Changes in indices of antioxidant status, lipid peroxidation and ...

Clinical Science (1999) 96, 105–115 (Printed in Great Britain)

Changes in indices of antioxidant status, lipid peroxidation and inflammation in human skeletal muscle after eccentric muscle actions R. CHILD*, S. BROWN†, S. DAY‡, A. DONNELLY§, H. ROPERR and J. SAXTON¶ *Muscle Research Centre, Department of Medicine, University Clinical Departments, Liverpool University, c/o The Duncan Building, Daulby Street, Liverpool L69 3GA, U.K., †School of Health and Sport Science, University of North London, London N7 8DB, U.K., ‡Department of Health, Staffordshire University, Stafford ST4 2DF, U.K., §Department of Physical Education and Sport Science, Limerick University, Limerick, Ireland, RDepartment of Paediatrics, Birmingham Heartlands Hospital, Birmingham B9 5SS, U.K., and ¶Institute of Sports Medicine and Exercise Science, Sheffield University, Sheffield S10 2TA, U.K.

A

B

S

T

R

A

C

T

This study investigated the effects of chronic muscle inflammation on indices of antioxidant status and muscle injury after eccentric exercise. Eight subjects each performed 70 maximal voluntary eccentric muscle actions on an isokinetic dynamometer, using the knee extensors of a single leg. Venous blood samples were collected into serum and EDTA tubes 5 and 3 days before exercise, immediately before exercise, and then again on days 3, 4, 5, 6, 7, 10 and 12 after the bout. Needle biopsies were taken from the vastus lateralis of six subjects, a week before exercise (baseline), and again on days 4 and 7 post-exercise. The concentrations of malondialdehyde in plasma and muscle were used as markers of lipid peroxidation. Creatine kinase activity, βglucuronidase activity and total antioxidant capacity were determined in serum. In muscle, aqueous and bound total antioxidant capacity, the aqueous sulphydryl concentration, and βglucuronidase and glucose-6-phosphate dehydrogenase activity were determined. No changes were detected in serum total antioxidant capacity, serum creatine kinase and β-glucuronidase after the baseline biopsy. After exercise serum creatine kinase and β-glucuronidase were elevated although other serum measures were unchanged. In muscle, aqueous and bound total antioxidant capacity, sulphydryls, glucose-6-phosphate dehydrogenase and β-glucuronidase were all elevated. Despite evidence of inflammation in this study, muscle antioxidant status was not compromised, and malondialdehyde was unaltered in muscle and plasma. Therefore, this study provides no evidence that chronic muscle inflammation compromises antioxidant status or increases lipid peroxidation.

INTRODUCTION Eccentric actions involve generating force while lengthening active muscle. Such activities can result in disruption to connective tissue [1,2] and diverse myocellular components including the sarcolemma, myofibrils and cytoskeleton [3–5]. Exercise-induced muscle damage is often accompanied by immediate decrements in maxi-

mum voluntary contractile force and the 20 : 100 Hz force ratio, with a delayed rise in muscle soreness and serum creatine kinase (CK ; EC 2.7.3.2) activity on subsequent days. Such changes have been used as indices of muscle damage in humans [2,6–8]. Faulkner et al. [9] proposed that the delayed rise in serum CK and muscle soreness may be manifestations of secondary muscle damage. Mechanical factors are closely associated with the

Key words : enzyme efflux, free radicals, muscle damage, muscle sulphydryls, oxidative stress. Abbreviations : βG, β-glucuronidase ; CK, creatine kinase ; CV, coefficient of variation ; G6PDH, glucose-6-phosphate dehydrogenase ; TAC, total antioxidant capacity. Correspondence : Dr Robert Child.

# 1999 The Biochemical Society and the Medical Research Society

105

106

R. Child and others

initiation of muscle injury during eccentric actions [5,10–13]. Secondary damage, however, may have metabolic origins [5,11,14] and several reports have implicated free radicals in the damage process [13,15,16]. After initial injury, oxidative stress could be increased by the presence of neutrophils and macrophages in muscle. The production of free radicals by such infiltrates has the potential to damage lipid membranes and produce cell necrosis [17,18]. Free-radical-mediated injury, resulting from muscle inflammation, has been used to explain several muscle damage phenomena. Warren et al. [13] proposed that free radical damage might be responsible for the delay in force loss, sometimes observed after eccentric exercise [8,16] ; while Cannon et al. [19] suggested free radicals from cellular infiltrates could result in delayed elevations in CK. The mechanism underlying free-radical-mediated tissue injury may involve increases in oxidative stress overwhelming antioxidant defences [18] and thereby allowing uncontrolled oxidation of cellular constituents. Malondialdehyde has been used as a marker of fatty acid oxidation in several studies of inflammatory myopathy [20–22] and is one of several cytotoxic peroxidation products. The simultaneous assessment of free radical protection in muscle has not been performed, although such an approach may help to elucidate the possible role of free radicals in exercise myopathy. When individual antioxidants are assessed, it is necessary to determine the concentration of a spectrum of compounds to provide an index of total free radical protection. Developments in assay techniques now allow quantification of the total free-radical scavenging capacity of complex antioxidant solutions [23,24] and tissue extracts [25]. Indices of free radical damage in human muscle have previously been evaluated after eccentric and concentric exercise [26]. These authors found no evidence that eccentric exercise resulted in free radical damage. However, there was also no evidence that the eccentric exercise bout produced mechanically mediated muscle injury, or resulted in muscle inflammation. These events may be essential prerequisites for free-radical-mediated muscle damage to occur after eccentric actions. This study tested the hypothesis that chronic muscle inflammation might compromise muscle antioxidant status, thereby predisposing damaged muscle to free-radical-mediated secondary injury.

MATERIALS AND METHODS Subjects After approval from the local ethics committee and completion of informed consent documents, eight physically active but untrained volunteers (four male and four female, aged 21–31 years) participated in the study. # 1999 The Biochemical Society and the Medical Research Society

Percutaneous electrical myostimulation Knee extensor muscles were stimulated using a Bioscience 200 stimulator (Bioscience, Sheerness, Kent, U.K.) producing a unidirectional square wave pulse width of 0.5 ms. Electrical potential was applied via large copper electrodes (minimum dimensions 15 cmi10 cm, 0.2 mm thick). These were contained within felt pouches (soaked in water) and affixed proximally and distally over the knee extensors with elasticated straps. The voltage used for myostimulation was sufficient to elicit 30 % of the maximum voluntary isometric contractile force of the knee extensors, at a knee angle of 1.92 radians on the day of testing.

Isometric knee extensor force measurements Volunteers were familiarized to isometric force testing on at least two visits to the laboratory before exercise. Isometric force measures were performed in duplicate on a Kin-Com isokinetic dynamometer (Chattecx Corp., TN, U.S.A.), at a knee angle of 1.92 radians. The following measurements were made in duplicate immediately before and 5 min after exercise. 1. Maximum knee extensor contractile force was determined by motivating subjects to maximally contract the knee extensors for 3 s. During contraction electrical myostimulation was applied to the knee extensor muscles at a frequency of 100 Hz and at a voltage sufficient to elicit 30 % of each subject’s maximum voluntary contractile force. This was superimposed on the voluntary contraction for 1 s to ensure the maximum force generating potential of the muscle was accurately determined. 2. The mean tetanic force produced during a 1-s pulse of 100 Hz electrical myostimulation. 3. The mean tetanic force produced during a 1-s pulse of 20 Hz electrical myostimulation. Measures 2 and 3 were used to calculate the 20 : 100 Hz force ratio which was used as an index of low-frequency fatigue.

Exercise Each subject performed 70 maximal voluntary eccentric muscle actions on a Kin-Com isokinetic dynamometer, using the knee extensors of a randomly selected leg. Exercise was performed in a prone position using a knee joint range of motion from almost full extension to almost full flexion. Exercising in this position increased the length at which the knee extensors performed the eccentric muscle actions, which has previously been shown to increase indirect indices of muscle damage [27]. A preload was used so that each subject produced maximal knee extensor force before movement of the

Human exercise myopathy and antioxidant status

dynamometer lever arm ; this allowed isometric force to rise before the eccentric muscle action. Myostimulation parameters in the prone position were identical to those used to determine the pre-exercise maximum knee extensor contractile force (i.e. a voltage sufficient to elicit 30 % of the maximum knee extensor contractile force, using a stimulation frequency of 100 Hz). This was applied for 0.5 s before movement of the lever arm and throughout the eccentric action ; therefore muscle force was generated by both voluntary activation and percutaneous electrical myostimulation. Each eccentric action was performed at an angular velocity of 1.75 radians\s, with each repetition separated by a 10-s rest period. During this time the leg was returned to the start position by the experimenter at 1.05 radians\s.

Muscle soreness Soreness was assessed before exercise and on each subsequent day at eight muscle regions (six extensor and two flexor), depicted on a questionnaire. The subject determined soreness by self-palpation of the sites in a seated position with the muscle relaxed, then reported perceived soreness on a scale between ‘ normal ’ (1) and ‘ very very sore ’ (10). Each subject’s soreness values (for the six knee extensor muscle sites) were summed each day, and the total used as the criterion score.

Blood collection Ten-millilitre blood samples were drawn from an antecubital vein 5 days, 3 days and immediately pre-exercise, then again on days 3, 4, 5, 6, 7, 10 and 12 after the exercise bout. Blood was dispensed in potassium–EDTA tubes (to be centrifuged immediately at 1500 g for 10 min), or serum tubes (to be centrifuged using the same protocol after clotting at room temperature). Serum and plasma were removed and stored at k80 mC until analysis.

Biopsy collection and preparation Muscle biopsies were taken under local anaesthesia (2 % lignocaine) using a Bergstrom-type needle (diameter 6.0–6.5 mm). Each sample was taken from the distal region of the vastus lateralis using a slightly different location on each occasion. Baseline biopsies were taken 4 days before exercise from a randomly selected leg, with further samples from the exercised leg 4 and 7 days after exercise. Each sample was quickly divided for histological and biochemical analysis, frozen in isopentane (chilled in liquid nitrogen) and stored at k80 mC until analysis.

Biochemical analysis Biopsies were first dissected free of visible connective tissue and fat. The remaining tissue was then homogenized on ice, in helium-degassed ice-cold 50 mmol\l phosphate buffer at pH 7.4 (approximately 10 % weight for volume). This was performed using a glass on glass homogenizer (Kontes Duall 20, Kontes Glass Co., NJ, U.S.A.), which was turned by hand to minimize elevations in temperature during tissue grinding. The homogenate was then transferred to pre-weighed vials and centrifuged at 13 000 g for 10 min at 3 mC. The supernatant was then divided into aliquots, frozen in liquid nitrogen and stored at k80 mC. After removal of the remaining supernatant, the pellet was washed twice in ice-cold degassed 50 mmol\l phosphate buffer at pH 7.4. Each wash consisted of resuspending the pellet in 1 ml of buffer, vortexing for 1 min (to form a milky solution), centrifugation at 13 000 g and removal of the supernatant layer. The pellet was then frozen in liquid nitrogen and freeze dried at k20 mC for 5 h, after which the combined weight of the vial and pellet were noted. The pellets were then stored in sealed containers at k80 mC. Exercise-induced muscle injury has been shown to decrease the aqueous protein content of the damaged tissue [28], and expression of the concentration or activity of a compound relative to aqueous protein overestimates any real increases under these conditions. To prevent such artifactual elevations, the changes reported in Table 2 are expressed relative to tissue weight.

Creatine kinase activity Serum CK was used as an indirect index of exerciseinduced muscle damage and was measured using a diagnostic kit (Sigma No. 47–10, Sigma, Poole, Dorset, U.K.). At least duplicate analyses were made of each sample, and CK activities were calculated as a mean of two values that differed by no more than 10 % of the lower value. The inter-assay coefficient of variation (CV) was 3.8 %.

β-Glucuronidase activity

β-Glucuronidase (βG ; EC 3.2.1.31) activity in serum was used as an indirect marker of inflammation, and as an index of tissue injury in muscle. Activity was determined via the cleaving of phenolphthalein from phenolphthalein mono-β-glucuronic acid, using a diagnostic kit (Sigma No. 325). Liberation of 1 µg of phenolphthalein per litre per hour, at 56 mC, was defined as one Sigma unit of activity. The inter-assay CV was 3.7 %.

Histological analysis

Glucose-6-phosphate dehydrogenase activity

Specimens were mounted on cryostat blocks, cut into transverse sections (5–6 µm thick) at k20 mC, mounted on glass slides, air dried and stained with haematoxylin\eosin.

Muscle glucose-6-phosphate dehydrogenase (G6PDH ; EC 1.1.1.49) was measured as a marker of cellular infiltration, as reported previously [13,29]. Activity was determined via changes in absorbance at 340 nm, result# 1999 The Biochemical Society and the Medical Research Society

107

108

R. Child and others

ing from the reduction of NADP+ to NADPH using a diagnostic kit (Sigma No. 345-UV). The procedure involved adding 25 µl of aqueous muscle supernatant to 0.5 ml of G6PDH reagent and 1 ml of G6PDH substrate solution ; activity was then measured at 37 mC. Results were converted to international units (i.u.) of activity using the appropriate temperature correction factor. One unit was defined as that amount of activity which converted 1 µmol of substrate per minute at 30 mC. The inter-assay CV was 6.3 %.

for 5 min. The supernatant layer was carefully removed for analysis without further preparation. The homogenization buffer, methyl sulphoxide and serum diluent were taken from the same stock and verified to have no antioxidant properties (within the detection limits of the assay), before and after sample analysis. The inter-assay CV for serum was 4.0 %, with intra-assay CVs of 0.4 %, 1.4 % and 0.9 % for muscle aqueous extracts, muscle bound extracts and serum.

Aqueous sulphydryls Serum urate The concentration of urate in serum was determined as this is the most abundant antioxidant in extracellular fluids [24]. Urate was measured using a diagnostic kit (Sigma No. 686), based on the uricase–peroxidase system. The inter-assay CV was 4.0 %.

Total antioxidant capacity Total antioxidant capacity (TAC) was evaluated to provide a global index of antioxidant protection. This was determined using reagents, equipment and procedures described by Whitehead et al. [24], utilizing an enhanced chemiluminescence technique to measure free radical scavenging capacity. Light emission occurs when the chemiluminescent substrate (luminol) is oxidized by hydrogen peroxide or perborate, in a reaction catalysed by horseradish peroxidase. The stability and intensity of light output is enhanced by the addition of paraiodophenol. Continuous light output depends upon on the constant production of free radical intermediates derived from para-iodophenol, luminol and oxygen. For this reason light emission is sensitive to antioxidant compounds, but will be restored when the added antioxidants have been consumed. As the generation of radical intermediates is constant, the time period of light suppression is directly related to the amount of antioxidant present. Therefore, this assay is a sensitive measure of soluble chain-breaking antioxidants (i.e. those molecules capable of reducing free radicals that would otherwise initiate and propagate free radical chain reactions). Antioxidant capacity was determined relative to 6-hydroxy-2,5,7,8-tetramethylchroman-2-carboxylic acid (Trolox), a soluble vitamin E analogue, and expressed as Trolox equivalents (Trolox Eq.). Each analysis was performed in duplicate. Serum antioxidant capacity was determined as described by Whitehead et al. [24]. Aqueous muscle antioxidants were determined without further preparation of the aqueous muscle supernatant. The antioxidant capacity of the biopsy pellet was determined after re-weighing (to determine there was no moisture uptake). The pellet was then resuspended in 200 ml\g methyl sulphoxide ACS reagent (Sigma) at room temperature, vortexed for 5 min and centrifuged at 13 000 g # 1999 The Biochemical Society and the Medical Research Society

Total aqueous sulphydryls were assessed, as these are the most concentrated antioxidant compounds in skeletal muscle tissue [15,30,31]. These were determined using a colorimetric kit (GSH-400, Bioxytech, Bonneuil, France) by changes in absorbance at 356 nm. This assay is based on the formation of thioethers, resulting from the reaction of 4-chloro-1-methyl-7-trifluromethyl-quinolinium and sulphydryls present in the sample. The interassay CV was 1 %.

Malondialdehyde The concentration of malondialdehyde was used as an index of lipid peroxidation and was determined following methods described by Young and Trimble [32], involving derivatization with thiobarbituric acid. We have demonstrated that this assay is sensitive enough to detect increases in the plasma malondialdehyde concentration after distance running [33]. The malondialdehyde– thiobarbituric acid adduct was eluted using a Perkin– Elmer HPLC system (Perkin–Elmer, Connecticut, U.S.A.), which consisted of an ISS 200 Advanced LC sample processor, Model 250 Binary LC pump and LS 40 fluorescence detector. Chromatographic separation was performed using an ODS2 analytical column (internal diameter 25 cmi4.6 mm), with a 3-cm guard cartridge (Phase Separations, Deeside, Flintshire, U.K.). The injection volume was 80 µl for muscle samples and 50 µl for plasma. Adduct fluorescence was determined at 553 nm after excitation at 532 nm. Quantification was performed by integration of peak area, using 1,1,3,3-tetramethoxypropane as a standard. Use of these procedures and equipment resulted in an inter-assay CV of 8.6 %, with sensitivity to 0.1 µmol\l, which was five times lower than the malondialdehyde concentration of the most dilute biological sample. The method described is highly specific for the malondialdehyde–thiobarbituric acid adduct and involved rigorous steps to separate this adduct from contaminating compounds. These included the precipitation of proteins, elution of contaminants using HPLC, and the utilization of the fluorescent rather than the absorbent properties of the malondialdehyde adduct. In combination, these procedures prevent interference from a variety of thiobarbituric acid-reactive compounds


including biliverdin, bilirubin, urea, creatinine and glucose [32,34]. This is an important methodological improvement over spectrophotometric assays, which have been criticized for their poor specificity [18,34].

(a)

Serum CK (i.u./l)

20 000

Statistics The ordinal soreness measurements were analysed using Wilcoxon matched-pairs tests. Functional measurements were analysed using paired t-tests, while biochemical data were analysed using repeated measures analysis of variance (ANOVA). Where significant differences were established using an ANOVA, Newman–Keuls post-hoc tests were used to determine significance at specific time points. Biochemical measurements in biopsy specimens were analysed after logarithmic transformation, to minimize variance heterogeneity over time. Significance was accepted when P 0.05. Soreness data are presented as means with the range, all other data are reported as meanspS.E.M. Blood and functional data are for eight subjects, whereas muscle biochemical data are for six subjects.

**

** P < 0.0001, ANOVA

15 000 ** 10 000

5000

0

Malondialdehyde concn. (µmol/l)

(b) 2.5 2.0 1.5 1.0 0.5 0

Biopsy

EXERCISE

Biopsy

Biopsy NS

BASELINE

–5 –3 0

POST EXERCISE

3 4 5 6 Time (days)

7

10 12

Figure 1 Serum creatine kinase (CK) activity (a) and plasma malondialdehyde (b), in response to muscle biopsies and exercise

RESULTS

**P

Isometric force

Muscle soreness

Exercise resulted in a decrease in maximum contractile force from 259p25 to 131p12 Nm (P 0.001, t-test), and a decline in the 20 : 100 Hz force ratio from 0.74p 0.03 to 0.29p0.04 (P 0.001, t-test). The reliability of duplicate measures of maximum contractile force, force produced by myostimulation at 100 Hz and force produced by myostimulation at 20 Hz was 4.0 %, 4.1 % and 4.9 % respectively.

Changes in knee extensor muscle soreness are shown in Table 1. Soreness was elevated above pre-exercise values between days 2 and 6 after exercise (P 0.05, Wilcoxon test).

0.01, compared with pre-exercise. NS, not significant.

Blood biochemistry No changes were detected in serum TAC, serum urate (Table 1) or plasma malondialdehyde (Figure 1b).

Table 1 Serum total antioxidant capacity (TAC), urate concentration, β-glucuronidase activity and muscle soreness, before and after eccentric exercise

*P

0.05 compared with pre-exercise. Time (days) Pre-exercise

Post-exercise

Parameter

k5

k3

0

1

2

3

4

5

6

7

8

9

10

11

Serum TAC (µmol Trolox Eq./l) Serum urate concn. (µmol/l) Serum β-glucuronidase activity (Sigma units/ml) Muscle soreness (arbitrary units)

477 (31) 271 (16) 10.1 (3.2) –

501 (38) 284 (19) 6.2 (2.0) –

439 (38) 282 (19) 7.9 (2.3) 6.0 (0)

–

–

–

–

–

23 (14–33)

36* (27–43)

487 (38) 265 (21) 12.8 (4.6) 8.0 (6–13)

–

–

484 (30) 275 (14) 9.9 (3.2) 12* (8–18)

6.0 (0)

6.0 (0)

473 (45) 280 (15) 12.5 (3.4) –

–

–

485 (24) 271 (13) 30.2 (12.4) 18* (13–24)

–

–

458 (34) 275 (14) 10.5 (3.8) 27* (16–36)

–

–

502 (42) 285 (19) 12.1 (3.6) 39* (24–48)

– – –

12 487 (36) 291 (18) 14.3 (3.6) –


109

110

R. Child and others

Figure 2

Representative light micrograph 5 days before exercise, showing polygonal fibres without cellular infiltration

Figure 3 Representative light micrograph 4 days after exercise, showing increasing degrees of cellular infiltration in fibres I, II and III

(a) Fibres with central nuclei on day 4 after exercise, (b) increased numbers of nuclei in the endomysium. # 1999 The Biochemical Society and the Medical Research Society


Figure 4

Light micrograph 7 days after exercise, showing extensive cellular infiltration

(a) Apparent loss of muscle fibres, (b) normal fibre surrounded by nuclei, (c) necrotic fibre. Table 2

*P

Changes in muscle inflammatory indices and antioxidants after eccentric exercise

0.05, **P

0.01 relative to baseline. NS, not significant.

Parameter

Baseline

Day 4

Day 7

Statistical changes

β-Glucuronidase (log pg phenolphthalein:h−1:g−1 wet weight muscle) G6PDH (log pmol NADPH:min−1:g−1 wet weight muscle) Total sulphydryls (log pmol/g wet weight muscle) Aqueous TAC (log pmol Trolox Eq/g wet weight muscle) Bound extract TAC (log pmol Trolox Eq/g dry weight muscle) Malondialdehyde (log pmol/g wet weight muscle)

4.94 (0.04) 4.72 (0.15) 2.47 (0.03) 6.11 (0.05) 7.38 (0.06) 5.07 (0.15)

5.17 (0.06) 5.07 (0.06) 2.45 (0.03) 6.12 (0.03) 7.55* (0.06) 4.72 (0.06)

5.42** (0.11) 5.36* (0.19) 2.66 (0.05) 6.26* (0.02) 7.70** (0.06) 5.36 (0.19)

P

0.001, ANOVA

P

0.01, ANOVA

P

0.05, ANOVA

P

0.01, ANOVA

P

0.001, ANOVA

Changes in the serum activities of CK and βG are shown in Figure 1(a) and Table 1 respectively. Serum CK did not differ between the baseline biopsy and pre-exercise. After exercise, however, serum CK was elevated above baseline values (P 0.0001, ANOVA) in all subjects. Newman– Keuls tests established elevations on days 3, 4 and 5 after exercise (P 0.01), with the highest activity on day 4. Serum βG was unaltered after the baseline muscle biopsy (P 0.05, ANOVA), but after exercise it was increased above pre-exercise values (P 0.05, ANOVA). The

NS

highest activity was measured on day 5, but this was not significantly different from pre-exercise levels (P 0.05, Newman–Keuls test).

Histology Histological changes in biopsy specimens are shown in Figures 2, 3 and 4. Baseline samples appeared normal, with tightly packed polygonal fibres arranged into welldefined fascicles. Nuclei were sparsely distributed in the perimysium and endomysium, with no central nuclei # 1999 The Biochemical Society and the Medical Research Society

111

112

R. Child and others

within fibres (Figure 2). After exercise inflammation was observed in biopsy specimens from all subjects, although the magnitude of cellular infiltration was highly variable between subjects. Some biopsies showed fibres with central nuclei on day 4 after exercise, with increased numbers of nuclei in the endomysium at this time. Despite these changes most fibres maintained their normal appearance (Figure 3). By day 7 after exercise cellular infiltration was more extensive, and in some subjects resembled an inflammatory muscle myopathy (Figure 4).

Muscle biochemistry Changes in muscle enzymes, antioxidants and malondialdehyde are shown in Table 2. After exercise G6PDH was increased from baseline (P 0.01, ANOVA), being significantly elevated on day 7 (P 0.05, Newman–Keuls test). β-Glucuronidase was increased from baseline (P 0.001, ANOVA), being significantly elevated on day 7 (P 0.05, Newman–Keuls test). Muscle sulphydryls increased from baseline (P 0.05, ANOVA), although the elevation on day 7 did not attain statistical significance (P 0.05, Newman–Keuls test). The TAC of the aqueous muscle extract increased from baseline (P 0.01, ANOVA), and was significantly elevated on day 7 (P 0.01, Newman–Keuls test). The TAC of the ‘ bound ’ muscle extract increased from baseline (P 0.001, ANOVA), and was significantly elevated on days 4 and 7 (P 0.05 and P 0.01 respectively, Newman–Keuls test). No changes were observed in the muscle malondialdehyde concentration.

DISCUSSION A high-force eccentric exercise model was used to investigate the possibility of free-radical-induced muscle injury, resulting from chronic inflammation. The protocol employed had the advantage of damaging a specific muscle group, thereby allowing indirect damage markers to be related to changes in the damaged musculature. Classic responses in muscle damage indices were observed after exercise, including reduced muscle contractility, delayed elevations in serum CK (Figure 1a), serum βG and muscle soreness (Table 1). As the serum indices of muscle damage were not influenced by the baseline muscle biopsy, it appears that this procedure did not produce significant muscle injury. Therefore the rise in serum CK and βG after the bout of eccentric muscle actions was probably a consequence of exercise-induced muscle damage. In the exercised muscle a progressive increase in cellular infiltration was observed. Such changes have been well-documented in humans after muscle-damaging exercise [2,4,6,7]. # 1999 The Biochemical Society and the Medical Research Society

Techniques for muscle damage quantification using light microscopy typically use the number and location of nuclei to classify fibres as ‘ damaged ’ (e.g. see [35]). Such classification systems have many limitations, which have been discussed previously [9,36]. Faulkner et al. [9] proposed that the decline in maximum force production was the most valid measure of muscle damage. In the present investigation the magnitude of force loss and decline in the 20 : 100 Hz force ratio suggested that exercise resulted in muscle injury. As these contractile measurements were made shortly after exercise, fatigue may also have contributed to the observed force deficits. It has been suggested that generation of maximal muscle forces might produce further disruption to fibres which were damaged, or undergoing repair [4]. Therefore, to minimize such experimental damage artefacts, further assessment of muscle function was not performed in the days after exercise. The muscle activity of the lysosomal enzyme βG was used as a marker of muscle injury in the present study. In rodents, activity of this enzyme closely reflects exerciseinduced muscle damage [36,37], and is significantly correlated with the histopathological state of the muscle [38]. Phagocytes possess high βG [38–40] and increased numbers of macrophages in muscle fibres and connective tissue are thought to reflect damage to these structures [2,29]. Elevated βG in damaged muscle has also been attributed to upregulation in fibres without cellular infiltrates [38], and appears to localize at cytosolic sites known to be damaged by exercise, including the myofibrillar I band [41]. As such ultrastructural disruption is only visible using electron microscopy and connective tissue damage is difficult to quantify using light microscopy, muscle βG may be a more sensitive marker of muscle damage than histological analysis using light microscopy. Thus, the use of muscle βG as a marker of injury in the present study may have overcome limitations of quantifying muscle damage using histological and contractile measures. The rise in muscle βG suggests there was an increase in damage to the knee extensors between days 4 and 7 after exercise. Macrophages also demonstrate high G6PDH activity relative to muscle tissue, and increased activity of this enzyme has been used as a marker of cellular infiltration in exercisedamaged rodent muscle [13,29]. In the present study, the changes in G6PDH suggested that cellular infiltration increased between days 4 and 7 after exercise. Biochemical evidence of inflammation (Table 2) was verified by histological analysis, which showed increased cellular infiltration between days 4 and 7 after exercise. Therefore, muscle βG and G6PDH appeared to reflect the increased numbers of nuclei in muscle biopsy specimens. The magnitude of cellular infiltration was highly variable between subjects, as in previous studies [7]. The aggregation of nuclei in the perimysium and endomysium is similar to previous reports in humans [2,42] and may


reflect exercise-induced connective tissue damage [2,29]. The time scale of these changes is also consistent with previous reports of exercise myopathy in humans [19,42]. Cellular infiltration and the release of acid hydrolases such as βG are characteristic of inflammation. The occurrence of such events suggests that neutrophils and macrophages were activated, which is associated with increased production of oxygen radicals [43,44]. A variety of inflammatory disease states have been associated with reduced serum TAC [23,24], while rhabdomyolysis has been associated with hyperuricaemia [45]. Despite evidence of an inflammatory response and rhabdomyolysis (Figure 1a), no changes were observed in serum TAC or urate in this study (Table 1). After exercise muscle antioxidant protection was enhanced (Table 2). As reduced glutathione is an abundant cytosolic antioxidant in tissues [15,46], the rise in aqueous TAC in muscle (Table 2) may have partially reflected increases in this antioxidant. This is consistent with the concurrent rise in aqueous sulphydryls (Table 2). Rajguru et al. [31] reported elevated sulphydryls in rodent skeletal muscle 30 min after swimming exercise, which the authors suggested might reflect increased muscle cysteine or glutathione. Similarly, Maulik et al. [47] found endotoxin-induced oxidative stress increased ascorbate and thiol components of the antioxidant reserve in rat myocardium. The authors proposed that this response could contribute to the protection of the heart against oxidative injury, and similar responses are thought to prevent degeneration in cultured human fibroblasts [48]. The elevation in aqueous TAC in the present study may provide evidence that similar changes occur in human muscle after damaging exercise. Such a response might reduce undesired oxidation by infiltrating cells and\or could facilitate tissue repair [31,46]. The rise in the TAC of the bound muscle extract may be consistent with an increase in membrane-bound antioxidants in the present investigation. If such a response occurred in the sarcolemma it might increase resistance to oxidative cellular injury, or facilitate biosynthesis in cells which sustained sub-lethal injury [31]. The disparate time course for changes in muscle TAC suggests that changes in bound antioxidants are not simply a reflection of changes in the aqueous environment (Table 2), although which bound antioxidants were actually elevated is a matter of conjecture. The distinction between antioxidants in muscle fibres and cellular infiltrates cannot be determined using the techniques employed in the present study. However, as antioxidants are highly concentrated within the cell [30,46,49], it seems unlikely that the large increases in the TAC of the biopsy specimens could be a direct consequence of the relatively small number of infiltrating cells observed in most sections. Therefore, it appears that human skeletal muscle may respond to inflammation by increasing its antioxidant reserve. Similar changes have

been observed in a variety of experimental models which, like inflammation, increase oxidative stress [31,47,50]. The malondialdehyde concentration in muscle and plasma was unaltered in response to the muscle biopsy procedure or exercise. This result is surprising when considering the extent of inflammation present in the muscle. One possibility is that the increase in muscle antioxidants minimized undesired peroxidative damage. Inflammation without evidence of peroxidation in muscle has previously been reported in exercise myopathy in mice [20], and in muscular dystrophy in both mice [22] and humans [21]. This is the first study to report inflammation without evidence of peroxidation in human exercise myopathy. The time points for blood and biopsy collection were selected based on reports which showed a paucity of cellular infiltrates within skeletal muscle fibres 2 days after eccentric exercise [2] and increasing numbers of myocellular infiltrates between 4 and 7 days after damaging exercise [7,42]. However, it has also been demonstrated that neutrophil infiltration into the human vastus lateralis can occur less than an hour after downhill running, and that significant numbers of neutrophils are still present 5 days later [51]. Although neutrophils may have been present in the biopsy samples collected 4 days after exercise in the present investigation, the peak in neutrophil activation may have occurred some time before this [51]. Therefore, the possibility that our biopsy sampling strategy missed the peak in neutrophil-mediated oxidative stress cannot be discounted. Further studies are required to establish whether free-radicalmediated damage occurs in the first 48 h after exercise and to determine if such injury arises as a consequence of neutrophil activation.

CONCLUSIONS Despite evidence of inflammation in this study, muscle antioxidant status was not compromised, and malondialdehyde was unaltered in muscle and plasma. Therefore, this investigation provides no evidence that chronic muscle inflammation compromises antioxidant status or increases lipid peroxidation.

ACKNOWLEDGMENTS We gratefully acknowledge technical guidance from Carol Rea, Helen Tomason and Gary Thorpe of the Wolfson Applied Technology Group, Queen Elizabeth Hospital, Birmingham, U.K. # 1999 The Biochemical Society and the Medical Research Society

113

114

R. Child and others

REFERENCES 1 Takala, T., Vuori, J., Anttinen, H., Vaananen, K. and Myllyla, R. (1986) Prolonged exercise causes an increase in the activity of galactosylhydroxylysyl glucosyltransferase and in the concentration of type III procollagen aminopropeptide in human serum. Pflugers Arch. 407, 500–503 2 Stauber, W. T., Clarkson, P. M., Fritz, V. K. and Evans, W. J. (1990) Extracellular matrix disruption and pain after eccentric muscle action. J. Appl. Physiol. 69, 868–874 3 Hikida, R. S., Staron, R. S., Hagerman, F. C., Sherman, W. and Costill, D. L. (1983) Muscle fibre necrosis associated with human marathon runners. J. Neurol. Sci. 59, 185–203 4 Friden, J., Sjostrom, M. and Ekblom, B. (1983) Myofibrillar damage following intense eccentric exercise in man. Int. J. Sports Med. 4, 170–176 5 Tidball, J. G., Salem, G. and Zernicke, R. (1993) Site and mechanical conditions of failure of skeletal muscle in experimental strain injuries J. Appl. Physiol. 74, 1280–1286 6 Schwane, J. A., Johnson, S. R., Vandenakker, C. B. and Armstrong, R. B. (1983) Delayed onset muscular soreness and plasma CPK and LDH activities after downhill running. Med. Sci. Sports Exercise 15, 51–56 7 Jones, D. A., Newham, D. J., Round, J. M. and Tolfree, S. E. (1986) Experimental human muscle damage : morphological changes in relation to other indices of damage. J. Physiol. 375, 435–448 8 Brown, S., Child, R. B., Donnelly, A., Saxton, J. and Day, S. (1996) Changes in human skeletal muscle contractile function following stimulated eccentric exercise. Eur. J. Appl. Physiol. 72, 515–521 9 Faulkner, J. A., Brooks, S. V. and Opiteck, J. A. (1993) Injury to skeletal muscle fibres during contractions : conditions of occurrence and prevention. Physical Ther. 73, 911–921 10 McCully, K. K. and Faulkner, J. A. (1986) Characteristics of lengthening contractions associated with injury to skeletal muscle fibres. J. Appl. Physiol. 61, 293–299 11 Armstrong, R. B., Warren, G. L. and Warren, J. A. (1991) Mechanisms of exercise-induced muscle fibre injury. Sports Med. 12, 184–207 12 Friden, J. and Lieber, R. L. (1992) Structural and mechanical basis of exercise-induced muscle injury. Med. Sci. Sports Exercise 24, 521–530 13 Warren, J. A., Jenkins, R. R., Packer, L., Witt, E. H. and Armstrong, R. B. (1992) Elevated muscle vitamin E does not attenuate eccentric exercise-induced muscle injury. J. Appl. Physiol. 72, 2168–2175 14 Duncan, C. and Jackson, M. J. (1987) Different mechanisms mediate structural changes and intracellular enzyme efflux following damage to skeletal muscle. J. Cell Sci. 87, 183–188 15 Duarte, J. A., Carvalho, F., Bastos, M. L., Soares, J. M. and Appell, H.-J. (1994) Do invading leucocytes contribute to the decrease in glutathione concentrations indicating oxidative stress in exercised muscle, or are they important for its recovery ? Eur. J. Appl. Physiol. 68, 48–53 16 Zerba, E., Komorowski, T. E. and Faulkner, J. A. (1990) Free radical injury to skeletal muscles of young, adult, and old mice. Am. J. Physiol. 258, C429–C435 17 Kramer, J. H., Mak, I. T. and Wegliki, W. B. (1984) Differential sensitivity of canine cardiac and microsomal enzymes to inhibition by free radical-induced lipid peroxidation. Circ. Res. 55, 120–124 18 Halliwell, B. and Chirico, S. (1993) Lipid peroxidation : its mechanism, measurement and significance. Am. J. Clin. Nutr. 57 (Suppl.), 715S–725S 19 Cannon, J. G., Orencole, S. F., Fielding, R. A. et al. (1990) Acute phase response in exercise : interaction of age and vitamin E on neutrophils and muscle enzyme release. Am. J. Physiol. 28, R1214–R1219 20 Salminen, A. and Vihko, V. (1983) Lipid peroxidation in exercise myopathy. Exp. Mol. Pathol. 38, 380–388 21 Burr, I. M., Asayama, K. and Fenichel, G. M. (1987) Superoxide dismutases, glutathione peroxidase, and catalase in neuromuscular disease. Muscle Nerve 10, 150–154


22 Asayama, K, Hayashibe, H., Dobashi, K. and Kato, K. (1989) Lipid peroxide and antioxidant enzymes in muscle and non-muscle of dystrophic mouse. Muscle Nerve 12, 742–748 23 Mulholland, C. W. and Strain, J. J. (1991) Serum total free radical trapping ability in acute myocardial infarction. Clin. Biochem. 24, 437–441 24 Whitehead, T. P., Thorpe, G. H. G. and Maxwell, S. R. J. (1992) An enhanced chemiluminescent assay for antioxidant capacity in biological fluids. Anal. Chim. Acta 266, 265–277 25 Vendetti, P., Di Meo, S., de Martino Rosaroll, P. and De Leo, T. (1995) Determination by enhanced luminescence technique of liver antioxidant capacity. Arch. Physiol. Biochem. 4, 484–491 26 Saxton, J. M., Donnelly, A. E. and Roper, H. P. (1994) Indices of free-radical mediated damage following maximum voluntary eccentric and concentric muscular work. Eur. J. Appl. Physiol. 68, 189–193 27 Child, R. B., Saxton, J. M. and Donnelly, A. E. (1998) Comparison of eccentric knee extensor muscle actions at two muscle lengths on indices of damage and anglespecific force production in humans. J. Sports Sci. 16, 301–308 28 Komulainen, J., Pitkanen, R. and Vihko, V. (1993) Muscle water content and exercise-induced damage in mice after submaximal running. Med. Sci. Res. 21, 111–113 29 Tullson, P. and Armstrong, R. B. (1981) Muscle hexose monophosphate shunt activity following exercise. Experientia 37, 1311–1312 30 Kretzschmar, M. and Muller, D. (1993) Aging training and exercise : a review of effects on plasma glutathione and lipid peroxides. Sports Med. 15, 196–209 31 Rajguru, S. U., Yeargans, G. S. and Seidler, N. W. (1994) Exercise causes oxidative damage to rat skeletal muscle microsomes while increasing cellular sulfhydryls. Life Sci. 54, 149–157 32 Young, I. S. and Trimble, E. R. (1991) Measurement of malondialdehyde in plasma by high performance liquid chromatography with fluorimetric detection. Ann. Clin. Biochem. 28, 504–508 33 Child, R. B., Wilkinson, D. M., Fallowfield, J. L. and Donnelly A. E. (1998) Changes in serum antioxidant capacity and plasma malondialdehyde concentration in response to a simulated half marathon run. Med. Sci. Sports Exercise, in the press 34 Knight, J. A., Pieper, R. K. and McClellan, L. (1988) Specificity of the thiobarbituric acid reaction : its use in studies of lipid peroxidation. Clin. Chem. 34, 2433–2438 35 Aharata, K. and Engel, A. G. (1984) Monoclonal antibody analysis of mononuclear cells in myopathies I. Quantification of subjects according to diagnosis. Ann. Neurol. 16, 193–208 36 Komulainen, J. and Vihko, V. (1994) Exercise-induced necrotic muscle damage and enzyme release in the four days following prolonged submaximal running in rats. Pflu$ gers Arch. 428, 346–351 37 Vihko, V. and Salminen, A. (1986) Propagation and repair of exercise induced skeletal fibre injury. In Biochemical Aspects of Physical Exercise (Benzi, G., Packer, L. and Silibandi, N., eds.), pp. 337–346, Elsevier, Amsterdam 38 Salminen, A. and Kihlstrom, M. (1985) Lysosomal changes in mouse skeletal muscles during repair of exercise injuries. Muscle Nerve 8, 269–279 39 Vihko, V., Rantamaki, J. and Salminen, A. (1978) Exhaustive physical exercise and acid hydrolase activity in mouse skeletal muscle. Histochemistry 57, 237–249 40 Shannon, A. D., Adams, E. P. and Courtice, F. C. (1974) The lysosomal enzymes acid phosphatase and βglucuronidase in muscle following a period of ischaemia. Aust. J. Exp. Biol. Med. Sci. 52, 157–171 41 Shannon, A. D. and Courtice, F. C. (1975) The lysosomal enzyme N-acetyl-β-glucosaminidase in rabbit muscle following a period of ischaemia. Pathology 7, 25–33 42 Round, J. M., Jones, D. A. and Cambridge, G. (1987)


43 44 45 46

47

Cellular infiltrates in human skeletal muscle : exercise induced damage as a model for inflammatory muscle disease ? J. Neurol. Sci. 82, 1–11 Henson, P. M. and Johnston, R. B. (1987) Tissue injury in inflammation. J. Clin. Invest. 79, 669–674 Weiss, S. J. (1989) Tissue destruction by neutrophils. N. Engl. J. Med. 320, 365–376 Knochel, J. P. (1982) Rhabdomyolysis and myoglobinuria. Annu. Rev. Med. 33, 435–443 Allen, R. G. (1991) Oxygen-reactive species and antioxidant responses during development : the metabolic paradox of cellular differentiation. Proc. Soc. Exp. Biol. Med. 196, 117–129 Maulik, N., Wantanabe, M., Engelman, D. et al. (1995) Myocardial adaptation to ischemia by oxidative stress induced by endotoxin. Am. J. Physiol. 269, C907–C916

48 Mbemba, F. A., Houbion, A., Raes, M. and Remacle, J. (1985) Subcellular localisation and modification with aging of glutathione, glutathione peroxidase and glutathione reductase activities in human fibroblasts. Biochem. Biophys. Acta 838, 211–220 49 Ji, L. L. (1995) Exercise and oxidative stress : role of the cellular antioxidant systems. Exercise Sports Sci. Rev. 23, 135–166 50 Wu, Y., Taylor, M. and Sun, F. (1996) Alterations in reactive oxygen, pH, and calcium in astrocytoma cells during lethal injury. Am. J. Physiol. 270, C115–C124 51 Fielding, R. A., Manfredi, T. J., Ding, W., Fiatarone, M. A., Evans, W. J. and Cannon, J. G. (1993) Acute phase response in exercise III. Neutrophil and IL-1β accumulation in skeletal muscle. Am J. Physiol. 265, R166–R172 Received 6 April 1998/20 July 1998; accepted 20 July 1998


115