INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Lung transplantation has become an established mode of treatment for many forms of end-stage pulmonary disease. The presently utilized techniques include heart lung transplantation (HLT), single lung transplantation (SLT), bilateral lung transplantation (BLT), and more recently, live donor bilateral lobar transplantation. When fully recovered postoperatively, HLT and BLT recipients typically have near-normal spirometry with mildly impaired diffusion capacity. SLT spirometry typically remains abnormal, reflecting the pathology of the remaining native lung, with reduced diffusion capacity (1).

The reported peak oxygen consumption (O₂peak) after lung transplantation ranges from 41% to 57% of predicted levels with remarkable similarity between transplant type and no major influence of type of pretransplant pulmonary disease (2). Cardiopulmonary exercise testing in these patients is characterized by low peak work rate, early lactate threshold, and reduced O₂peak (1). A number of potential causes of exercise limitation have been postulated including: poor motivation; reduced cardiac output in SLT (particularly in patients with previous pulmonary hypertension) and HLT; ventilatory (especially flow) limitation in some SLT recipients; and low hemoglobin levels [Hb] (1). A peripheral skeletal muscle defect resulting in reduced uptake and/or utilization of delivered oxygen, however, has been postulated as an important factor (1, 2).

Peripheral muscle defects may reflect persistence of a pretransplant skeletal muscle injury, perhaps due to the deconditioning/disuse. Skeletal muscle fiber-type changes with marked reduction in type I fiber proportion have been reported in severe chronic obstructive pulmonary disease (COPD) (3). Abnormalities in oxidative metabolism of skeletal muscle have been reported in congestive heart failure (CHF) (4), COPD (5), and cystic fibrosis (6). Posttransplant factors resulting in a possible peripheral muscle defect may include protein catabolism (particularly in response to sepsis), immobilization, and the effects of medication, especially corticosteroids and cyclosporin A (CyA). Recent studies have implicated CyA in the impairment of mitochondrial oxidation (both in vivo and in vitro) in the skeletal muscle of rats (7, 8). More recently, Evans and coworkers (9) examined exercise metabolism in human lung transplant recipients using ³¹P-nuclear magnetic resonance spectroscopy and demonstrated a low pH in quadriceps femoris muscle at rest and an earlier decline in muscle pH during exercise. These patients also had a lower lactate threshold during increment exercise, and Evans and coworkers speculated that reduced oxidative capacity of peripheral skeletal muscle may be the cause of exercise limitation. Furthermore, the investigators speculated that this may reflect the persistence of a pretransplant abnormality of skeletal muscle oxidative capacity.

In the current study we hypothesized that, in spite of a number of potential mechanisms for exercise limitation in the lung transplant (LTx) recipient, reduced oxidative capacity of peripheral skeletal muscle is an important and common cause. We investigated oxidative metabolism in skeletal muscle of patients after lung transplantation and measured mitochondrial oxidative phosphorylation, metabolic enzyme activity (oxidative and glycolytic), fiber type proportion, and resting muscle metabolites.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Subjects

Seven stable LTx recipients (4 women and 3 men, 3 to 24 mo after lung transplantation) and seven age- and sex-matched control subjects volunteered to participate in this study (Table 1). The transplant recipients had stable lung function, with no evidence of recent infection or rejection, having fully rehabilitated after the LTx procedure. The respective diagnoses, operative procedures, and immunosuppressive medication are shown in Table 2. The controls were sedentary and did not exercise regularly. The Victoria University of Technology Human Research Ethics Committee and The Alfred Group of Hospitals Ethics Committee approved all protocols and procedures. Written informed consent was obtained from all participants.

Incremental Exercise Tests

At least 2 h after a light breakfast, each subject performed an incremental exercise test on an electronically braked cycle ergometer to determine their O₂peak. Exercise tests for LTx recipients and control subjects were conducted at different sites for subject convenience. The LTx recipients were tested on a Sensormedics 2900 system (Sensormedics Co., Yorba Linda, CA). The control subjects were tested on a system which consisted of a Lode N.V. cycle ergometer (Groningen, The Netherlands) with the expired gases directed through a Hans-Rudolph 2-way nonrebreathing valve (Hans-Rudolph Co., Kansas City, MO) into a 6-L mixing chamber. Expired volume was measured using a flow transducer (KL Engineering, Sunnydale, CA), mixed expired oxygen and carbon dioxide were analyzed by rapid responding gas analyzers (model S-3A; Amtek, Pittsburgh, PA). Ventilatory gas data were calculated over 15 s (TurboFit, Ventura, CA). After resting on the cycle for 5 min, subjects performed an exercise test, in which the work rate was increased by 25 watts and 16.3 watts each minute in control subjects and patients, respectively, to volitional exhaustion. Before commencement of the exercise test a 20-gauge indwelling catheter was inserted into a vein in the dorsum of the hand within 4 cm of the wrist. Arterialization was maintained by immersion of the hand, covered by a waterproof plastic glove, in a water bath maintained at 45° C. Arterialization was confirmed by blood gas analysis with oxygen saturation persistently > 90%. Blood samples were drawn each minute during exercise as well as at 1, 2, 5, 10, 20, and 30 min after exercise (results in recovery not shown) for measurement of arterialized plasma lactate concentration.

Vastus Lateralis Muscle Biopsy

Between 2 and 7 d after the incremental exercise test a resting vastus lateralis muscle biopsy was obtained. In the transplant group this procedure was performed while the patients were sedated using pethidine and midazolam for a routine bronchoscopy. With the subject resting supine, 2% lidocaine local anesthetic was infiltrated into the skin, subcutaneous tissue, and fascia overlying the belly of the vastus lateralis. A small incision was then made into the skin and a 5-mm-diameter muscle biopsy needle was used to obtain approximately 200 mg of skeletal muscle. Approximately 70 to 100 mg of the muscle sample was used for this study and the remainder was used for other studies on sarcoplasmic reticulum function and Na⁺/K⁺ adenosine triphosphatase (ATPase) activity which will be reported separately. One portion of muscle (15 to 20 mg), for fiber type analysis, was embedded in mounting Tissue Tek medium (Sokurn Finetec, Torrance, CA), immediately immersed in isopentane cooled in liquid nitrogen, and subsequently stored in liquid nitrogen. Two pieces of muscle (each 15 to 20 mg) were immediately frozen in liquid nitrogen and stored in liquid nitrogen for later analyses of enzyme activities and metabolites. The remaining piece of fresh muscle (25 to 40 mg) was placed on ice and used for the determination of mitochondrial adenosine triphosphate (ATP) production rate (MAPR).

Muscle Fiber Types

Muscle fiber types were determined using the myofibrillar ATPase method as described by Dubowitz and Brookes (10). The muscle cross-sections (10 µm) were preincubated at pH 10.35 and 4.30. After staining for myosin ATPase, a minimum of 200 fibers were visually counted to determine the percentage of each fiber type.

Muscle and Blood Metabolite Analyses

The muscle adenine nucleotide ATP, adenosine diphosphate (ADP), adenosine monophosphate (AMP), and their degradation product inosine monophosphate (IMP) were analyzed by reverse-phase high-performance liquid chromatography (HPLC) (ICI Instruments, Sydney, Australia). Phosphocreatine (PCr), creatine (Cr), and lactate were determined enzymatically using fluorometric detection (11). Lactate concentration was determined on deproteinized plasma. The lactate threshold was assessed using a log-log transformation plot of plasma lactate concentration against O₂ (12).

Enzymatic Assays

Metabolic enzyme activities measured were: the Krebs cycle enzymes citrate synthase (CS) and 2-oxogluterate dehydrogenase (OGDH); the fatty acid -oxidative enzyme 3-hydroxyacyl-CoA-dehydrogenase (HAD); the glycogenolytic enzyme phosphorylase (PHOSPH); and the glycolytic enzymes phosphofructokinase (PFK), pyruvate kinase (PK), and hexokinase (HK). Assays for PHOSPH, PFK, PK, HK, and HAD were performed on tissue homogenates using fluorometric methods (13). CS activity was determined using the spectrophotometric method (14) and OGDH according the method described by Read and coworkers (15). Enzyme activities are expressed in µmol · min¹ · g¹ wet weight of muscle.

Mitochondrial Preparation

Fresh muscle was placed on a plate over ice, and dissected free of fat and connective tissue, cut into small pieces, and weighed. The muscle sample was twice homogenized for 5 s at low speed using an electric homogenizer (S/N TH-1276; OMNI International, Warrenton, VA). The homogenizing solutions used for the preparation of isolated mitochondria and the isolation procedures were as previously described (16).

MAPR Measurement

MAPR was determined at 25° C, using the method and substrate concentrations described by Wibom and Hultman (16). Aliquots of mitochondrial suspension were added to cuvettes containing purified ADP, ATP monitoring reagent (AMR) (FL-MXB; Sigma, St. Louis, MO), and a variety of substrates to represent fat, carbohydrate, and protein metabolism. The substrate combinations used in the present study were as follows: pyruvate + malate (P + M), palmitoyl-L-carnitine + malate (PC + M), succinate + rotenone (S + R), -ketoglutarate (-KG), and pyruvate + malate + plamitoyl-L-carnitine + -ketoglutarate (PPKM). A blank containing ADP and mitochondrial suspension was used for estimation of ATP production by the adenylate kinase reaction and other nonspecific reactions. MAPR was determined from the rate of light production which was measured on a custom- designed luminometer comprising a photomultiplier tube attached to a luminescence spectrometer (model AB2; Aminco Bowman, Urbana, IL). At the end of the MAPR assay, 190 nmol ATP standard (FL-AAS; Sigma) in 10 µl was added to the test cuvette as an internal ATP standard, to allow for calculation of MAPR. Rates are expressed as mmol · min¹ · g¹ mitochondrial protein and mmol · min¹ · kg¹ muscle, referring to the protein content in the mitochondrial suspension and the muscle wet weight, respectively. All measurements were made in duplicate and were completed within 4 h of biopsy.

Measurement of Glutamate Dehydrogenase Activity

Glutamate dehydrogenase (GDH), a glutamate oxidative deamination enzyme, is a specific mitochondrial marker enzyme and was used to determine mitochondrial yield in the suspensions. Mitochondrial GDH activities were determined on the mitochondrial suspension before and after disruption of the mitochondria. From these measurements the activity of GDH in intact mitochondria (GDHim) could be determined. Total muscle GDH activity (GDHt) was determined on a separate piece of muscle (16). The ratio of GDHim to GDHt gave the yield of intact mitochondria in the suspensions. The GDH activity was determined at 35° C (17) with fluorometric detection (model AB2; Aminco Bowman, Urbana, IL). The protein content in the mitochondrial suspension was measured according to the method of Lowry (18).

Statistical Analysis

Data from patients and control subjects were compared by an unpaired Student's t test for independent variables. Data are expressed as means ± SEM.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Exercise Capacity and Pulmonary Function

Lung function tests revealed mild mixed obstructive and restrictive ventilation defects in the LTx recipients as a group with normal spirometry in the control subjects (Table 1). In LTx recipients, despite the smaller work rate increment, the duration of exercise was significantly shorter (p < 0.05) when compared with control subjects (Table 3). The LTx patient group had an earlier lactate threshold (Figure 1), occurring at 8.3 ± 0.7 ml · kg¹ · min¹ (O₂/predO₂max = 23%) versus 16.7 ± 1.2 ml · kg¹ · min¹ (O₂/predO₂max = 50%) for control subjects (p < 0.0005), and at termination of exercise, exhibited a lower peak work rate, O₂ and heart rate (p < 0.05, Table 3). Ventilatory limitation appeared to occur during exercise in one LTx patient (a measured maximal minute ventilation/maximal voluntary ventilation [E_max/MVV] of 110% of the predicted value; MVV = 35 × FEV_1.0), but in the remainder E_max/MVV was < 82% of predicted value. No patient desaturated at termination of exercise. In six of seven patients leg fatigue was the predominate symptom at termination of exercise; the remaining patient reported equally severe leg fatigue and shortness of breath when exercise ceased.

View larger version (17K): [in this window] [in a new window]   Figure 1.   Plasma lactate concentration, expressed in mmol · L1, versus work rate during incremental exercise testing. LTx recipients (solid circles; n = 7) and matched control subjects (solid squares; n = 7).

Muscle Fiber Types

The average number of muscle fibers counted was 243 ± 29 versus 276 ± 34 in LTx recipients and control subjects, respectively. LTx recipients exhibited a lower proportion of type I muscle fibers (24.9 ± 4.4% versus 56.1 ± 2.4%, p < 0.001) and therefore a higher proportion of type II muscle fibers (75.1 ± 4.4% versus 43.3 ± 2.6%) than the control subjects (Figure 2).

View larger version (12K): [in this window] [in a new window]   Figure 2.   Skeletal muscle fiber-type proportions in LTx recipients (n = 7) versus matched control subjects (n = 7). Values are expressed as mean ± SEM. *p < 0.01.

Resting Metabolites

In resting skeletal muscle, lactate and IMP concentrations were higher (both p < 0.01), and ATP lower (p < 0.05), in LTx recipients compared with the control subjects. The concentrations of muscle creatine, PCr, ADP, AMP, and total creatine (creatine + PCr) did not differ between the two groups. The total adenine nucleotides (TAN = ATP + ADP + AMP) and the ratio of ATP/ADP were lower in LTx recipients than control subjects (p < 0.05; Table 4).

Enzyme Activities

LTx recipients demonstrated lower activities of the mitochondrial enzymes CS, HAD, GDH, and OGDH (p < 0.005) and higher activity in the glycolytic enzyme, PFK, compared with the control subjects (p < 0.05). There were no significant differences in the activities of PHOSPH, HK, and PK between the two groups (Table 5).

MAPR

GDHt and the mitochondrial protein yield were significantly lower in LTx recipients compared with control subjects (p < 0.05). The protein concentration in the mitochondrial suspension, mitochondrial protein in the assay's medium, and the yield of mitochondria did not differ statistically between the two groups (Table 6). MAPR, in the presence of P + M, PC + M, and -KG were significantly lower in LTx recipients than in the control subjects when expressed relative to mitochondrial suspension protein (p < 0.05, Table 7). There was a tendency for a lower MAPR per unit weight of mitochondrial suspension protein in LTx recipients, in the presence of PPKM (p = 0.053), but MAPR in the presence of S + R did not differ compared with the control subjects. When expressed relative to muscle weight, MAPR, in the presence of all substrates, was lower in LTx recipients than in control subjects (p < 0.05, Table 8).

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Compared with normal matched control subjects, LTx recipients had a shorter exercise duration, a lower O₂peak, and an earlier rise in plasma lactate during the incremental exercise test. As a group LTx recipients demonstrated a mild mixed obstructive/restrictive ventilatory defect. However, only one of seven LTx recipients reached ventilation limitation during exercise and none significantly desaturated. Quadriceps femoris muscle biopsy showed lower proportion of type I fibers, lower ATP content, and reduced oxidative enzyme activity in LTx recipients. In addition, LTx recipients had reduced skeletal muscle oxidative phosphorylation as evidenced by a reduced muscle MAPR in the presence of a range of substrates, both when expressed relative to muscle mass and relative to mitochondrial protein content.

The marked alteration in muscle fiber-type proportion is likely to be a major contributor to the reduced exercise tolerance in LTx recipients. In general, type I fibers are fatigue- resistant, and rich in mitochondria with high levels of oxidative enzymes. The type IIb population of type II fibers have fewer mitochondria and tend to be better suited for anaerobic metabolism possessing increased glycolytic capabilities. Therefore, in LTx recipients, a shift of type I to type IIb fibers would reduce the oxidative capacity of skeletal muscle, with an associated increased reliance upon anaerobic metabolism.

Recently, it has been reported (19) that chronic heart failure patients without peripheral vascular disease have a lower proportion of type I fibers and decreased activities of mitochondrial enzymes CS and HAD in skeletal muscle before cardiac transplantation. Although the activity of oxidative enzymes was increased, the abnormalities in muscle fiber-type proportions were not changed 12 mo posttransplant. In this present study, the low type I fiber type proportion and reduced oxidative enzyme activity may reflect the persistence of pretransplant abnormalities previously reported (3, 20), but this needs to be confirmed.

A low ATP concentration and accumulation of lactate and IMP in resting skeletal muscle of LTx recipients most likely reflect reduced muscle mitochondrial oxidative phosphorylation and/or a higher tendency for glycolysis in LTx recipients' muscle. The ratio of ATP/ADP is an indicator of the efficiency of transfer from energy-producing processes to energy utilizing processes in cells. The ratio of ATP to ADP can regulate the catabolism of ATP in the mitochondria. A decrease in the ATP/ADP concentration ratio can lead to activation of the enzyme AMP deaminase, which catalyzes the conversion of AMP to IMP (21). This study has shown significantly higher IMP and lactate concentrations in the resting skeletal muscle of LTx recipients. Accordingly, the abnormalities of metabolism in skeletal muscle are also most likely secondary to the impaired mitochondrial function and the reduced proportion of type I fibers.

The LTx recipients exhibited impaired in vitro mitochondrial ATP production rate whether using carbohydrate, fatty acids, or amino acids as substrates. The low mitochondrial function appears to be the result of a combination of lower mitochondrial mass and volume (indicated by low mitochondrial protein content in muscle and perhaps due [22] to reduction in type I fiber proportion), and functional impairment of the individual mitochondria in vitro (indicated by low MAPR expressed in terms of mitochondrial protein in the suspension).

The cause of the reduced MAPR and enzyme activities in LTx recipients' muscle, observed in the present study is not clear. Previous studies have demonstrated altered mitochondrial capacity in skeletal muscle related to muscle disuse (23), immobilization (24), physical deconditioning (25, 26), severe chronic pulmonary diseases (3, 5, 27), and chronic heart diseases (4, 28). In addition, the impaired mitochondrial electron transport in vitro could be a source of free radicals, which have been demonstrated to cause loss of contractile function and accelerated cell death in cardiomyocytes (29). Therefore, in the present study, it is possible that impairment of the mitochondrial electron transfer chain (ETC) in skeletal muscle may produce free radicals, further aggravating mitochondrial dysfunction and injuring skeletal muscle.

It has been demonstrated that CyA inhibits respiration in isolated human kidney mitochondria (30) in a dose-dependent manner. Inhibition of O₂ consumption in mitochondria isolated from rat skeletal muscle has been reported in response to high concentration of CyA (25 µg/ml) (7). In addition, mitochondria isolated from skeletal muscle of rats fed a high dose of CyA (20 mg/kg/d for 14 d) were similarly affected (8). The latter report showed there was a significant linear correlation between endurance exercise time in rats and impaired ADP-stimulated mitochondrial respiration.

A combination of prednisolone and azathioprine may worsen CyA effects on rat renal mitochondrial function (31). This raises the prospect that effects on oxidative phosphorylation in skeletal muscle reported with CyA may be potentiated by other standard immunosuppressive agents. Furthermore, prolonged use of corticosteroids, such as prednisolone, can inhibit protein synthesis, increase protein catabolism, and lead to negative nitrogen balance in rat skeletal muscle (32) and is also strongly implicated in chronic myopathy and muscle wasting in humans (33).

The mechanism by which CyA inhibits mitochondrial function is still not clear. Hokanson and coworkers (7) have suggested that CyA may reduce mitochondrial ATP synthesis through effects on the mitochondrial membrane. Alternatively, CyA may influence mitochondrial function through its effects on the so-called inner mitochondrial membrane transport pore (MTP). CyA may block this pore which controls Ca²⁺ passage across the inner mitochondrial membrane, but the consequences of this effect on mitochondrial function are not clear. Some reports have suggested that CyA-related blockage of the MTP leads to a decrease in mitochondrial ATP synthesis and cell death (34, 35). Additionally, CyA has been shown to enhance the generation of reactive oxygen species (ROS) in vitro (36) and lipid peroxidation in vivo (37). Overproduction of free radicals might inactivate mitochondrial ETC components and ATPase (38). In contrast, other data have demonstrated beneficial or protective effects of CyA inhibition of the MTP by reducing the damage caused by anoxia and oxidant stress injury or ischemic reperfusion damage (39).

CyA causes renal and peripheral vasoconstriction, thus inducing renal damage and systemic hypertension (40). Systemic hypertension may cause skeletal muscle type I fiber shift to type II (41) and muscle mitochondrial capacity may decline similar to peripheral vascular diseases (42). In this study, five LTx recipients had mild hypertension and this may have been one factor affecting mitochondrial function in these individuals.

In conclusion, the current study has demonstrated that LTx recipients have reduced endurance tolerance, characterized by a low O₂peak, and early lactate threshold. Their resting skeletal muscle shows a lower proportion of type I fibers which may partially cause the low mitochondrial ATP production rate, low mitochondrial oxidative enzyme activities, and the demonstrated abnormal metabolism (including low ATP content associated with high levels of lactate and IMP). After correction for reduced mitochondrial protein, a marked reduction in MAPR persists. Thus, the low MAPR does not seem to be entirely explained by the decreased type I fiber proportion and an additional abnormality of oxidative phosphorylation appears to be present. It appears most likely that a pretransplant reduction in type I fiber proportion persists after recovery from lung transplantation and in addition, oxidative phosphorylation is impaired by cyclosporin A (perhaps potentiated by prednisolone and azathioprine). Although there are many potential causes of exercise limitation in the LTx recipient, impaired oxidative capacity of the peripheral skeletal muscles shown in this study may prove the most prevalent and important.