Abstract
In the diaphragms of chronic obstructive pulmonary disease (COPD) patients, the nature of oxidatively modified proteins and superoxide anion production were explored.
Diaphragm specimens were obtained through thoracotomy because of localised lung lesions in COPD patients (16 severe and eight moderate) and 10 control subjects. Lung and respiratory muscle functions were evaluated. Oxidised proteins were identified using immunoblotting and mass spectrometry. Protein and activity levels of the identified proteins were determined using immunoblotting and activity assays. Lucigenin-derived chemiluminescence signals in a luminometer were used to determine superoxide anion levels in muscle compartments (mitochondria, membrane and cytosol) using selective inhibitors.
In severe COPD patients compared with controls, respiratory muscle function was impaired; creatine kinase, carbonic anhydrase III, actin and myosin were oxidised; myosin carbonylation levels were increased five-fold; creatine kinase content and activity and myosin protein were reduced; superoxide anion levels were increased in both mitochondria and membrane compartments; and the percentage of superoxide anion inhibition achieved by rotenone was significantly greater.
In severe COPD patients, oxidation of diaphragm proteins involved in energy production and contractile performance is likely to partially contribute to the documented respiratory muscle dysfunction. Furthermore, generation of the superoxide anion was increased in the diaphragms of these patients.
- Chronic obstructive pulmonary disease
- muscle compartments
- oxidised proteins
- respiratory muscles
- superoxide anion
Chronic obstructive pulmonary disease (COPD) is a highly prevalent condition that imposes a substantial burden in terms of disability and impairment of quality of life 1. COPD is also a leading cause of death worldwide and is projected to increase within the next decade, since the prevalence of cigarette smoking continues to rise and the population ages 2, 3. Poor exercise tolerance and reduced quality of life, partially owing to muscle dysfunction, are major complaints in patients with COPD.
Several reports have shown that in severe COPD the main inspiratory muscle, the diaphragm, undergoes adaptive modifications that probably render the overloaded muscle more resistant to fatigue 4–8. However, there is growing evidence that in vitro fibre contractile function is also impaired in the diaphragms of patients with COPD 9, even at early stages of their disease, as shown by reduced myosin heavy chain (MyHC) content and increased protein degradation via the ubiquitin–proteasome pathway 10, 11. Oxidative stress, among other factors, was also suggested to be a contributor to this process of muscle dysfunction and wasting 12–14, although its specific role is yet to be fully elucidated. The oxidative post-translational modifications of proteins may cause loss of protein function as a result of oxidation of residues in the protein active site. In other cases, oxidation of critical residues may lead to increased proteolytic degradation of the modified proteins 15. Indeed, it has also been suggested that the reactive oxygen species (ROS) cascade regulates pathophysiological signalling, leading to proteolysis and apoptosis when synthesised at high levels within the myofibres 16.
In a previous study 14, it was shown that diaphragm oxidative stress could be a contributor to the intrinsic respiratory muscle dysfunction of severe COPD patients. However, in that study, neither the nature of the oxidatively modified proteins nor the generation of ROS within the diaphragm fibres of COPD patients were explored. On this basis, it is now hypothesised that excessive ROS production in the diaphragm muscles of severe COPD patients chronically exposed to inspiratory overloads would target specific proteins involved in muscle contractile function, thereby contributing to the recognised in vivo ventilatory muscle dysfunction. Accordingly, our objectives were: first, to identify the nature of the oxidatively modified proteins (protein carbonylation) in the diaphragms of patients with both severe and moderate COPD and in control subjects; and, secondly, to explore the generation of superoxide anion within several muscle compartments in the diaphragms of the same patients and control individuals.
METHODS
For additional information concerning the Methods section, please refer to the online supplementary material.
Subjects
In total, 24 Caucasian male patients with stable COPD (eight moderate and 16 severe patients) 17 and 10 healthy male age-matched sedentary controls were recruited on an out-patient basis (see online supplementary material for additional information). Moderate and severe COPD were defined in accordance with the Global Initiative for Chronic Obstructive Lung Disease guidelines 17. All subjects underwent thoracotomy for a localised lung neoplasm. All patients were clinically stable at the time of the study. Eight COPD patients (three moderate and five severe) also participated in a former study aimed at assessing the associations between oxidative stress and respiratory muscle dysfunction in COPD 14. The current investigation was designed in accordance with both the ethical standards on human experimentation of the IMIM-Hospital del Mar (Barcelona, Spain) and the World Medical Association guidelines for research on human beings. The Ethics Committee on Human Investigation at IMIM-Hospital del Mar approved all experiments. Informed written consent was obtained from all individuals.
Nutritional and functional assessment
Nutritional evaluation included body mass index (BMI) and analytical parameters. Both pulmonary and respiratory muscle functions were evaluated.
Biopsies
During thoracotomy, because of localised lung lesions, diaphragm biopsy specimens were obtained from the anterior costal diaphragm lateral to the insertion of the phrenic nerve 5, 14.
Biological muscle studies
Immunoblotting of one-dimensional electrophoresis
The levels of oxidative and nitrosative stress and those of muscle creatine kinase, carbonic anhydrase III, α-actin and MyHC were assessed, as described elsewhere 13, 14, 18.
Identification of carbonylated proteins: one-dimensional and two-dimensional electrophoresis and mass spectrometry
Oxidised proteins (protein carbonylation) were separated and identified in the diaphragms of COPD patients and control subjects, as previously described 11, 13, 18.
Creatine kinase activity assay
Total muscle creatine kinase activity was measured with a commercial kit (Diagnostic Chemicals Ltd, Charlotte Town, PE, Canada), as described elsewhere 13, 18.
Muscle fibre counts and morphometry
Proportions of type I, type II and hybrid fibres were determined, as previously described 14. The size of type I and type II fibres was also measured.
Statistical analysis
Data are presented as mean±sd. One-way ANOVA together with Tukey's test to adjust for multiple comparisons was employed in order to compare variable results among the three groups. Pearson's correlation coefficient was employed to assess relationships among different variables within COPD patients. A p-value ≤0.05 was considered significant.
RESULTS
Characteristics of the study subjects
Table 1⇓ indicates the main characteristics of the study subjects. No significant differences in age or nutritional status, as assessed by BMI and analytical parameters, were observed between control subjects and either severe or moderate COPD patients. However, forced expiratory volume in 1 s (FEV1), forced vital capacity (FVC), the FEV1/FVC ratio, diffusion capacity and arterial oxygen tension were significantly lower, while residual volume (RV) and the RV/total lung capacity ratio were significantly higher in severe COPD patients. Global respiratory muscle strength, as measured by maximal inspiratory pressure (PI,max) and maximal transdiaphragmatic pressure (Pdi,max), was moderately reduced in severe COPD patients.
Muscle structure
In the diaphragms of severe COPD patients, proportions of both type I and hybrid fibres were higher, while those of type II fibres were lower than in the muscles of either moderate COPD or control subjects (table 1⇑). When considering all COPD patients as a group, proportions of type I fibres negatively correlated with FEV1 (r = -0.499, p = 0.042).
Identification of carbonylated proteins
Total carbonyl group formation
As illustrated in figure 1a⇓, anti-dinitrophenyhydrazone antibody detected different positive protein bands, with apparent masses ranging from 67 to 29 kDa, in the muscles of both patients and controls. The diaphragms of severe COPD patients showed higher levels of total carbonyl content than those of the controls or moderate patients (fig. 1b⇓).
Carbonylated proteins
The identification of the different carbonylated proteins revealed that muscle creatine kinase, muscle carbonic anhydrase III, α-1 sarcomeric actin and MyHC were specifically oxidised in the diaphragms of both groups of patients and control subjects (table 2⇓; fig. 1c⇑ and d; fig. 2⇓). Carbonyl group formation in MyHC protein was significantly greater in the diaphragms of the severe COPD patients compared with either moderate COPD or control subjects (fig. 1c⇑ and d). Indeed, there was a five-fold increase in MyHC carbonylation in the diaphragms of the patients with severe disease compared with the controls (fig. 1c⇑ and d). Albumin (∼58–67 kDa) was also consistently oxidised in all the diaphragm specimens (fig. 2⇓).
Expression and activity of identified carbonylated proteins
Muscle creatine kinase content was significantly lower in the diaphragms of severe COPD patients compared with either moderate COPD patients or control subjects (fig. 3a and b⇓). Protein levels of carbonic anhydrase III and actin were not significantly different between COPD patients and controls (fig. 3a, c and d⇓). Moreover, creatine kinase activity was significantly reduced in the diaphragms of both severe and moderate COPD patients compared with controls (fig. 3e⇓). Importantly, MyHC content was significantly reduced in the diaphragms of the severe COPD patients compared with either moderate COPD or control subjects (fig. 4⇓). In fact, in the severe patients, diaphragm MyHC levels were 32% of those detected in the controls.
Protein tyrosine nitration
Several nitrated proteins were identified with apparent masses ranging from 64 to 30 kDa (fig. 5a⇓). Levels of diaphragm protein nitration did not differ among the study groups (fig. 5b⇓).
Diaphragm superoxide anion production
Diaphragm levels of superoxide anion were significantly higher in the mitochondrial fraction of severe COPD patients compared with those in the controls (fig. 6a⇓). Likewise, muscle superoxide anion levels were greater in the membrane fraction of severe COPD patients compared with either moderate COPD patients or controls (fig. 6b⇓). However, diaphragm superoxide anion production from the cytosolic fraction did not significantly differ between COPD patients and control subjects (fig. 6c⇓).
Interestingly, the percentage of superoxide anion inhibition elicited by rotenone (respiratory chain inhibitor) was significantly greater in the diaphragms of both severe and moderate COPD patients compared with controls (fig. 7a⇓). Conversely, the percentage of superoxide anion inhibition elicited by the inhibitors apocynin (specific nicotinamide adenine dinucleotide phosphate hydrogen (NADPH) oxidase inhibitor) and oxypurinol (inhibitor of xanthine oxidase), respectively, did not differ between COPD patients and controls (fig. 7b⇓ and c).
DISCUSSION
The novel findings in this study are that in the diaphragms of severe COPD patients compared with control subjects are the following: 1) total muscle protein carbonylation rather than protein nitration was increased; 2) muscle creatine kinase, carbonic anhydrase III, actin and myosin were oxidised; 3) carbonylation levels in MyHC were five-fold greater; 4) creatine kinase content and activity were reduced and MyHC levels were 32% of those in the controls; 5) superoxide anion levels were increased in the mitochondria and membrane compartments; and 6) the percentage of superoxide anion inhibition achieved by the mitochondrial inhibitor rotenone was significantly greater.
Respiratory muscle dysfunction and structure
In the current investigation, the severe COPD patients exhibited a moderate reduction in respiratory muscle strength, as measured by Pdi,max and PI,max. Although this level of respiratory muscle dysfunction may not be clinically relevant at rest, it could have significant clinical implications under certain circumstances, such as during an exacerbation or physical exercise. Importantly, our observations were consistent with previous reports, where in vivo respiratory muscle and diaphragm strength were also shown to be compromised 14, 21, 22. In keeping with this, it has recently been proposed 23 that in severe COPD, several molecular mechanisms contribute to the aetiology of this respiratory muscle dysfunction, such as myosin loss 10, 11, sarcomeric injury 24, oxidative stress 14 and alterations in cross-bridge cycling kinetics 8, resulting in reduced diaphragm isometric force. This raises serious questions concerning whether the adaptive mechanisms encountered in the diaphragms of severe COPD patients 4–8 are sufficient to make this muscle more efficient, since impaired contractile properties of the diaphragm muscle fibres have been clearly demonstrated 9–11.
In the present study, the diaphragm fibre type shift towards a fatigue-resistant phenotype exhibited by the severe COPD patients is in agreement with previous investigations 4, 14. Moreover, we provide evidence of the proportions of hybrid fibres in the human diaphragm. The significant increase in hybrid fibre proportions observed in the diaphragms of the severe patients is likely to be another adaptive mechanism to their respiratory overloads. Interestingly, the size of the diaphragm fibres did not differ between patients and controls. In fact, discrepancies exist in the literature regarding diaphragm fibre cross-sectional areas in COPD patients. While in several studies the size of the diaphragm fibres was reduced in these patients 4, 6, 25, no significant differences were found between patients and controls in other investigations 5, 24, 26. The number of patients examined, which was greater in the latter studies and in the current study, the methodologies employed in each study, such as the use of frozen or paraffin-embedded tissues, or the identification of the muscle fibre types using histochemical or immunohistochemical procedures, and the absence of nutritional abnormalities in our patients, which have been shown to influence diaphragm fibre sizes 25, 27, could account for such discrepancies.
Diaphragm carbonylated proteins
Over the past decade, a growing body of evidence has shown that oxidative stress is one of the mechanisms clearly involved in the skeletal muscle dysfunction of COPD patients 12–14. Carbonyl formation is an important detectable marker of protein oxidation in tissues. The present study is the first to report that carbonylation of diaphragm proteins in severe COPD and in control subjects includes proteins involved in key muscle cellular processes, such as adenosine triphosphate metabolism (creatine kinase), hydration of carbon dioxide (carbonic anhydrase III) and contractile function (actin and MyHC).
Diaphragm carbonic anhydrase III content was not different in COPD patients to that in controls. This is in keeping with a previous study 13, in which quadricep carbonic anhydrase levels did not differ between COPD patients and controls. Creatine kinase activity, however, was reduced in the diaphragms of both moderate and severe COPD patients. Indeed, this is consistent with previous reports, in which creatine kinase was shown to be a major target for the in vitro and in vivo exposure of ROS, resulting in enzyme inactivation 28, 29. In our study, although patients with moderate COPD did not exhibit a significant reduction in diaphragm creatine kinase content, it could be hypothesised that moderate levels of ROS in their muscles may have already influenced enzyme activity, since their levels were indeed similar to those of the severe patients. Moreover, it has been well established that slow-twitch oxidative fibres exhibit the lowest creatine kinase activity in skeletal muscles 30. On this basis, it is possible to conclude that the decreased creatine kinase activity in the diaphragms of our severe COPD patients could be partly due to their fibre-type shift towards a more fatigue-resistant phenotype. Eventually, although in the current investigation it is not possible to define the precise contribution of reduced creatine kinase activity to respiratory muscle dysfunction in severe COPD patients, it is likely to play a role, since absence of creatine kinase activity in mice led to profound reductions in exercise performance 31 and to myocardial dysfunction 32.
In contrast to the current findings, in a previous study 7 total creatine kinase activity levels in the diaphragms of severe COPD patients were shown not to differ from those in the control muscles. Discrepancies between the two studies could be explained by the different methodologies employed to determine creatine kinase activity and by the fact that measurements were only conducted in saponin-skinned fibres in the previous study 7. In line with the current findings, the quadriceps of severe COPD patients also exhibited a significant reduction in creatine kinase content and activity both at rest 13 and after exercise training 33. However, creatine kinase activity in the deltoid 34 and external intercostal muscles 35 did not differ between severe to moderate COPD patients and control subjects.
Several studies have already demonstrated oxidation of the cytoskeletal protein actin in different models 36, 37. The complex modifications induced by ROS on muscle actin are characterised by severe disruption of the actin filaments, hampering their interaction with the myosin protein 38, thus suggesting that oxidation of contractile actin may contribute to muscle dysfunction in these models 36, 38. In our study, although muscle actin levels did not differ between COPD patients and controls, disruption of the contractile actin in the diaphragms of the severe patients cannot be ruled out.
Importantly, in line with the observations of Ottenheijm and co-workers 10, 11, the diaphragm of our severe COPD patients exhibited significantly lower levels (32%) of MyHC protein than the controls. The novel finding of the current study is that in the diaphragms of the severe patients, carbonylation levels of MyHC protein were five times greater than in the control muscles. Although increased MyHC carbonylation and reduced MyHC content in the diaphragms of our severe patients cannot be causally related, it is likely that the greater oxidative damage exhibited by MyHC protein in these muscles may have contributed to its higher protein breakdown 10, 11. In fact, activity of the ubiquitin–proteasome pathway was shown to be increased in patients with mild to moderate COPD 11 and inflammatory cytokines and ROS are well-known triggers of muscle proteolysis 16. Taken together, these findings suggest that, in the diaphragms of severe COPD patients, highly carbonylated MyHC protein is prone to further degradation by the ubiquitin–proteasome pathway. Clearly, a future study should be designed in order to explore the exact contribution of oxidants to increased muscle protein breakdown and myosin loss in the diaphragms of severe COPD patients.
Diaphragm superoxide anion formation
The current investigation is the first to report in vitro measurements of the superoxide anion being produced within distinct muscle fractions in the human diaphragm. Interestingly, superoxide anions are the initial substrates of a series of reactions resulting in the formation of more powerful ROS in tissues. In the present study, in vitro generation of the superoxide anion by the mitochondrial and membrane compartments was significantly increased in the diaphragms of severe COPD patients compared with controls. Another important finding in this study is that rotenone, a selective mitochondrial respiratory chain inhibitor, elicited significantly greater inhibitory effects of ROS generation in the diaphragms of the patients. Importantly, in exercising muscles, oxygen utilisation increases several fold, eventually resulting in leakage of superoxide anions from the mitochondria to other muscle structures. In patients with severe COPD, the main inspiratory muscle, the diaphragm, must incessantly generate muscle contractions of relatively high intensity in order to overcome the increased inspiratory loads imposed by the mechanical alterations of their thoracic cage. This may result in the continuous generation of abnormally high levels of superoxide anion within the diaphragm mitochondria, which would not possibly revert to “control” levels in these patients.
In view of the significant increase in superoxide anion generation in the membrane compartment of diaphragms in the severe COPD patients, it may be suggested that muscle NADPH oxidase enzyme complex was a likely contributor to superoxide anion generation in these muscles, as has previously been shown in septic muscles 20. In contrast to the mitochondrial and membrane compartments, the amount of superoxide anion produced in the cytosolic fraction was relatively low in the diaphragms of both patients and controls. Furthermore, oxypurinol had no major effects on ROS generation within the diaphragm cytosolic compartment in any of the study groups, suggesting that xanthine oxidase is probably not involved in the generation of superoxide anion in the human diaphragm.
Study limitations
Although lung volume reduction surgery also makes it possible to obtain diaphragm specimens, only very severe COPD patients undergo this type of surgery. Therefore, diagnostic–therapeutic thoracotomy is the only approach available for studying mild to moderate COPD and normal lung function subjects. Accordingly, the population of our study shares a common morbidity; the presence of a localised lung neoplasm. Nevertheless, we do not believe that this condition has made any significant contribution to the development of oxidative stress in these diaphragms, since extremely restrictive criteria were employed to select our population. Another limitation encountered in our study is directly related to the methodologies involved in the in vitro measurements of muscle superoxide anion, which have been challenged on the basis of the potential production of superoxide anion by the redox recycling of lucigenin. However, as previously demonstrated 20, we do not believe that lucigenin redox recycling has made any substantial contribution to the superoxide anion concentrations detected in our study.
In the present study, both patients and controls were recruited over a period of several years. In order to avoid unnecessary methodological concerns due to analyses conducted sequentially rather than simultaneously, all muscle specimens were always stored frozen at -80°C until further use in the laboratory. Although we acknowledge that freezing and fractionation of the muscle samples might have somehow influenced the superoxide anion measurements, we do not believe that this has had any substantial effects in our study for the following reasons: 1) All superoxide anion analyses were conducted simultaneously for both patients and controls under identical laboratory conditions by the same investigator; 2) a fragment of the muscle specimens was always kept frozen in our -80°C freezer until muscle fractions were obtained and superoxide anion experiments were conducted; and 3) fractionation of the muscle samples using specific buffers that do not destroy the different cell compartments was already demonstrated to yield reliable lucigenin-derived chemiluminescence signals 20. Further information can be found in the online supplementary material.
Conclusions
Our study is the first to provide evidence that in severe COPD patients, oxidation of diaphragm proteins involved in energy production and contractile performance is likely to partially contribute to the documented respiratory muscle dysfunction. Furthermore, generation of the superoxide anion was increased in the diaphragms of these patients.
Support statement
This study was supported by FIS, the Spanish Health Ministry (grants 04/00161 and 06/1043), SEPAR 2004, the Catalan Government (2005-SGR01060) and CIBERES (all Spain). E. Barreiro received the ERS COPD Research Award 2008.
Statement of interest
None declared.
Acknowledgments
The authors are grateful to M.L. Blanco (IMIM-Hospital del Mar), M.J. Masdeu (Hospital Parc Taulí, Sabadell), J.A. Barberà (Hospital Clinic), and J. Martínez-Llorens (IMIM-Hospital del Mar) for their help with the patient recruitment, C. Chiva and M. Rodríguez from the Protemics Laboratory (Universtat Pompeu Fabra) for their help with the protein identification, to F. Sánchez and S. Mas (both IMIM-Hospital del Mar; all Barcelona, Spain) for their technical assistance in the laboratory, and to R. Marshall for help with editing.
Footnotes
This article has supplementary material accessible from www.erj.ersjournals.com
- Received May 10, 2008.
- Accepted January 17, 2009.
- © ERS Journals Ltd