- Email: [email protected]
Exercise causes oxidative damage to rat skeletal muscle microsomes while increasing cellular sulfhydryls
Life Sciences,Vol. 54, pp. 149-157 Printed in the USA
Pergamon Press
EXERCISE CAUSES OXIDATIVE DAMAGE TO RAT SKELETAL MUSCLE MICROSOMES WHILE INCREASING CELLULAR SULFHYDRYLS
Shailesh U. Rajguru, George S. Yeargans and Norbert W. Seidler Department of Biochemistry, University of Health Sciences, College of Osteopathic Medicine, 2105 Independence Blvd., Kansas City, MO 64124 (Received in final form October 28, 1993) Summary The physiological and biochemical demands on contracting muscle make this tissue particularly susceptible to molecular and cellular damage. We looked at membrane structures in cardiac and skeletal muscle and in erythrocytes for exercise-induced lipid peroxidation. These tissues were removed from each of the rats used in this study. We also examined and compared the effects of exercise on the redox status of blood plasma, erythrocytes and cardiac and skeletal muscle from the same rats. We used a swim stress protocol to exercise the rats to exhaustion. Some form of chemical modification or oxidative damage to membranes was observed in all of the tissues tested. Cardiac muscle microsomes from exercised rats exhibited increased malondialdehyde and decreased phospholipid (control, 249.1 vs exercised, 120.6 nmols phospholipid/mg protein). Skeletal muscle microsomes showed decreased sulfhydryls, decreased phospholipid (control, 1,276.9 vs exercised, 137.7 nmols phospholipid/mg protein), increased malondialdehyde and greater protein crosslinking after exercise. Erythrocyte membranes also exhibited exercised-induced protein oxidation. However, the total cellular sulfhydryl content remained the same in erythrocytes and cardiac tissue but increased in blood plasma (control, 10.8 vs exercised, 24.7/~mols SH/dl plasma) and skeletal muscle after exercise. We conclude that exercise profoundly effects membrane structures. The body compensates for this lipid peroxidation and protein damage by increasing total cellular sulfhydryls in blood plasma and skeletal muscle which would aid in repair of the damaged membranes. Exercise is an important lifestyle change which may bring about significant benefits in health and quality of life. The biochemical basis for the improvement of health is poorly understood. There is an apparent paradox in the study of exercise. Exercise promotes increased cardiovascular endurance, strengthens the heart and lungs, decreases blood pressure, and reduces stress; however, exercise causes temporary muscle damage(l) as determined by pain sensation and serum creatine kinase activity. Additionally, there is a decrease in the mitochondrial redox status with an apparent down regulation of antioxidant enzymes after Corresponding Author: Dr. N.W. Seidler 0024-3205/94 $6.00 + .00 Copyright © 1993Pergamon Press Ltd All rights reserved.
150
Exercise-inducedOxidativeDamage
Vol. 54, No. 3, 1994
exercise (2,3). To better understand this paradox we examined the membranes from various tissues and measured the redox status of these tissues. We harvested the tissues precisely 30 minutes after a single bout of exhaustive exercise of untrained rats. The observations presented here may not pertain to the chronic effects of training, and the considerations addressed are likely more relevant to individuals who sporadically exercise than to athletes. We compared specific chemical modifications in the whole cell homogenate and microsomal fractions of cardiac and skeletal muscle, in red cell lysates and membranes and in blood plasma obtained from exercised and control rats. In light of our findings we discuss the significance of cellular defense and repair in exercise. Methods Exercise Protocol. Male Sprague-Dawley rats used in these experiments were 49-56 days old unless otherwise indicated. The rats were divided into experimental and control groups. The experimental rats were exercised using a swim stress protocol(4). A soft lead wire (1% of the rat's body weight) was attached to the rat's tail before placing the rat into a tank of pure water at room temperature and allowing the rat to swim to exhaustion (typically 10-15 minutes). The control rats were also subjected to the lead weight and placed in the water tank but they were removed immediately and not exercised. The animals were not subjected to water prior to the day of the study. Deionized water was stored in sealed containers at room temperature overnight in a thermostatically controlled room. The water temperature was kept at 23-25°C. For each series of experiments exercised and control rats were paired, and their tissues were processed in parallel. Precisely 30 minutes following treatment the rats were sacrificed and their tissues harvested. Tissue Samples. Blood was drawn into heparinized syringes from the dorsal artery of anesthetized rats (50mg sodium pentobarbital per kg body weight). The tissues were then excised from the same rats after being euthanized while under anesthesia. The hearts were removed and the predominantly white muscle fibers of the upper hind limb were excised. The tissues were immediately processed and all steps were performed at 4 ° C. The subcellular fractionation and isolation of the samples as described below was completed in approximately 6 hours. Samples were quick-frozen in ethanol and solid carbon dioxide and stored at -20° C. All assays were performed within three months of their isolation. Skeletal and Cardiac Muscle Homogenates. In this procedure(5) the tissues were homogenized in a medium containing 0.05 M KC1, 10 mM Tris-HC1, pH 7.4, in a blender. After centrifugation at 5,900 x g (10 min) to sediment the extracellular material, the supernatant was passed through cheese cloth to further remove connective tissue remnants. The remaining whole cell homogenate was quickly frozen and stored in small aliquots at minus 20 ° C until assayed. Some of the homogenate was used to obtain microsomes as described below. Skeletal and Cardiac Muscle Microsomes. Isolation of microsomes involved differential centrifugation. The whole cell homogenate was centrifuged at 8,700 x g (20 min) to remove mitochondria. The supernatant was passed through cheese cloth and glass wool and then centrifuged at 45,900 x g (30 min). The sediment was resuspended in 0.6 M KC1, 10 mM TrisHC1, p H 7.4 to extract the contractile proteins and centrifuged at 45,900 x g (30 min). The pellet was then resuspended in 0.3 M sucrose, 20 mM Tris-maleate, pH 7.4 (storage medium) and centrifuged at 49,000 x g (45 min). The final pellet was resuspended in the storage medium. Aliquots were quickly frozen and kept at minus 20 ° C until assayed.
Vol. 54, No. 3, 1994
Exerclse-induced OxidativeDamage
151
Assays. Samples were tested for sulfhydryl content by a procedure(6) using dithionitrobenzene (DTNB). The sample was added to a denaturing solution containing sodium dodecyl sulfate to solubilize proteins in order to expose hidden sulfhydryls. The medium was slightly alkaline to ionize the sulfhydryl groups in order to make them more reactive to DTNB. Sample controls (minus DTNB) were used to subtract background absorbances. The malondialdehyde content of the samples was determined using a thiobarbituric acid-reactivity assay(7). The samples were deproteinated by adding trichloroacetic acid, heating and centrifuging to remove precipitated protein. Phospholipid quantitation involved determining the amount of total phosphorus using an ash method(8). Protein concentration was determined using the Biuret method(9). All observations were averaged and statistically analyzed using Student's t-tests. Significance was evaluated by inspecting the t-values using a two-tailed criterion and accepted at the 95 % confidence level. Protein Electrophoresis. Electrophoresis was performed according to Laemmli(10). Samples were denatured by heating in sodium dodecylsulfate (SDS) and 13-mercaptoethanol prior to electrophoresis. The protein bands were stained with Coomassie Brilliant Blue R250. The slab gels consisted of 5-20% polyacrylamide gradients. The stacking gel above the resolving gel had a 2.5% polyacrylamide concentration. Results Cardiac Muscle. There was no difference in size, weight or appearance of the cardiac tissue from the control and exercised rats. The total protein yield in homogenate (whole cell) and rnicrosomal (internal membrane) fractions was similar in the two groups. The amount of phospholipid in the microsomes, however, decreased significantly after exercise as indicated in Table I. The loss of membrane phospholipid is consistent with an observed increase in malondialdehyde which is shown in Figure 1. Malondialdehyde is a breakdown product of fatty acids that have undergone spontaneous peroxidation. Malondialdehyde is one of the most stable lipid peroxidation products. Its appearance is dependent on lipid decomposition, and its disappearance is due to spontaneous reactivity with primary amines which is negligible during storage at -20° C. TABLE I Effect of exercise on phospholipid to protein ratios in cardiac and skeletal muscle microsomes Phospholipid/Protein (nmols PL/mg prot) Group a
Cardiac Muscle
Skeletal Muscle
Control
249.1(1)
1,276.9 _ 526.7(2)
Exercised 120.6 _+ 41.9(7)b 137.7 -- 29.3(5)c The data are presented as mean _+ standard error of the mean. a The number of rats is given in parentheses. b p < 0.05; c P < 0.01 There was no detectable difference in protein composition of the cardiac microsomes from control and exercised rats as determined by polyacrylamide gel electrophoresis performed under reducing and denaturing conditions (data not shown). As shown in Figure
152
Exercise-induced Oxidative Damage
Vol. 54, No. 3, 1994
A
.-I I1. ¢0 "o
,-o G) -o.9.o -°= e8 s
=E_m O
E eCardiac
Muscle
Skeletal Muscle
Erythrocytes
FIG. 1 Levels of malondialdehyde in control and exercised tissues. Values were based on determinations using microsomes from cardiac and skeletal muscle and plasma membranes from erythrocytes. The bars represent averages from multiple assays of tissues from three control rats (open bars) and from six exercised rats (shaded bars). Error lines are the relevant halves of the standard errors of the mean. PL: phospholipid. Asterisks indicate significant differences between control and exercised groups (* P < 0.05; ** P < 0.01). 8O 7O
2o
=.51o 0 Hornogenate$ CARDIAC
Microsomes MUSCLE
Homogenates
Mlcrosomes
SKELETAL
MUSCLE
Hemolysate$
Membranes
ERYTHROCYTES
FIG. 2 Total sulfhydryl content in tissue fractions from control and exercised rats. The bars represent mean values from multiple assays of the indicated tissue samples from three control rats (open bars) and from six exercised rats (shaded bars). Error lines are given as the relevant halves the standard errors of the mean. Asterisks indicate significant differences between control and exercised groups (* P < 0.05; ** P < 0.01).
Vol. 54, No. 3, 1994
Exerdse-indueed Oxidative Damage
153
2 we observed an increase in sulfhydryl content in exercised cardiac microsomes as compared to control cardiac microsomes. This observation is consistent with increased binding of phospholambam, a cysteine-rich regulatory protein. Figure 2 presents the total sulfhydryl content in cardiac homogenates which remained essentially unchanged after exercise. Skeletal Muscle. The appearance of the exercised muscle upon excision was not different from control muscle. The exercised muscles showed a tendency to sporadically contract during surgical removal of the tissue. However, none of the muscles from the control rats exhibited spontaneous twitching during tissue extraction. The concentrations of protein in the control and exercised skeletal muscle homogenates and microsomes were the same. However, we found a ten-fold decrease (Table I) in the phospholipid content of exercised skeletal muscle microsomes when compared with control samples. Skeletal muscle membranes were oxidatively modified. Figure 1 shows that microsomes from exercised skeletal muscle contained higher malondialdehyde levels than microsomes from nonexercised skeletal muscle. Elevated malondialdehyde indicates accelerated lipid peroxidation and a greater susceptibility to oxidative modification due to reaction with proteins forming adducts and crosslinked products. Electrophoretic separation of microsomal proteins presented in Figure 3 indicated distinctive exercise-induced changes in band patterns that reflect increased covalent crosslinking of proteins in the microsomal structures after exercise. We observed a decrease in 43 and 110 kDa proteins together with an increase in protein bands migrating at 145 and 285 kDa. This suggests that the 145 kDa band may represent crosslinking of the 43 and ll0kDa proteins. The 285 kDa band may be dimerization of this 145 kDa crosslinked product. The identity of the 43 kDa protein is as yet unknown. The intensity and position of the 110 kDa band is consistent with the Ca 2+ ATPase of the sarcoplasmic reticulum which is enriched in this fraction. The protein profile of skeletal muscle homogenates confirmed increases in crosslinking and loss of the same proteins after exercise (data not shown). Figure 2 also illustrates the increased protein damage to skeletal muscle microsomes from exercised rats indicated by decreased levels of protein sulfhydryls compared to nonexercised controls. Interestingly, the sulfhydryl content of skeletal muscle homogenates shown in Figure 2 was highest in the exercised samples. The values for the homogenate fractions represent the total cellular (microsomal and cytosolic) sulfhydryl content which suggests a dramatic increase in sulfhydryl compounds in the cytosol. The large increase in cytosolic sulfhydryls in exercised skeletal muscle may indicate activation of a repair mechanism. Blood. As mentioned above, blood was obtained from the dorsal artery which is covered by a layer of connective tissue that had to be peeled away in order to expose the artery. Interestingly, the diameter of the artery from the exercised rat was visibly larger than that of the control rats. However, there was no visible difference between control and exercised rats in the diameter of the vein that runs parallel with this artery. While an objective measurement was not made in this case, the observation raises provoking questions. Experimentation in this direction will likely yield important information. The packed cell volume was the same in exercised and control samples (data not shown). No hemolysis occurred during the isolation of the erythrocytes from either the control or the exercised group. As shown in Figure 1 there was no detectable difference in
154
Exercise-induced Oxidative Damage
Vol. 54, No. 3, 1994
malondialdehyde levels in erythrocyte membranes obtained from control and exercised rats. We have found, however, a decrease (Figure 2) in sulfhydryl content in exercised erythroeyte membranes compared with control samples.
High Molecular Weight Crosslinked Products
ii !! y ~iii
285 kDa
i ii ii!
ill
~ - - 145 kDa
, ~i~!!ii~ill ~ !! iii!!~'!i'i'i'~ii'ii
'! Ca2* - A T P a s e (110 kDa)
i~~il,~, ,~~ii~,
~ - - 43 kDa
I
I =
I
I "I-
SDS polyacrylamide gel electrophoresis of skeletal muscle microsomal proteins. Microsomes were isolated from control (-) and exercised (+) rats that were 268 days old. Arrows indicate major differences.
Vol. 54, No. 3, 1994
Exercise-induced OxidativeDamage
155
Electrophoresis of denatured erythrocyte membrane proteins from the exercised samples revealed the presence of high molecular weight crosslinked products that did not enter the resolving gel (data not shown). This crosslinked protein material was not found in the control samples. TABLE II Exercised-induced change in the level of total sulfhydryl groups in blood plasma
Sulfhydryl Levels Group
N
(nmols SH/mg prot)
Control
3
1.31 _+ 0.08
(~mols SH/dl plasma) 10.8 _ 1.55
Exercised 6 3.17 -+ 0.62 a 24.7 -+ 3.89 a The data are presented as mean _+ standard error of the mean. a p < 0.01 When membrane-free hemolysates were assayed for sulfhydryl content, we observed no change in sulfhydryls between control and exercised samples which is shown in Figure 2. Even when solubilized samples of whole cells were tested, there was no detectable difference in sulfhydryls between the control and exercised groups (data not shown). As presented in Table II, the blood plasma levels of sulfhydryl-containing compounds increased after exercise. Discussion We report changes in the chemical composition of various tissues that occur after exercise. The tissue response represents oxidative and compensatory repair processes. Exercise had its greatest effects on membrane structures. We found oxidized erythrocyte membrane proteins, which supports evidence that exercise causes hemolysis(ll) presumably by affecting membrane integrity. The microsomes from cardiac and skeletal muscle exhibited increased lipid peroxidation products and decreased phospholipid content, and the membrane proteins of skeletal muscle microsomes were oxidized. Sulfhydryl levels in whole erythrocytes, erythrocyte lysates, blood plasma cardiac muscle, homogenates and skeletal muscle homogenates after exercise did not decrease. Our observations that skeletal muscle exhibited higher levels of sulfhydryls after exercise is consistent with evidence that various skeletal muscle transport systems become stimulated upon exercise(12). Increased uptake of cysteine or glutathione would account for the elevation in sulfhydryls observed in skeletal muscle. Increased levels of skeletal muscle sulfhydryls may also be due to activation of pathways for cysteine synthesis. These findings suggest that while there is a down regulation of some antioxidant enzymes(3) a repair process appears to be induced as evidenced by the elevation in skeletal muscle redox status. There was a simultaneous elevation in plasma levels of sulfhydryls which may affect surface membrane transport mechanisms of skeletal muscle. This latter finding is consistent with an observed increase in plasma haptoglobin following exercise(ll) but may also be due to increased cysteine absorption in the gut during the increased blood flow through the splanchnic circulation following exercise.
156
Exerclse-inducedOxidativeDamage
Vol. 54, No. 3, 1994
Since the variation among individuals in the experimental group was generally greater than that for the control group, we examined tissues from a larger number of experimental rats which explains the disparity in animal numbers. We did however observe very little interesse variation and therefore used few animals in some tests such as phospholipid analysis which was performed to confirm lipid decomposition. Modulation of intracellular Ca2+ in muscle cells(13) involves the participation of Ca2+ channels, Na+-Ca2+ exchange systems and ATP-dependent Ca2+ pumps of the sarcolemma and transverse tubules and Ca2+ channels and ATP-dependent Ca 2+ pumps of the sarcoplasmic reticulum. The microsomal preparation that we tested is enriched in sarcoplasmic reticulum. The exercise-induced changes in membrane lipid and crosslinking of the ll0kDa protein which is thought to be the SR Ca2+ pump may greatly effect Ca 2+ homeostasis. Protection and regeneration of lipid-based structures appear to be vital for an exercising individual. While there is evidence that short-term vitamin E supplementation does not decrease lipid peroxide formation(4), vitamin E supplementation has proven beneficial(14). Dietary considerations for athletes may not be the same as that for more those who sporadically exercise. Since our observations are based on effects of acute exercise on untrained animals, the dietary needs for nutrients that protect and regenerate membranes may not be pertinent to the athlete. The sulfhydryl groups are the most reactive of the amino acid side chains. This property makes cysteines important components of catalytic sites. The structure and function of membrane transport systems(15), glycolytic enzymes(16) and contractile proteins(17) are dependent on the redox status of the cell as reflected in the oxidative state of the cysteine residues. Furthermore, non-protein sulfhydryl-containing compounds such as glutathione, which is found at approximately 5mM concentration in the cell, are necessary for the repair of lipid peroxides. It is this fact that has led to the hypothesis that lipid peroxidation accelerates the degenerative processes in certain myopathies(18). In conclusion, damage to subcellular membrane structures was observed in all of the tissues tested. Cardiac and skeletal muscle microsomes from exercised rats exhibited increased lipid peroxidative products and decreased membrane phospholipids. Skeletal muscle microsomes and erythrocyte membranes showed increased protein oxidation and greater protein crosslinking after exercise. The total cellular sulfhydryl content remains the same in cardiac tissue and erythrocytes and is increased in skeletal muscle and blood plasma after exercise. The subcellular membranes in skeletal muscle were chemically modified after exercise and we think that skeletal muscle compensated by increasing total cellular sulfhydryls which would aid in repair (reduced glutathione removes lipid peroxides). The intracellular increase in sulfhydryls in skeletal muscle may be a direct consequence of blood plasma elevations in sulfhydryls. While the increase in sulfhydryls in several tissues correlates with the acute exercise of untrained animals, further research is required to determine the causal relationships. This study suggests that some of the long term health benefits of exercise may be due to increased levels of blood plasma sulfhydryl-containing compounds which may act as detoxifying agents by removing cytotoxic lipid breakdown products. The mechanism for this elevation and the organs involved in regulating blood levels of sulfhydryl-containing compounds are yet to be determined.
Vol. 54, No. 3, 1994
Exercise-inducedOxidativeDamage
157
Acknowledgements The authors wish to thank Kamin McCune and Bill Glosser for their efforts in producing this manuscript and gratefully acknowledge the advice and support of Dr. Douglas Rushing and Dr. Anthony Silvagni. The work reported here was supported by funds from the University of Health Sciences - College of Osteopathic Medicine. References 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18.
P.M. CLARKSON and I. TREMBLAY, J. Appl. Physiol. 65 1-6 (1988). T.E. GRAHAM and B. SALTIN, J. Appl. Physiol. 66 561-566 (1989). L.L. JI, F.W. STRATMEN, and H.A. LARDY, Arch. Biochem. Biophys. 263 137-149 (1988). P.S. BRADY, L.J. BRADY, and D.E. ULLREY, J. Nutr. 109 1103-1109 (1979). L.R. JONES, A.D. WEGENER, and H.K.B. SIMMERMAN, Methods Enzymol. 157 360-369 (1988). A.F.S.A. HABEEB, Methods Enzymol. 2__.5457-464 (1972). J.A. BUEGE and S.D. AUST, Methods Enzymol. 5__22302-310 (1978). G.R. BARTLETI', J. Biol. Chem. 234 466-468 (1959). F. MARCUS, J. RITYENHOUSE, T. CHATTERJEE, and M.M. HOSEY, Methods Enzymol. 9_00352-357 (1982). U.K. LAEMMLI, Nature 227 680-685 (1970). L.M. WEIGHT, M.J. BYRNE, and P. JACOBS, Clin. Science 8._/1147-152 (1991). S. IOTH, R. FUNICELLO, P. ZANIOL and B. BARBIROLI, Biochem. Biophys. Res. Comm. 176 1204-1209 (1991). A. MARTONOSI, Molecular Aspects of Transport Proteins, J.J.H.H.M. De Pont (ed.), 57-116, Amsterdam: Elsevier Science Publishers, (1992). M. MEYDANI, W. EVANS, G. HANDELMAN, R.A. FIELDING, S.N. MEYDANI, M.A. FIATARONE, J.B. BLUMBERG, AND J.G. CANNON, Ann. N.Y. Acad. Sci. 669 363-364 (1992). R.P. HEBBEL, O. SHALEV, W. FOKER, and B.H. RANK, Biochem. Biophys. Acta. 862 8-16 (1986). A.E. BRODIE, and D.J. REED, Arch. Biochem. Biophys. 276 212-218 (1990). J. HIRATSUKA, Biol. Chem. 267 14941-14948 (1992). P.J. RUSSO and N.W. SEIDLER, Med. Hypotheses 3__99147-151 (1992).
Pergamon Press
EXERCISE CAUSES OXIDATIVE DAMAGE TO RAT SKELETAL MUSCLE MICROSOMES WHILE INCREASING CELLULAR SULFHYDRYLS
Shailesh U. Rajguru, George S. Yeargans and Norbert W. Seidler Department of Biochemistry, University of Health Sciences, College of Osteopathic Medicine, 2105 Independence Blvd., Kansas City, MO 64124 (Received in final form October 28, 1993) Summary The physiological and biochemical demands on contracting muscle make this tissue particularly susceptible to molecular and cellular damage. We looked at membrane structures in cardiac and skeletal muscle and in erythrocytes for exercise-induced lipid peroxidation. These tissues were removed from each of the rats used in this study. We also examined and compared the effects of exercise on the redox status of blood plasma, erythrocytes and cardiac and skeletal muscle from the same rats. We used a swim stress protocol to exercise the rats to exhaustion. Some form of chemical modification or oxidative damage to membranes was observed in all of the tissues tested. Cardiac muscle microsomes from exercised rats exhibited increased malondialdehyde and decreased phospholipid (control, 249.1 vs exercised, 120.6 nmols phospholipid/mg protein). Skeletal muscle microsomes showed decreased sulfhydryls, decreased phospholipid (control, 1,276.9 vs exercised, 137.7 nmols phospholipid/mg protein), increased malondialdehyde and greater protein crosslinking after exercise. Erythrocyte membranes also exhibited exercised-induced protein oxidation. However, the total cellular sulfhydryl content remained the same in erythrocytes and cardiac tissue but increased in blood plasma (control, 10.8 vs exercised, 24.7/~mols SH/dl plasma) and skeletal muscle after exercise. We conclude that exercise profoundly effects membrane structures. The body compensates for this lipid peroxidation and protein damage by increasing total cellular sulfhydryls in blood plasma and skeletal muscle which would aid in repair of the damaged membranes. Exercise is an important lifestyle change which may bring about significant benefits in health and quality of life. The biochemical basis for the improvement of health is poorly understood. There is an apparent paradox in the study of exercise. Exercise promotes increased cardiovascular endurance, strengthens the heart and lungs, decreases blood pressure, and reduces stress; however, exercise causes temporary muscle damage(l) as determined by pain sensation and serum creatine kinase activity. Additionally, there is a decrease in the mitochondrial redox status with an apparent down regulation of antioxidant enzymes after Corresponding Author: Dr. N.W. Seidler 0024-3205/94 $6.00 + .00 Copyright © 1993Pergamon Press Ltd All rights reserved.
150
Exercise-inducedOxidativeDamage
Vol. 54, No. 3, 1994
exercise (2,3). To better understand this paradox we examined the membranes from various tissues and measured the redox status of these tissues. We harvested the tissues precisely 30 minutes after a single bout of exhaustive exercise of untrained rats. The observations presented here may not pertain to the chronic effects of training, and the considerations addressed are likely more relevant to individuals who sporadically exercise than to athletes. We compared specific chemical modifications in the whole cell homogenate and microsomal fractions of cardiac and skeletal muscle, in red cell lysates and membranes and in blood plasma obtained from exercised and control rats. In light of our findings we discuss the significance of cellular defense and repair in exercise. Methods Exercise Protocol. Male Sprague-Dawley rats used in these experiments were 49-56 days old unless otherwise indicated. The rats were divided into experimental and control groups. The experimental rats were exercised using a swim stress protocol(4). A soft lead wire (1% of the rat's body weight) was attached to the rat's tail before placing the rat into a tank of pure water at room temperature and allowing the rat to swim to exhaustion (typically 10-15 minutes). The control rats were also subjected to the lead weight and placed in the water tank but they were removed immediately and not exercised. The animals were not subjected to water prior to the day of the study. Deionized water was stored in sealed containers at room temperature overnight in a thermostatically controlled room. The water temperature was kept at 23-25°C. For each series of experiments exercised and control rats were paired, and their tissues were processed in parallel. Precisely 30 minutes following treatment the rats were sacrificed and their tissues harvested. Tissue Samples. Blood was drawn into heparinized syringes from the dorsal artery of anesthetized rats (50mg sodium pentobarbital per kg body weight). The tissues were then excised from the same rats after being euthanized while under anesthesia. The hearts were removed and the predominantly white muscle fibers of the upper hind limb were excised. The tissues were immediately processed and all steps were performed at 4 ° C. The subcellular fractionation and isolation of the samples as described below was completed in approximately 6 hours. Samples were quick-frozen in ethanol and solid carbon dioxide and stored at -20° C. All assays were performed within three months of their isolation. Skeletal and Cardiac Muscle Homogenates. In this procedure(5) the tissues were homogenized in a medium containing 0.05 M KC1, 10 mM Tris-HC1, pH 7.4, in a blender. After centrifugation at 5,900 x g (10 min) to sediment the extracellular material, the supernatant was passed through cheese cloth to further remove connective tissue remnants. The remaining whole cell homogenate was quickly frozen and stored in small aliquots at minus 20 ° C until assayed. Some of the homogenate was used to obtain microsomes as described below. Skeletal and Cardiac Muscle Microsomes. Isolation of microsomes involved differential centrifugation. The whole cell homogenate was centrifuged at 8,700 x g (20 min) to remove mitochondria. The supernatant was passed through cheese cloth and glass wool and then centrifuged at 45,900 x g (30 min). The sediment was resuspended in 0.6 M KC1, 10 mM TrisHC1, p H 7.4 to extract the contractile proteins and centrifuged at 45,900 x g (30 min). The pellet was then resuspended in 0.3 M sucrose, 20 mM Tris-maleate, pH 7.4 (storage medium) and centrifuged at 49,000 x g (45 min). The final pellet was resuspended in the storage medium. Aliquots were quickly frozen and kept at minus 20 ° C until assayed.
Vol. 54, No. 3, 1994
Exerclse-induced OxidativeDamage
151
Assays. Samples were tested for sulfhydryl content by a procedure(6) using dithionitrobenzene (DTNB). The sample was added to a denaturing solution containing sodium dodecyl sulfate to solubilize proteins in order to expose hidden sulfhydryls. The medium was slightly alkaline to ionize the sulfhydryl groups in order to make them more reactive to DTNB. Sample controls (minus DTNB) were used to subtract background absorbances. The malondialdehyde content of the samples was determined using a thiobarbituric acid-reactivity assay(7). The samples were deproteinated by adding trichloroacetic acid, heating and centrifuging to remove precipitated protein. Phospholipid quantitation involved determining the amount of total phosphorus using an ash method(8). Protein concentration was determined using the Biuret method(9). All observations were averaged and statistically analyzed using Student's t-tests. Significance was evaluated by inspecting the t-values using a two-tailed criterion and accepted at the 95 % confidence level. Protein Electrophoresis. Electrophoresis was performed according to Laemmli(10). Samples were denatured by heating in sodium dodecylsulfate (SDS) and 13-mercaptoethanol prior to electrophoresis. The protein bands were stained with Coomassie Brilliant Blue R250. The slab gels consisted of 5-20% polyacrylamide gradients. The stacking gel above the resolving gel had a 2.5% polyacrylamide concentration. Results Cardiac Muscle. There was no difference in size, weight or appearance of the cardiac tissue from the control and exercised rats. The total protein yield in homogenate (whole cell) and rnicrosomal (internal membrane) fractions was similar in the two groups. The amount of phospholipid in the microsomes, however, decreased significantly after exercise as indicated in Table I. The loss of membrane phospholipid is consistent with an observed increase in malondialdehyde which is shown in Figure 1. Malondialdehyde is a breakdown product of fatty acids that have undergone spontaneous peroxidation. Malondialdehyde is one of the most stable lipid peroxidation products. Its appearance is dependent on lipid decomposition, and its disappearance is due to spontaneous reactivity with primary amines which is negligible during storage at -20° C. TABLE I Effect of exercise on phospholipid to protein ratios in cardiac and skeletal muscle microsomes Phospholipid/Protein (nmols PL/mg prot) Group a
Cardiac Muscle
Skeletal Muscle
Control
249.1(1)
1,276.9 _ 526.7(2)
Exercised 120.6 _+ 41.9(7)b 137.7 -- 29.3(5)c The data are presented as mean _+ standard error of the mean. a The number of rats is given in parentheses. b p < 0.05; c P < 0.01 There was no detectable difference in protein composition of the cardiac microsomes from control and exercised rats as determined by polyacrylamide gel electrophoresis performed under reducing and denaturing conditions (data not shown). As shown in Figure
152
Exercise-induced Oxidative Damage
Vol. 54, No. 3, 1994
A
.-I I1. ¢0 "o
,-o G) -o.9.o -°= e8 s
=E_m O
E eCardiac
Muscle
Skeletal Muscle
Erythrocytes
FIG. 1 Levels of malondialdehyde in control and exercised tissues. Values were based on determinations using microsomes from cardiac and skeletal muscle and plasma membranes from erythrocytes. The bars represent averages from multiple assays of tissues from three control rats (open bars) and from six exercised rats (shaded bars). Error lines are the relevant halves of the standard errors of the mean. PL: phospholipid. Asterisks indicate significant differences between control and exercised groups (* P < 0.05; ** P < 0.01). 8O 7O
2o
=.51o 0 Hornogenate$ CARDIAC
Microsomes MUSCLE
Homogenates
Mlcrosomes
SKELETAL
MUSCLE
Hemolysate$
Membranes
ERYTHROCYTES
FIG. 2 Total sulfhydryl content in tissue fractions from control and exercised rats. The bars represent mean values from multiple assays of the indicated tissue samples from three control rats (open bars) and from six exercised rats (shaded bars). Error lines are given as the relevant halves the standard errors of the mean. Asterisks indicate significant differences between control and exercised groups (* P < 0.05; ** P < 0.01).
Vol. 54, No. 3, 1994
Exerdse-indueed Oxidative Damage
153
2 we observed an increase in sulfhydryl content in exercised cardiac microsomes as compared to control cardiac microsomes. This observation is consistent with increased binding of phospholambam, a cysteine-rich regulatory protein. Figure 2 presents the total sulfhydryl content in cardiac homogenates which remained essentially unchanged after exercise. Skeletal Muscle. The appearance of the exercised muscle upon excision was not different from control muscle. The exercised muscles showed a tendency to sporadically contract during surgical removal of the tissue. However, none of the muscles from the control rats exhibited spontaneous twitching during tissue extraction. The concentrations of protein in the control and exercised skeletal muscle homogenates and microsomes were the same. However, we found a ten-fold decrease (Table I) in the phospholipid content of exercised skeletal muscle microsomes when compared with control samples. Skeletal muscle membranes were oxidatively modified. Figure 1 shows that microsomes from exercised skeletal muscle contained higher malondialdehyde levels than microsomes from nonexercised skeletal muscle. Elevated malondialdehyde indicates accelerated lipid peroxidation and a greater susceptibility to oxidative modification due to reaction with proteins forming adducts and crosslinked products. Electrophoretic separation of microsomal proteins presented in Figure 3 indicated distinctive exercise-induced changes in band patterns that reflect increased covalent crosslinking of proteins in the microsomal structures after exercise. We observed a decrease in 43 and 110 kDa proteins together with an increase in protein bands migrating at 145 and 285 kDa. This suggests that the 145 kDa band may represent crosslinking of the 43 and ll0kDa proteins. The 285 kDa band may be dimerization of this 145 kDa crosslinked product. The identity of the 43 kDa protein is as yet unknown. The intensity and position of the 110 kDa band is consistent with the Ca 2+ ATPase of the sarcoplasmic reticulum which is enriched in this fraction. The protein profile of skeletal muscle homogenates confirmed increases in crosslinking and loss of the same proteins after exercise (data not shown). Figure 2 also illustrates the increased protein damage to skeletal muscle microsomes from exercised rats indicated by decreased levels of protein sulfhydryls compared to nonexercised controls. Interestingly, the sulfhydryl content of skeletal muscle homogenates shown in Figure 2 was highest in the exercised samples. The values for the homogenate fractions represent the total cellular (microsomal and cytosolic) sulfhydryl content which suggests a dramatic increase in sulfhydryl compounds in the cytosol. The large increase in cytosolic sulfhydryls in exercised skeletal muscle may indicate activation of a repair mechanism. Blood. As mentioned above, blood was obtained from the dorsal artery which is covered by a layer of connective tissue that had to be peeled away in order to expose the artery. Interestingly, the diameter of the artery from the exercised rat was visibly larger than that of the control rats. However, there was no visible difference between control and exercised rats in the diameter of the vein that runs parallel with this artery. While an objective measurement was not made in this case, the observation raises provoking questions. Experimentation in this direction will likely yield important information. The packed cell volume was the same in exercised and control samples (data not shown). No hemolysis occurred during the isolation of the erythrocytes from either the control or the exercised group. As shown in Figure 1 there was no detectable difference in
154
Exercise-induced Oxidative Damage
Vol. 54, No. 3, 1994
malondialdehyde levels in erythrocyte membranes obtained from control and exercised rats. We have found, however, a decrease (Figure 2) in sulfhydryl content in exercised erythroeyte membranes compared with control samples.
High Molecular Weight Crosslinked Products
ii !! y ~iii
285 kDa
i ii ii!
ill
~ - - 145 kDa
, ~i~!!ii~ill ~ !! iii!!~'!i'i'i'~ii'ii
'! Ca2* - A T P a s e (110 kDa)
i~~il,~, ,~~ii~,
~ - - 43 kDa
I
I =
I
I "I-
SDS polyacrylamide gel electrophoresis of skeletal muscle microsomal proteins. Microsomes were isolated from control (-) and exercised (+) rats that were 268 days old. Arrows indicate major differences.
Vol. 54, No. 3, 1994
Exercise-induced OxidativeDamage
155
Electrophoresis of denatured erythrocyte membrane proteins from the exercised samples revealed the presence of high molecular weight crosslinked products that did not enter the resolving gel (data not shown). This crosslinked protein material was not found in the control samples. TABLE II Exercised-induced change in the level of total sulfhydryl groups in blood plasma
Sulfhydryl Levels Group
N
(nmols SH/mg prot)
Control
3
1.31 _+ 0.08
(~mols SH/dl plasma) 10.8 _ 1.55
Exercised 6 3.17 -+ 0.62 a 24.7 -+ 3.89 a The data are presented as mean _+ standard error of the mean. a p < 0.01 When membrane-free hemolysates were assayed for sulfhydryl content, we observed no change in sulfhydryls between control and exercised samples which is shown in Figure 2. Even when solubilized samples of whole cells were tested, there was no detectable difference in sulfhydryls between the control and exercised groups (data not shown). As presented in Table II, the blood plasma levels of sulfhydryl-containing compounds increased after exercise. Discussion We report changes in the chemical composition of various tissues that occur after exercise. The tissue response represents oxidative and compensatory repair processes. Exercise had its greatest effects on membrane structures. We found oxidized erythrocyte membrane proteins, which supports evidence that exercise causes hemolysis(ll) presumably by affecting membrane integrity. The microsomes from cardiac and skeletal muscle exhibited increased lipid peroxidation products and decreased phospholipid content, and the membrane proteins of skeletal muscle microsomes were oxidized. Sulfhydryl levels in whole erythrocytes, erythrocyte lysates, blood plasma cardiac muscle, homogenates and skeletal muscle homogenates after exercise did not decrease. Our observations that skeletal muscle exhibited higher levels of sulfhydryls after exercise is consistent with evidence that various skeletal muscle transport systems become stimulated upon exercise(12). Increased uptake of cysteine or glutathione would account for the elevation in sulfhydryls observed in skeletal muscle. Increased levels of skeletal muscle sulfhydryls may also be due to activation of pathways for cysteine synthesis. These findings suggest that while there is a down regulation of some antioxidant enzymes(3) a repair process appears to be induced as evidenced by the elevation in skeletal muscle redox status. There was a simultaneous elevation in plasma levels of sulfhydryls which may affect surface membrane transport mechanisms of skeletal muscle. This latter finding is consistent with an observed increase in plasma haptoglobin following exercise(ll) but may also be due to increased cysteine absorption in the gut during the increased blood flow through the splanchnic circulation following exercise.
156
Exerclse-inducedOxidativeDamage
Vol. 54, No. 3, 1994
Since the variation among individuals in the experimental group was generally greater than that for the control group, we examined tissues from a larger number of experimental rats which explains the disparity in animal numbers. We did however observe very little interesse variation and therefore used few animals in some tests such as phospholipid analysis which was performed to confirm lipid decomposition. Modulation of intracellular Ca2+ in muscle cells(13) involves the participation of Ca2+ channels, Na+-Ca2+ exchange systems and ATP-dependent Ca2+ pumps of the sarcolemma and transverse tubules and Ca2+ channels and ATP-dependent Ca 2+ pumps of the sarcoplasmic reticulum. The microsomal preparation that we tested is enriched in sarcoplasmic reticulum. The exercise-induced changes in membrane lipid and crosslinking of the ll0kDa protein which is thought to be the SR Ca2+ pump may greatly effect Ca 2+ homeostasis. Protection and regeneration of lipid-based structures appear to be vital for an exercising individual. While there is evidence that short-term vitamin E supplementation does not decrease lipid peroxide formation(4), vitamin E supplementation has proven beneficial(14). Dietary considerations for athletes may not be the same as that for more those who sporadically exercise. Since our observations are based on effects of acute exercise on untrained animals, the dietary needs for nutrients that protect and regenerate membranes may not be pertinent to the athlete. The sulfhydryl groups are the most reactive of the amino acid side chains. This property makes cysteines important components of catalytic sites. The structure and function of membrane transport systems(15), glycolytic enzymes(16) and contractile proteins(17) are dependent on the redox status of the cell as reflected in the oxidative state of the cysteine residues. Furthermore, non-protein sulfhydryl-containing compounds such as glutathione, which is found at approximately 5mM concentration in the cell, are necessary for the repair of lipid peroxides. It is this fact that has led to the hypothesis that lipid peroxidation accelerates the degenerative processes in certain myopathies(18). In conclusion, damage to subcellular membrane structures was observed in all of the tissues tested. Cardiac and skeletal muscle microsomes from exercised rats exhibited increased lipid peroxidative products and decreased membrane phospholipids. Skeletal muscle microsomes and erythrocyte membranes showed increased protein oxidation and greater protein crosslinking after exercise. The total cellular sulfhydryl content remains the same in cardiac tissue and erythrocytes and is increased in skeletal muscle and blood plasma after exercise. The subcellular membranes in skeletal muscle were chemically modified after exercise and we think that skeletal muscle compensated by increasing total cellular sulfhydryls which would aid in repair (reduced glutathione removes lipid peroxides). The intracellular increase in sulfhydryls in skeletal muscle may be a direct consequence of blood plasma elevations in sulfhydryls. While the increase in sulfhydryls in several tissues correlates with the acute exercise of untrained animals, further research is required to determine the causal relationships. This study suggests that some of the long term health benefits of exercise may be due to increased levels of blood plasma sulfhydryl-containing compounds which may act as detoxifying agents by removing cytotoxic lipid breakdown products. The mechanism for this elevation and the organs involved in regulating blood levels of sulfhydryl-containing compounds are yet to be determined.
Vol. 54, No. 3, 1994
Exercise-inducedOxidativeDamage
157
Acknowledgements The authors wish to thank Kamin McCune and Bill Glosser for their efforts in producing this manuscript and gratefully acknowledge the advice and support of Dr. Douglas Rushing and Dr. Anthony Silvagni. The work reported here was supported by funds from the University of Health Sciences - College of Osteopathic Medicine. References 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18.
P.M. CLARKSON and I. TREMBLAY, J. Appl. Physiol. 65 1-6 (1988). T.E. GRAHAM and B. SALTIN, J. Appl. Physiol. 66 561-566 (1989). L.L. JI, F.W. STRATMEN, and H.A. LARDY, Arch. Biochem. Biophys. 263 137-149 (1988). P.S. BRADY, L.J. BRADY, and D.E. ULLREY, J. Nutr. 109 1103-1109 (1979). L.R. JONES, A.D. WEGENER, and H.K.B. SIMMERMAN, Methods Enzymol. 157 360-369 (1988). A.F.S.A. HABEEB, Methods Enzymol. 2__.5457-464 (1972). J.A. BUEGE and S.D. AUST, Methods Enzymol. 5__22302-310 (1978). G.R. BARTLETI', J. Biol. Chem. 234 466-468 (1959). F. MARCUS, J. RITYENHOUSE, T. CHATTERJEE, and M.M. HOSEY, Methods Enzymol. 9_00352-357 (1982). U.K. LAEMMLI, Nature 227 680-685 (1970). L.M. WEIGHT, M.J. BYRNE, and P. JACOBS, Clin. Science 8._/1147-152 (1991). S. IOTH, R. FUNICELLO, P. ZANIOL and B. BARBIROLI, Biochem. Biophys. Res. Comm. 176 1204-1209 (1991). A. MARTONOSI, Molecular Aspects of Transport Proteins, J.J.H.H.M. De Pont (ed.), 57-116, Amsterdam: Elsevier Science Publishers, (1992). M. MEYDANI, W. EVANS, G. HANDELMAN, R.A. FIELDING, S.N. MEYDANI, M.A. FIATARONE, J.B. BLUMBERG, AND J.G. CANNON, Ann. N.Y. Acad. Sci. 669 363-364 (1992). R.P. HEBBEL, O. SHALEV, W. FOKER, and B.H. RANK, Biochem. Biophys. Acta. 862 8-16 (1986). A.E. BRODIE, and D.J. REED, Arch. Biochem. Biophys. 276 212-218 (1990). J. HIRATSUKA, Biol. Chem. 267 14941-14948 (1992). P.J. RUSSO and N.W. SEIDLER, Med. Hypotheses 3__99147-151 (1992).