione peroxidase activity failed to increase in patients with COPD following exercise (18). Our study provides first evidence that the expression of an important mitochondrial enzyme responsible for the dismutation of superoxide anions, Mn-SOD, is significantly elevated in leg muscles of patients with COPD. This enhanced expression may represent a response to increased ROS production inside the mitochondria of these muscles. Although increased mitochondrial density can also lead to increased muscle Mn-SOD protein level in total muscle lysates, this is not a likely explanation for our finding because limb muscles of patients with COPD have been reported to have lower percentage of mitochondrial rich (type I) fibers compared with control subjects (1).
The NO Pathway The functional significance of protein tyrosine nitration in regulating skeletal muscle function remains unclear. Many studies have documented tyrosine nitration of proteins involved in energy production, apoptosis, fatty acid metabolism, oxidative stress, and structural integrity (26). In the majority of these proteins, tyrosine nitration causes loss of protein function primarily as a result of nitration of critical tyrosine residues involved in the catalytic activity of these proteins (as for example in Mn-SOD) or as a result of increased proteolytic degradation of these proteins (27). Further studies are needed to identify tyrosine-nitrated proteins inside human skeletal muscles and to explore the functional significance of protein tyrosine nitration to various muscle functions including contractile performance and the redox state. The exact mechanisms responsible for the elevation of constitutive NOS expression in leg muscles of severe patients with COPD are unknown. We propose that hypoxia, which induces significant upregulation of both eNOS and nNOS expression in skeletal muscles, might be involved (28). The presence of normal PaO2 at rest in our patients does not exclude the possibility that muscle hypoxia may develop during exercise and may contribute to increased gene expression of both nNOS and eNOS, though we have not found an increase in the latter. Another possible mechanism is proinflammatory cytokines such as tumor necrosis factors and interleukins, which have been proposed to mediate limb muscle dysfunction in patients with COPD (1). In pathologic conditions where these cytokines are elevated, such as in severe sepsis, an increase in muscle eNOS and nNOS expression has been reported (29). Disuse or lack of physical activity are not likely to have contributed to the rise in constitutive NOS expression in the leg muscle of patients with COPD because of the strong association between increased muscle activity and muscle nNOS expression (30).
Limitations of the Study Another limitation of this study is related to methodologies. Direct confirmation of ROS involvement in muscle dysfunction of patients with COPD is technically difficult, and previous studies focused mainly on measuring antioxidant levels, primarily those of glutathione (reduced and oxidized). Recently, investigators have introduced a new and reliable experimental approach to evaluate protein oxidation as a marker of ROS production within cells that is based on measuring protein carbonyl moieties by either HPLC or an immunoblotting technique (31). Similarly, thus far several methods have been employed to evaluate lipid peroxidation such as malondialdehyde and 8-isoprostane formations; however, these methods require relatively large muscle samples. Indeed, we opted to measure 4-hydroxy-2-nonenal formation in this study because this aldehyde is considered to be the most toxic product of lipid peroxidation and a major mediator of free radical cell damage and to use Western blot analysis to detect HNE protein adducts, because it requires a relatively small amount of muscle sample. HNE crosslinks with various proteins such as Na+-K+-ATPase and glucose-6-phosphate dehydrogenase, resulting in inactivation of these proteins in addition to changes in membrane fluidity and increased nonspecific permeability to ions such as Ca++. In summary, our study indicates that lipid peroxidation rather than protein oxidation is elevated in vastus lateralis muscles of patients with COPD with different degrees of airway obstruction. These muscles express enhanced protein tyrosine nitration, probably at the expense of nNOS enzyme activity, compared with muscles obtained from control subjects. These results are strongly suggestive of the development of reactive oxygen and nitrogen species-derived limb muscle dysfunction.
S.N.A.H. is a Chercheur Nationaux of the F.R.S.Q. The authors are thankful to Ms. Anna Llorens and to Mr. L. Franchi for their technical assistance. E.B. was supported by ASTRA, SOCAP (Spain), Red Respira (RTIC C03/11, FIS, Instituto de Salud Carlos III), and BIOMED BMH4-CT98–3406 (EU). This study was supported by a grant from the Canadian Institute of Health Research. Received in original form April 16, 2003 Received in final form June 9, 2003
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Top
Abstract
Introduction
O2max • maximal mechanical power output, Wmax
,ß-unsaturated aldehyde and is the most cytotoxic product of lipid peroxidation. HNE levels were evaluated by probing the membranes with polyclonal anti-HNE antibody (Calbiochem, San Diego, CA) (
0.05 was considered significant. 
