INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Several lines of evidence suggest that mitochondrial metabolism is impaired in skeletal muscles of patients with chronic obstructive pulmonary disease (COPD). First, studies using ³¹P-magnetic resonance spectroscopy have shown abnormal phosphocreatine turnover and intracellular pH values during exercise and recovery in these patients (1, 2). Interestingly, these functional abnormalities are ameliorated by oxygen therapy (3). Second, a decrease in the activity of some oxidative enzymes (e.g., citrate synthase) has been reported in biopsies taken from the quadriceps femoris of COPD patients (4, 5). Collectively, these results have been interpreted as evidence of abnormal oxidative metabolism. To date, however, no study has investigated cytochrome-c oxidase (COX) in these patients. COX is a key oxidative enzyme because, as the terminal complex of the mitochondrial electron transport chain, it catalyzes the oxidation of reduced cytochrome c by oxygen and modulates oxygen uptake (6, 7). Further, recent in vitro studies indicate that the activity of COX can be directly regulated by the presence of molecular oxygen (8). Thus, a better understanding of the role of COX in patients with COPD and chronic hypoxemia can provide an important link between the availability of oxygen to tissues and the regulation of oxygen uptake and energy production in these patients.

The human mitochondrial DNA (mtDNA) is a 16.569 base pair naked circular double-stranded extrachromosomal molecule that is located in the mitochondrial matrix (6, 7). It encodes a limited number of genes and, in particular, 3 of the 13 subunits of COX (6, 7). Recent studies show that tissue hypoxia modulates nuclear gene expression very significantly (9, 10). Yet, no previous study has analyzed the effects of chronic respiratory failure upon mtDNA expression, particularly in relation to COX, in patients with COPD.

This study compares the activity of COX and the expression of some mtDNA-related genes in patients with COPD and in healthy sedentary volunteers of similar age. It also assesses the relationship between them and the degree of arterial hypoxemia present.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Population

We studied 18 male patients with COPD under clinically stable conditions (none of them had suffered from an acute exacerbation during, at least, the past 4 mo), and six healthy sedentary volunteers of similar age (Table 1). Patients were divided in two groups according to the presence (n = 10) or absence (n = 8) of chronic respiratory failure (CRF), defined as an arterial PO₂ value lower than 60 mm Hg. To avoid potentially confounding effects, we studied only male subjects, and we excluded patients with known neuromuscular disorders, cardiac failure, diabetes mellitus, alcoholism, and/or those who were receiving treatment with oral steroids. No patient was participating in a rehabilitation program. All of them signed the informed consent, after being fully aware of the nature, characteristics, and risks of the study. This investigation was approved by the local ethics committee of our hospital.

Lung Function

Forced spirometry (G. S. Collins, Braintree, MA) and arterial blood gases while breathing room air (Instrumentation Laboratory, IZASA, Barcelona, Spain) were measured in all participants (11). Spirometric reference values were from a mediterranean population (12). In patients with COPD, total lung capacity (TLC) and residual volume (RV), by body plethismography (E. Jaegger, Würtburg, Germany), and single-breath carbon monoxide diffusing capacity (DL_CO) (G. S. Collins) were also measured. Reference values were those of Roca and coworkers (13, 14).

Muscle Biopsy

We used standard percutaneous needle biopsy techniques (15) to obtain several muscle biopsies (~ 8) in each subject from the lateral portion of muscle quadriceps femoris (at the midthigh level), under local anesthesia. COX activity was investigated immediately after biopsy in fresh samples. Other samples were frozen in liquid nitrogen and stored at 80° C until analysis of mtDNA gene expression.

COX Activity Determination

Fresh muscle samples were weighed (wet weight: 30 to 90 mg) and homogenized in 250 mM sucrose/1 mM HEPES/0.2 mM EDTA buffer (pH 7.0) in a Teflon/glass homogenizer, held in an ice bath. Aliquots of this homogenate were assayed for total protein content (16) and COX activity (E.C.1.9.3.1). Analysis of COX activity was performed at 37° C according to the protocol described by Wharton and Tzagoloff (17) in a spectrophotometer with continuous optical absorbency registry (Shimadzu, Nagoya, Japan). Specific COX activity was normalized for total protein (µkats/mg protein). All reagents were obtained from Sigma Chemical (St. Louis, MO).

Mitochondrial Gene Expression

To assess mtDNA expression in these patients, we measured the mRNA content for subunit I of COX (mRNA COX-I) and the RNA component of the 12S ribosomal subunit (12S rRNA). Both measurements were normalized to the mRNA of actin, which is generally accepted in muscle as a housekeeping gene, and which is probably not influenced by hypoxia (18). RNA was extracted from muscle samples (~ 20 mg) following a modification of the method of Birnboim (19, 20). RNA transcripts were measured using reverse transcription (RT) and polymerase chain reaction (PCR). This was performed in a total volume of 10 µl containing 1.5 µl of RNA solution, extracted from approximately 1.5 mg of muscle tissue, 1 mM nucleotides (deoxyribonucleoside triphosphates, dNTPs [Pharmacia, Freiburg, Germany]) 0.2 µg/ 10 µl reverse primers for COX-I [TAAGGGAGGGTAGACTGTTC], 12S [GAACAGGCTCCTCTAGAGGG], skeletal actin [GGGCGAGATCTTGATCTTC] and 2 µl of 5× reaction buffer (Promega, Madison, WI). The reaction mixture was denatured (65° C, 3 min) and cooled on ice before 20 U RNasin (0.5 µl of RNasin 40 U/µl) (AGS, Heidelberg, Germany) and 20 U of avian myeloblastosis virus (AMV) reverse transcriptase (2 µl of 10 U/µl) (AGS) were added for 1 h at 42° C. PCR (21) was performed with 2.5 µl of RT reaction mixture in a total volume of 50 µl, containing 0.05 µg/50 µl forward primers (COX-I [TCGCCGACCGTTGACTATTC]; 12S [CAGCCACCGCGGTCACACGA]; actin [CGCGACATCAAAGAGAAGCT]), 2 U Taq polymerase (AGS, Heidelberg, Germany) and reaction buffer to which radioactive phosphorus-deoxycytidine triphosphate (³²P-dCTP) (20 mCi) (DuPont de Nemours, Bad Homburg, Germany) was added. Denaturation temperature was 95° C (30 s), annealing was at 50° C (60 s), and synthesis at 72° C (30 s). Aliquots of 1 µl were taken from the reaction mixture before the first (as a measure of background radioactivity) and after consecutive cycles (cycle 10-20) in order to determine the exponential phase of the PCR and to find the optimal point for sampling. To this end, a pool RNA solution was used which was obtained from the samples to be analyzed later. The amplification products were separated according to their length (COX-I, 391 bp; 12S, 318 bp; actine, 367 bp) in 15% polyacrylamide gels, stained with ethidium bromide. The bands were then cut out from the gel and the incorporated radioactivity was determined by liquid scintillation counting (Canberra Packard, Dreieich, Germany). Then, PCR was repeated four times for each individual sample, which was amplified to the prechosen cycle (typically cycle 18) and the incorporated radioactivity was determined. Ratios were calculated for mRNA COX-I/mRNA actin and 12S rRNA/mRNA actin for each reaction and from these, the mean value for each sample was calculated.

Statistical Analysis

Results are shown as mean ± SEM. A nonparametric analysis of variance (Kruskal-Wallis), followed by appropriate post hoc contrasts (Mann-Whitney U test) was used to assess the statistical significance of differences between groups (control subjects, COPD without CRF, and COPD with CRF). Potential relationships between variables of interest were evaluated using the Spearman correlation test. A p value lower than 0.05 was considered significant.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

All patients were ex-smokers. Pharmacological treatment included inhaled short-acting ₂ agonists (n = 16), inhaled ipratropium bromide (n = 8), oral theophylline (n = 12), and inhaled steroids (n = 6). All patients had severe airflow limitation. The degree of airflow obstruction, however, was similar in patients with and without CRF (Table 1). By definition, patients with CRF had abnormal arterial blood gas values (Table 1).

Total protein content was similar in patients and control subjects (Table 1). The activity of COX was higher in patients than in control subjects, particularly in those with CRF (Table 1). Figure 1 shows the relationship of arterial PO₂ and COX activity for all subjects studied. There was a strong significant inverse relationship (Rho = 0.59, p = 0.003) between these two variables (Figure 1).

View larger version (10K): [in this window] [in a new window]   Figure 1.   Relationship between arterial PO2 and the activity of COX (left) and the RNA component of the 12S ribosomal subunit (12S rRNA), normalized for the mRNA of actin (right), in all subjects studied. The Pearson regression line is drawn for clarification.

The mean value of COX-I mRNA was not different between patients and control subjects (Table 1). In contrast, 12S rRNA was significantly increased in patients with CRF as compared both with controls and patients without CRF (Table 1). We did not find any significant relationship between Pa_O2 and mRNA of COX-I, but the former was significantly related to 12s rRNA (Rho = 0.49, p = 0.03) (Figure 1).

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The main findings of our investigation were that the activity of COX and the expression of some mt-DNA-related genes (12S rRNA) were increased in skeletal muscle of patients with COPD and CRF (Table 1), and that both were inversely related to the degree of arterial hypoxemia present (Figure 1).

To our knowledge, no previous study has determined the activity of COX in patients with COPD. Several previous studies, though, have shown increased COX activity in skeletal muscle under conditions of tissue hypoxia, such as in patients with peripheral arterial insufficiency (22), and in healthy individuals at moderate altitude (23). In keeping with these observations, we found that the activity of COX was also increased in COPD patients, particularly in those with chronic respiratory failure (Table 1, Figure 1). This observation can help to explain why patients with COPD often show increased basal metabolic rate of uncertain origin (24) and, also, some recent experimental findings by Marrades and colleagues showing that leg oxygen consumption during exercise is higher in patients than in control subjects at the same workload (25); an increased COX activity would help to sustain metabolic flux rates by increasing the blood-to-mitochondrion PO₂ gradient and facilitating oxygen diffusion and tissue extraction (26, 27). Yet, our observation of a higher activity of COX in patients with COPD may seem at variance with two recent studies reporting decreased oxidative activity in skeletal muscle of these same patients (4, 5). However, these two studies did not measure the activity of COX. Rather they quantified the activity of enzymes involved in the citric acid cycle (citrate synthase, succinic acid dehydrogenase) or the -oxidation of fatty acids (3-hydroxyacyl coenzyme A dehydrogenase) (4, 5), which do not provide a quantitative measure of the capacity of oxidative metabolism (27, 28). Further, it is theoretically possible that different enzymes may be regulated differently (5). Therefore, our results do not support a generalized (i.e., homogeneous) decrease of mitochondrial oxidative metabolism in patients with COPD, as previously suggested (4, 5). On the contrary, if all available information (4, 5) is taken into account, our results suggest that the citric acid cycle and the electron transport chain may be somehow uncoupled in patients with COPD (29). We can not address directly the mechanisms underlying this interpretation, but potential mediators may include, among others, tissue hypoxia (9), reactive oxygen species (30), nitric oxide (31), and several cytokines (e.g., TNF-) (32). This hypothesis would require confirmation in future studies.

Under the experimental conditions used in our study to measure the activity of COX (cofactor and substrate excess at 37° C), changes in measured activity are paralleled by changes in enzyme content (33). Therefore, it was of interest to investigate whether or not these patients showed evidence of increase COX biogenesis. In mammals, COX consists of 13 subunits, three of which are coded by mtDNA (subunits I-III) and the remainder by nuclear DNA (subunits IV-XIII) (6, 7). Expression of the mitochondrial encoded subunits is a necessary, if not a rate-limiting step, in the biogenesis of the enzyme (20). We found that COX-I mRNA was not different between patients and control subjects but, in contrast, 12S rRNA was higher in patients with CRF (Table 1). Further, the latter (but not the former) was significantly related (p = 0.03) to the degree of arterial hypoxemia (Figure 1). The observation of a higher content of 12S rRNA in these patients (and its relationship with Pa_O2, Figure 1) suggests that chronic hypoxia may increase the number of mitochondrial ribosomes (polyribosome) (34). This mechanism would enhance translation efficiency for any given single RNA transcript of COX-I (35), and would contribute to explain, at least in part, the increased COX activity seen in our patients (Table 1). Admittedly, though, it has to be recognized that the regulation of COX biogenesis is complex (36) and possibly under multiple control sites (37). In fact, various mitochondrial components have variable turnover times, variable degree of nuclear and mitochondrial control, and variable specific genomic expression (33). Therefore, individual proteins can also be individually regulated (33).

In summary, our study extends previous investigations on skeletal muscle bioenergetics in patients with COPD (1, 2, 4, 5) by integrating new biochemical and genetic aspects. We found a strong inverse relationship between the degree of arterial hypoxemia and both the activity of COX and the expression of some mtDNA genes. As a working hypothesis, we propose that the electron transport chain and the citric acid cycle may be uncoupled in patients with COPD. Because of its potential clinical impact, future studies will be needed to confirm this observation and to unravel the underlying mechanisms. Finally, our results suggest that the observed increase in COX activity in these patients is regulated at the translational level, by increasing the number of mitochondrial ribosomes.