Acta Physiol Scand 1983, 117: 109-1 13

Endurance training reduces the susceptibility of mouse skeletal muscle to lipid peroxidation in vitro

A. SALMINEN and V. VIHKO Division of Muscle Research, Department of Cell Biology, University of Jyvaskyla, SF-40100 Jyvaskyla 10, Finland

SALMINEN, A. & VIHKO, V.: Endurance training reduces the susceptibility of mouse skeletal muscle to lipid peroxidation in vitro. Acta Physiol Scand 1983, 117: 109-113. Received 30 May 1982. ISSN 0001-6772. Division of Muscle Research, Department of Cell Biology, University of Jyvaskyla, Finland.

Selected estimates of the lipid peroxidative capacity were assayed in the red and white skeletal muscles of control and endurance-trained mice. Endurance training decreased the lipid peroxidation rate in vitro in both muscle types. The concentration of lipids susceptible to Fez+-induced lipid peroxidation was greater in the red than in the white skeletal muscle and increased after endurance training in the red muscle. Endurance training, however, decreased highly significantly the sensitivity of red muscle to in vitro stimulated lipid peroxidation. The activity of catalase and the concentration of vitamin E were consider- ably higher in the red muscle, whereas the activity of glutathione peroxidase was slightly higher in the white muscle. Endurance training caused no changes in these antioxidants. Endurance training increased the concentrations of reduced and total non-protein glutath- ione in the red skeletal muscle but not in the white muscle. The total sulfhydryl group contents were unaffected. Our results suggest that endurance training may increase the resistance of skeletal muscle to injuries caused by lipid peroxidation.

Key words: Endurance training, muscles, lipid peroxides, autoxidation, antioxidants

The oxygen uptake of skeletal muscles may in- crease up to 1W200 times during strenuous exer- tion (see Keul et al. 1972). It is well documented that mitochondria1 respiration generates in vitro, and probably also in vivo, free oxygen radicals as a byproduct of oxidative metabolism (Chance et al. 1979). Oxygen radicals, especially hydroxyl radi- cals, may initiate lipid peroxidation and thus cause cell injuries (e.g. Del Maestro 1980). Aerobic cells, such as muscle fibers, have enzymatic and non- enzymatic scavenger systems against oxygen stress (Chance et al. 1979, Del Maestro 1980).

Endurance training induces energy metabolic ad- aptations in skeletal muscle (Holloszy 1975). These include e.g. increases in the size and number of mitochondria. Furthermore, endurance training produces a resistance to exertion-induced myo- pathy (Vihko et al. 1979). The purpose of this study was to determine whether endurance training af- fects the susceptibility of white and red skeletal muscles to lipid peroxidation in vitro. Certain an- tioxidants were also assayed.

METHODS Endurance training Male NMRI-mice, aged 5 months, were trained on a motor-driven treadmill with 6" uphill tracks. The animals were made to run 1 h/day for 5 days a week over a period of 3 weeks. During the first week the running speed was increased to 25 m/min. The speed was set at 25 mimin during the 2nd week and at 28 m h i n during the 3rd week. The mice were killed by cervical dislocation on the day following the last exercise. The animal care of both trained and control mice was as described earlier (Vihko et al. 1978). The control mice (n= 15) weighed 39.6T 1.1 g (TSE) and the trained mice (n=15) 36.4k0.7 g. This dif- ference is statistically almost significant (p<O.OS).

Tissue preparation and assay methods The skeletal muscle samples, both red and white muscle types, were excised from the quadriceps femoris muscle. The red muscle sample was composed of the red parts of the proximal heads of the vastus lateralis, vastus medialis and rectus femoris muscles and the red fibers of vastus intermedius. The white muscle sample consisted of the distal and superficial parts of the vastus lateralis muscle. The red and white muscle samples were not exclusively composed of red or white muscle fibers, but were as completely red or white as possible within the limitations

Acta Phys id Scand 117

I 10 A. Salminen and V . Vihko

Table I . EjJects of endurance training on lipid peroxidative capucitj in bc'hite and red skeletal muscles of mice Peroxidation rate is expressed as rnalondialdehyde production ( mol MDAxkg-' muscle) during a 30 min incubation. Fe'+-activated total autoxidation is given as pmol MDAx kg- muscle. Sensitivity to Fe*+-autoxidation is given as per cent autoxidation products in 8 min compared to the total (60 min) value. Catalase activity is given as arbitrary units x g I muscle (see methods) and glutathione peroxidase activity as p o l oxidized NADPH x s - ' x kg-' supernatant protein. Vitamin E concentration is given as mg a-tocopherol x kg-' muscle. fl-Glucuronidase activity is expressed as iimol reaction products X s - ' x kg-l muscle. Citrate synthase and malate dehydrogenase activities are given as mmol x s - I x kg- ' muscle. Protein content is given as g protein x kg-l muscle. Values are means k SE

I"

Variable

Control mice Trained mice

White Red White Red ( n = 15) (n=15) (n= 1.5) (n=15)

Peroxidation rate Fe'+-activated autoxidation

Total (60 min) Sensitivity (8 mini60 min)

Catalase GSH-peroxidase Vitamin E 13-Glucuronidase Citrate slnthase Malate dehydrogenase Protein content

238i15

2 510k50 75.0i5.9 0.55 f0.03 369+15

4.32f0.36 0. I 1 t O . O 1 0.32i0.01 2.98t0.13 178f3

201t8h

3 200FSO' 60.7 26.3 1.83f0.06' 3505 17"

8.57f0.47' 0.16i0.01 0.95 f0.02 ' 5.!BiO.I3' 17912

188+6'

2 470k40 71.8i4.5 0.55 i0 .03 371k15

4.84k0.51 0.13k0.0Id 0.42 k O . O If 3.15k0.13 17752

l55fV

3 470f60' 3 1 .0k2 .9 1.87+0.06 3545 I 1

9.44k0.39

1 .I4 f 0.0 Is 6.7620.1p

0.2 I +0.01"

178f2

"p<O.05. ' p < 0 . 0 1 . p<0.001 (white vs. red muscle). dp<0 .05 . 'pc0.01, 'p<0.001 (control vs. trained muscle)

of the preparation procedure. The samples for the fluoro- metric assays (vitamin E and glutathione) were excised from the right thigh and those for the other assays from the left thigh. The red muscle samples weighed on average 70.7t 1.4 mg and the white samples 73.0k I .4 mg. There were no statistically significant differences between the control and the trained mice.

The muscle samples were quickly prepared. weighed. deep-frozen. and transferred to -80°C until being ana- lyzed within 3 weeks. The muscle samples for the fluoro- metric measurements were homogenized in ice-cold 0.25 M sodium phosphate buffer. pH 8.0, using an all-glass Potter-Elvehjem homogenizer. Homogenates were made to 4 % (wtivol). Immediately after the homogenization the reduced (GSH) and oxidized (GSSG) glutathione were measured with the o-phthalaldehyde method as described by Hissin & Hilf (1976). Vitamin E concentrations were assayed by the method of Taylor et al. (1976). Vitamin A interference was eliminated by H&O, treatment.

The muscle samples for the other measurements were homogenized (3 %, wtivol) in ice-cold 0.1 M potassium phosphate buffer. pH 7.4, using an all-glass Potter-El- vehjern homogenizer. The rate of lipid peroxidation in vitro was estimated in the homogenates using the method of Placer et al. (1966). The homogenates were incubated for 30 min in 0.2 M tris-maleate buffer, pH 5.9. at 37°C. Fe" -induced lipid peroxidation was measured as de- xribed by Kornbrust & Mavis (1980), using FeSO, (60 pM) and sodium ascorbate ( I .O mM) for the in vitro stimu- lation of lipid peroxidation. The total concentration of peroxidative lipids was measured by assaying thiobarbi- turic acid reactive substances produced in I h in an ice- bath by the Fe"-stimulation of muscle homogenates.

41 I i i P ~ . \ I I I / Si ( i i id 117

The susceptibility of muscle homogenates to lipid peroxi- dation was analyzed by comparing the content of peroxi- dation products after 8 min with the total peroxidation after 1 h .

The activity of catalase was assayed and calculated according to the method of Leighton et al. (1968). The homogenates were incubated at 27°C for 7 min. The activ- i t y of glutathione peroxidase was analyzed by a slight modification of the method of Paglia & Valentine (1967). Before the assay the homogenates were made 0.5 % with respect to Triton X-100 and centrifuged for 10 min at 600 g. The supernatant was used in the glutathione peroxidase assay. The concentration of total sulfhydryl groups was analyzed as described by Sedlak & Lindsay (1968). The activities of citrate synthase (EC 4.1.3.7), malate dehy- drogenase (EC 1 . I . I .37) and fl-glucuronidase (EC 3.2.1.31), and protein contents were analyzed as de- scribed earlier (Vihko et al. 1978). The reaction mixture of (3-glucuronidase was made 0.2 % with respect to Triton X- 100 in order to determine the total activity of this enzyme.

RESULTS

Differences between red and ,c,hite skeletal muscle The energy metabolic type of muscle samples was characterized by the activities of citrate synthase and malate dehydrogenase. The activities of these enzymes were 2-3 times higher in the red skeletal muscle (Table 1 ) . The activity of 6-glucuronidase was also significantly higher in the red muscle. The

Lipid peroxidation in skeletal muscle 1 1 1

Table 2 . Effects of endurance training on sulfiydryl groups in white and red skeletal muscles of mice Glutathione concentrations are expressed as mg x kg-' muscle. Total sulfhydryl groups are given as mmol GSH equivalents x kg-' muscle. Other legends are as in Table 1

Variable

Control mice Trained mice

White Red White Red (n= 15) (n= 15) (n=15) (n= 15)

Glutathione

Oxidized (GSSG) 110f3 10925 11424 109f4 Total 251+6 244 f 5 254f5 262f4' Ratio (GSHESSG) 1.2920.03 1.27f0.05 1.25f0.03 1.43_+0.06' Total -SH groups 15.2+0.5 16.5+0.4" 15.6f0.3 16.8k0.2

Reduced (GSH) 14123 135f2 140f2 153f3d

"p<0.05 (white vs. red muscle). 'p<0.05, cp<O.Ol, p<O.OOl (control vs. trained muscle)

protein contents were very similar in both muscle types.

The rate of lipid peroxidation in vitro was faster in the homogenates of the white muscle (Table 1) . The total concentration of lipids susceptible to Fe*+-induced lipid peroxidation was significantly greater in the red skeletal muscle (Table 1). Howev- er, the sensitivity to Fe*+-induced lipid peroxida- tion was similar in the two muscle types.

Of the antioxidants, the activity of catalase and the concentration of vitamin E were considerably higher in the red skeletal muscle, whereas the activ- ity of glutathione peroxidase was slightly higher in the white musc€e. The concentrations of reduced and oxidized glutathione and the ratio between the reduced and oxidized forms were similar in both skeletal muscles (Table 2 ) . The concentration of total sulfiydry1 groups was slightly greater in the red skeletal muscle.

Effects of endurance training on the peroxidative capa city Endurance training increased the activities of ci- trate synthase, malate dehydrogenase and P-glucur- onidase (Table 1). The changes were greater in the red than in the white muscle type. The protein contents were unaffected by endurance training.

Endurance training decreased considerably the rate of lipid peroxidation in vitro in both muscle types (Table €). Endurance training increased the concentration of peroxidizable lipids in the red muscle but not in the white muscIe (Table 1). Si- multaneously the sensitivity to Fe*+-induced lipid peroxidation decreased in the red skeletal muscle but was unchanged in the white muscle. The esti-

mates of antioxidative capacity, i.e. the activities of catalase and glutathione peroxidase and the con- centration of vitamin E, were unaffected by endur- ance training.

Endurance training increased the concentrations of reduced and total non-protein glutathione in the red muscle type (Table 2 ) . Simultaneously the ratio between reduced and oxidized glutathione in- creased slightly in the red muscle. The concentra- tion of total sulfhydryl groups was unaffected by endurance training in both the red and the white muscle.

DISCUSSION

Lipid peroxidation is induced in vitro when tissue homogenates and cell fractions are incubated under aerobic conditions (e.g. Placer et al. 1966, Korn- brust & Mavis 1980). Polyunsaturated fatty acids, e.g. in membrane phospholipids, are very sensitive to lipid peroxidation. Peroxide formation is in- creased by a variety of catalytic components (Hul- tin 1980, Kornbrust & Mavis 1980, Takayanagi et al. 1980). Ferrous iron, in the presence of reducing agents such as ascorbic acid, is the most potent catalyst of non-enzymic peroxidation. Ferrous ions are also essential in the enzymic, NADH- or NADPH-dependent, lipid peroxidation in heart mi- tochondria (Takayanagi et al. 1980) and skeletal muscle microsomes (Hultin 1980).

The concentration of peroxidizable fatty acids in tissues and cell fractions can be measured by carry- ing the ascorbateliron-induced peroxidation to com- pletion (Kornbrust & Mavis 1980). Our study indi- cated that red skeletal muscle contains consider-

Acta Physiol Scand I17

11 2 A . Salminen and V . Vihko

ably more peroxidizable lipids than the white mus- cle type. This is probably due to the higher mito- chondrial content in red skeletal muscle, because mitochondria are a rich source of phospholipids containing a high proportion of unsaturated fatty acids (Fiehn et al. 1971). This conclusion is sup- ported by the observation that endurance training increased the concentration of peroxidizable lipids in red muscle simultaneously with the increase in citrate synthase and malate dehydrogenase activi- ties. Several studies have indicated that endurance training increases the mitochondrial content of skeletal muscles (see Holloszy 1975). The increase in oxidative capacity was greater in red than in white skeletal muscle. The concentration of peroxi- dizable lipids was unchanged in white muscle.

The susceptibility of skeletal muscle homog- enates to lipid peroxidation was decreased in trained mice in spite of the increased concentration of peroxidizable lipids. This interesting change oc- curred in the rates of both the Fe2+-induced perox- idation and the unstimulated peroxidation. The de- fence system against free radical attack is very versatile in aerobic cells (e.g. Chance et al. 1979, Del Maestro 1980). The ascorbateiiron-induced per- oxidation, carried out in an ice-bath, is considered to be non-enzymatic. The changes in enzymic sys- tems both in the induction of and the protection against peroxidation are therefore unlikely to be the cause of the observed decreased sensitivity to lipid peroxidation. Kornbrust & Mavis (1980) observed that the content of vitamin E in microsomal mem- branes modulates the susceptibility of these mem- branes to lipid peroxidation. The concentration of vitamin E was twice as high in the red as in the white muscle. However, statistically significant changes between the two muscle types were not observed in the susceptibility to Fe?"-induced per- oxidation. Endurance training also decreased the susceptibility to lipid peroxidation without affecting vitamin E concentrations. There may exist other structural antioxidants which protect the polyunsat- urated lipids of membranes against peroxidation. The fatty acid composition and compartmentalisa- tion of membrane lipids perhaps also modulate their sensitivity to lipid peroxidation.

Endurance training also decreased the rate of lipid peroxidation in muscle homogenates incubat- ed without any exogenous stimulant. The peroxida- tion rate was slightly faster in white muscle. The origin of this peroxidation in vitro, as well as of the

training-induced protection, is unknown. The reac- tion was carried out at 37°C. The enzymic microso- ma1 (Hultin 1980) and mitochondrial (Takayanagi et al. 1980) peroxidation systems may therefore be activated simultaneously with the non-enzymic re- actions. Endurance training may impart protection against lipid peroxidation by reducing the initiation of peroxidation or by strengthening scavenger sys- tems. Endurance training produced no changes in the activities of the essential enzymatic scavengers, catalase and glutathione peroxidase, or in the con- centration of vitamin E. The concentration of re- duced glutathione, however, increased in the red but not in the white skeletal muscle. This change could improve the functioning of the glutathione redox system but is unlikely to be the cause of the induced resistance to lipid peroxidation.

It must be borne in mind that we used crude muscle homogenates and measured lipid peroxida- tion with the thiobarbituric acid method. Thus, for instance, the value of malondialdehyde as an indi- cator of lipid peroxidation in tissues may be limited because malondialdehyde is a secondary product of lipid peroxidation. The decomposition reactions of lipid hydroperoxides are very complex in biological materials. Our results suggest that endurance train- ing activates the mechanisms responsible for the protection against destructive peroxidations. Noh1 et al. (1981) observed that the activities of protec- tive enzymes, especially of catalase and glutathione peroxidase, increased in the cardiac mitochondria of rats exposed to prolonged hyperbaric oxygen stress. We could not, however, detect similar changes in the skeletal muscles of endurance trained mice.

This study was supported by the Academy of Finland and the Research Council for Physical Education and Sport (Ministry of Education, Finland). We thank Mrs Irene Helkala and Mr Matti Virtanen for skillful technical assist- ance.

REFERENCES CHANCE. B . , SIES, H. & BOVERIS, A. 1979. Hydro-

peroxide metabolism in mammalian organs. Physiol Rev 59: 527-605.

DEL MAESTRO, R. F. 1980. An approach to free radi- cals in medicine and biology. Acta Physiol Scand Suppl. 492: 153-168.

ELEPANO, M. 1971. Lipids and fatty acids of sarco- lemma, sarcoplasmic reticulum, and mitochondria from rat skeletal muscle. J Biol Chem 246: 5617-5620.

FIEHN, W . & PETER, J . B., MEAD, J . F. & GAN-

Lipid peroxidation in skeletal muscle 1 13

HISSIN, P. J . & HILF, R. 1976. A fluorometric method for determination of oxidized and reduced glutathione in tissues. Anal Biochem 74: 216226.

HOLLOSZY, J. 0. 1975. Adaptation of skeletal muscle to endurance exercise. Med Sci Sports 7: 155-164.

HULTIN, H. 0. 1980. Enzyme-catalyzed lipid oxidation in muscle microsomes. In: Autoxidation in food and biological systems (ed. M. G . Simic & M. Karel), pp. 505-527. Plenum Press, New York.

KEUL, J. , DOLL, E. & KEPPLER, D. 1972. Energy metabolism of human muscle. S. Karger, Basel.

KORNBRUST, D. J. & MAVIS, R. D. 1980. Relative susceptibility of microsomes from lung, heart, liver, kidney, brain and testes to lipid peroxidation: correla- tion with vitamin E content. Lipids IS: 315-322.

HUIN, P., COFFEY, J. W., FOWLER, S. & DE DUVE, C. 1968. The large-scale separation of peroxi- somes, mitochondria and lysosomes from the liver of rats injected with Triton WR-1339. J Cell Biol

NOHL, H., HEGNER, D. & SUMMER, K.-H. 1981. The mechanism of toxic action of hyperbaric oxygenation on the mitochondria of rat-heart cells. Biochem Phar- macol. 30: 1753-1757.

PAGLIA, D. E. & VALENTINE, W. N. 1967. Studies on the quantitative and qualitative characterization of erythrocyte glutathione peroxidase. J Lab Clin Med

LEIGHTON, F., POOLE, B., BEAUFAY, H., BAUD-

37: 482-513.

70: 158-169.

PLACER, Z. A., CUSHMAN, L. L. &JOHNSON, B. C. 1966. Estimation of product pf lipid peroxidation (ma- lonyl dialdehyde) in biochemical systems. Anal Bio- chem 16: 359-364.

SEDLAK, J . & LINDSAY, R. H. 1968. Estimation of total protein-bound, and nonprotein sulfhydryl groups in tissue with Ellman's reagent. Anal Biochem

TAKAYANAGI, R., TAKESHIGE, K. MINAKAMI, S. 1980. NADH- and NADPH-dependent lipid peroxida- tion in bovine heart submitochondrial particles. De- pendence on the rate of electron flow in the respiratory chain and an antioxidant role of ubiquinol. Biochem J 192: 853-860.

TAYLOR, S. L., LAMDEN, M. P. & TAPPEL, A. L. 1976. Sensitive fluorometric method for tissue toco- pherol analysis. Lipids 1 1 : 530-538.

VIHKO, V., SALMINEN, A. & RANTAMAKI, J. 1978. Oxidative and lysosomal capacity in skeletal muscle of mice after endurance training of different intensities. Acta Physiol Scand 104: 74-81.

VIHKO, V., SALMINEN, A. & RANTAMAKI, J. 1979. Exhaustive exercise, endurance training, and acid hy- drolase activity in skeletal muscle. J Appl Physiol

25: 192-205.

47: 43-50.

8-835883 Acta Physiol Scund 117