Glucose-6-phosphate dehydrogenase (G6PD) deficiency is a common genetic disorder (approximately 7% of the world population is affected) that can result in increased sensitivity of erythrocytes to oxidative stress leading to hemolytic anemia (19). Several hundred G6PD genetic variants have been identified, but fortunately most evoke no clinical disorder or hemolytic anemia. Only two variants, designated G6PD A and G6PD Mediterranean, lead to clinically significant hemolysis (19). In the A type, the activity of G6PD is normal, but the mutant enzyme has a half-life of 13 d instead of 62 d (19). In the Mediterranean type, the hemolysis is more severe because there is both impaired synthesis and reduced stability of the enzyme (19). Inheritance of the mutant gene is X linked. Thus, the defect is expressed in all erythrocytes of the affected male (19). In the heterozygous female, there are normal and deficient erythrocytes, owing to random inactivation of the X chromosome. Therefore, males are more vulnerable to oxidative injury than females (19).
The basic biochemical defect of G6PD deficiency is a decreased or absent ability of erythrocytes to generate NADPH (19). Among the pleiotropic effects of this deficiency is the inability of glutathione reductase, a NADPH-dependent enzyme, to recycle oxidized glutathione (GSSG) to reduced glutathione (GSH) (19). The loss of NADPH and GSH are thought to account for the enhanced rates of Heinz body formation and lipid peroxidation that are observed in G6PD-deficient subjects in response to both endogenous and exogenous oxidants (19).
It is now well established that heavy exercise can accelerate the generation of reactive oxygen species (ROS) that frequently exceeds the capacity of antioxidant defenses and that results in oxidative stress (16,27). Erythrocytes are particularly susceptible to oxidative damage because of the high polyunsaturated fatty acid content in their membranes and the high concentration of oxygen and hemoglobin (24). During exercise, where the production of ROS increases, erythrocytes are at an increased risk of oxidative damage (24). Erythrocytes and all other cells, to cope with the formation of ROS, have an antioxidant defense system that includes enzymatic and nonenzymatic antioxidants. From the antioxidant molecules, GSH/GSSG is the major redox couple in animal cells and of great importance for erythrocytes (29).
In all living organisms, the levels of ROS are controlled by a complex network of antioxidant defenses, which reduce (but do not completely prevent) oxidative damage to biomolecules. In human disease, this "oxidant-antioxidant" balance is tilted in favor of the ROS so that oxidative damage levels increase (10). In some diseases, this makes a significant contribution to tissue injury; therefore, increased oxidative stress may pose a health problem in the general population (10). Additionally, there is evidence that oxidative stress has demonstrated links to fatigue, muscle damage, and reduced immune function (10).
Cells of G6PD-deficient individuals, due to their lower levels of GSH and potentially lower ability to convert GSSG to GSH, fall short of one of the principal radical scavenger molecules against oxidative stress during exercise (Fig. 1). Given that the only way for GSH to be recycled is through the pentose phosphate pathway, a critical step of it catalyzed by the G6PD enzyme, erythrocytes (and other cells to a lesser extent) from G6PD-deficient individuals may have impaired ability to dispose of lipid peroxides and hydrogen peroxides (19). These compounds can cause oxidation of sulfhydryl groups in proteins and peroxidation of lipids in the membrane of a cell. If the cell is an erythrocyte, peroxidation of lipids can cause lysis of its membrane (i.e., hemolysis). If some of the sulfhydryl groups of hemoglobin become oxidized, the protein precipitates inside the erythrocyte, forming Heinz bodies (19). Despite the existence of a theoretically sound background for diminished resistance of G6PD individuals to oxidative challenge (and, therefore, increased susceptibility to erythrocyte lysis), we are not aware of any controlled study that has examined the antioxidant and oxidative stress responses to exercise in G6PD-deficient individuals.
The aim of the present study was to examine the effects of acute high-intensity exercise to exhaustion on blood oxidative stress markers in G6PD-deficient individuals and matched controls. We also investigated whether the duration of exercise is an important determinant of the magnitude of the exercise-induced oxidative stress, since, to our knowledge, there are no studies that have compared the response of oxidative stress indices after acute exercise bouts of different duration. We hypothesized that exercise would result in greater alterations in oxidative stress biomarkers in the G6PD-deficient group and that the exercise-induced oxidative stress would be duration dependent.
Subjects were recruited after advertising the study in the local media. Nine males with established G6PD deficiency and nine males with normal G6PD activity (matched with the G6PD-deficient subjects for age and maximal oxygen consumption; V̇O2max) participated in the present study. Subjects were nonsmokers and were not receiving anti-inflammatory medication or nutritional supplements. A written informed consent to participate in the study was provided by all participants after the volunteers were informed about all risks, discomforts, and benefits involved in the study. All experimental procedures were performed in accordance with the policy statement of the American College of Sports Medicine on research with human subjects as published by Medicine and Science in Sports and Exercise® and were approved by the human subjects committee from the University of Thessaly.
Subjects reported to the laboratory twice. Each participant reported to the laboratory in the morning after an overnight fast and abstained from alcohol and caffeine for 24 h. During their first visit, body mass was measured to the nearest 0.5 kg (Beam Balance 710, Seca, UK) with subjects lightly dressed and barefoot. Standing height was measured to the nearest 0.5 cm (Stadiometer 208, Seca). Percentage body fat was calculated from seven skinfold measures (average of two measurements of each site), using a Harpenden caliper (John Bull, UK), according to published guidelines (2). The anthropometric characteristics of the subjects are shown in Table 1.
Short-duration exercise protocol.
To establish that subjects of the control and G6PD-deficient group were of similar levels of fitness, V̇O2max was determined using a treadmill test to exhaustion. The test commenced at 9 km·h−1, with 2-min speed increments of 1 km·h−1 until exhaustion in a GXC200 treadmill (Powerjog, UK). HR, V̇O2 uptake, and CO2 production were continuously recorded via a computerized online gas analysis system (SensorMedics 2900c, SensorMedics Corporation) calibrated to known gases. The V̇O2max of the subjects is shown in Table 1. Hereafter, we refer to this V̇O2max test as the short-duration exercise protocol.
Long-duration exercise protocol.
Seven to 14 d after the short-duration exercise protocol, the subjects visited the laboratory for a second time to perform an acute bout of long-duration exercise until exhaustion. The subjects initially ran for 45 min on a treadmill at an intensity corresponding to 70-75% of their established V̇O2max. Speed adjustments during exercise were made to ensure that the subjects were exercising within 70-75% V̇O2max. After the first 45 min of stable-intensity running, speed was increased to 90% V̇O2max up to the athlete's exhaustion. V̇O2max was monitored during exercise every 5 min for 60 s. Exercise was performed at a temperature of 21 ± 2°C and 45 ± 4% relative humidity. To attenuate subjects' discomfort and plasma volume changes, volunteers had access to water ad libitum during the first 45 min.
Blood collection and handling.
Before each exercise protocol and within 2 min from the completion of exercise, a blood sample was drawn from a forearm vein. A portion of blood was collected into a tube containing ethylenediamine tetraacetic acid and was placed immediately on ice for the determination of hematological parameters, G6PD activity, and Heinz bodies. Whole-blood lysates were produced by adding 5% trichloroacetic acid to whole blood (1:1 vol/vol) collected in plain tubes, vortexed vigorously, and centrifuged at 4,000 × g for 10 min at 4°C. The supernatants were removed and centrifuged again at 28,600 × g for 5 min at 4°C. The clear supernatants were transferred in Eppendorf™ tubes and were used for GSH and GSSG determination. Another portion of blood was also collected in plain tubes, left on ice for 20 min to clot, and centrifuged at 1500 × g for 10 min at 4°C for serum separation. Serum was transferred in tubes and was used for the determination of thiobarbituric acid-reactive substances (TBARS), protein carbonyls, catalase, and total antioxidant capacity (TAC).
Assays in whole blood.
Postexercise changes in plasma volume were computed based on hematocrit and hemoglobin as described (6). The hematological parameters were measured in a Sysmex K-1000 (TOA Electronics, Japan) autoanalyzer. G6PD activity was determined with a commercially available kit from Sigma (St. Louis, MO). The intraassay coefficient of variation for G6PD activity was 2.9%.
For Heinz body determination, 0.5 mL of whole blood was mixed with 10 mL of sodium nitrite (Merck), and the solution was placed in a water bath at 37°C for 30 min. Thereafter, equal volume of the solution and methyl violet (Amresco) was placed on a plate and examined for Heinz bodies (which are the result of hemoglobin oxidation inside erythrocytes) under a light microscope with 100× magnification. Hematocrit, hemoglobin, G6PD activity, and Heinz bodies were determined on the day of blood collection.
Assays in whole-blood lysates and serum.
GSH was measured according to Reddy et al. (28), and GSSG was determined according to Tietze (30). TBARS were measured according to Keles et al. (17), protein carbonyls according to Patsoukis et al. (25), and catalase activity according to Aebi (1). TAC of serum was based on the scavenging of 1,1-diphenyl-2-picrylhydrazyl and was determined according to Janaszewska and Bartosz (15). The intraassay coefficient of variation for each measurement was GSH 4.1%, GSSG 6.1%, TBARS 4.7%, protein carbonyls 5.4%, catalase 7.0%, and TAC 3.4%. Total protein in serum was assayed using a Bradford reagent from Sigma.
Each assay in whole-blood lysates and serum was performed at least in triplicate on the same day to eliminate variation in assay conditions and within 1 month of the blood collection. Blood samples were stored in multiple aliquots at −30°C and thawed only once before analysis.
To control for the effect of previous diet on the outcome measures of the study, and to establish that participants of both groups had similar levels of macronutrient and antioxidant intake, they were asked to record their diet for 3 d preceding the short-duration exercise protocol and to repeat this diet before the long-duration exercise protocol. Each subject had been provided with a written set of guidelines for monitoring dietary consumption and a record sheet for recording food intake. Diet records were analyzed using the nutritional analysis system Science Fit Diet 200A (Sciencefit, Greece), and the results of the analysis are presented in Table 2.
Data are presented as mean ± SD. The distribution of all dependent variables was examined by the Kolmogorov-Smirnov test and was found not to differ significantly from normal. Data were analyzed using three-way (group × protocol × time) ANOVA with repeated measures on protocol and time. If a significant interaction was obtained, pairwise comparisons were performed through simple main effect analysis. Differences in diet among groups were examined through two-way (group × protocol) ANOVA with repeated measures on protocol. Differences in physical characteristics and G6PD activity between the control and G6PD group were examined by unpaired Student's t-test. The level of statistical significance was set at α = 0.05. The SPSS version 13.0 was used for all analyses (SPSS Inc.).
Anthropometry, V̇O2max, and diet.
There were no significant differences in any of the physical characteristics and fitness level between the two groups (P > 0.05). Additionally, no significant differences in daily energy and macronutrient and antioxidant intake were found between the two groups and between the first and the second exercise protocols within each group (P > 0.05). The duration of the short exercise protocol was 12.2 ± 1.4 min in the control and 11.8 ± 2.0 min in the G6PD-deficient group. Regarding the long exercise protocol, during its first 45 min, control subjects ran at an intensity corresponding to 74.7 ± 1.7% of their V̇O2max, whereas G6PD-deficient subjects ran at 74.4 ± 2.2% V̇O2max. The duration of the long exercise protocol was 50.2 ± 3.4 min in the control and 51.8 ± 4.0 min in the G6PD-deficient group. There were no significant differences in exercise intensity and duration within each protocol between groups (P >0.05).
The values of the hematological parameters measured are presented in Table 3. The main finding regarding hematology is the significantly lower values in hematocrit, hemoglobin, and red blood cell count in G6PD-deficient individuals compared to controls (P < 0.05 in all three variables). Nevertheless, the effect of both exercises on the levels of all hematological parameters was independent of the group (i.e., either G6PD deficient or not) of the participants (P > 0.05). On the other hand, white blood cell count and platelet count increased significantly after exercise of both durations (P > 0.05).
Postexercise plasma volume relative to preexercise was 0.99 ± 0.05 in the control group after the short exercise protocol and 1.01 ± 0.04 after the long exercise protocol (both not significant; P >0.05). The corresponding plasma volume changes in the G6PD-deficient group were 0.98 ± 0.03 and 1.01 ± 0.05, respectively (both not significant; P >0.05). When postexercise values of biochemical parameters measured were corrected for plasma volume changes, the results of the statistical comparisons were not different from those performed on the original values; therefore, the original values are presented.
G6PD activity and Heinz bodies.
G6PD activity was significantly higher (by 25 times; P < 0.0001) in control compared with G6PD-deficient individuals (9.23 ± 2.61 vs 0.37 ± 0.11 U·g−1 Hb). To examine whether exercise was capable of oxidizing hemoglobin, we examined erythrocytes microscopically for the presence of Heinz bodies, which are various forms of denatured hemoglobin. However, Heinz body formation was not seen in both groups either pre- or postexercise in both protocols.
The values of the biochemical parameters examined are presented in Figures 2-8. Regarding GSH status, exercise of both durations decreased GSH concentrations (main effect of time, P = 0.007), even though the decrements after the long exercise protocol were larger, and, as a result, a significant interaction of protocol × time appeared (P = 0.045). GSH levels at rest were 2.1 and 1.8 times higher in the control group compared with the G6PD-deficient group before the short and long exercise protocols, respectively (main effect of group, P = 0.002). The effect of exercise on GSSG was also dependent on exercise duration (interaction of protocol × time, P = 0.004). Specifically, the short exercise bout did not alter the levels of GSSG (P > 0.05), whereas the long exercise bout increased GSSG levels (P = 0.016) regardless of the G6PD status of the participant. GSSG levels at rest were higher by 2.0 and 1.4 times in control compared with the G6PD-deficient group before the short and long exercise protocols, respectively (main effect of group, P = 0.003). Due to the above-described effects of exercise on GSH and GSSG, their ratio did not change after the short exercise bout (P > 0.05), whereas it decreased after the long exercise bout in the controls (P = 0.050) and the G6PD-deficient group (P = 0.013) (interaction of protocol × time, P = 0.010).
TBARS, protein carbonyls, catalase, and TAC.
Exercise of both durations increased the concentrations of TBARS (main effect of time, P < 0.0001), protein carbonyls (P < 0.0001), catalase activity (P = 0.034), and TAC (P = 0.008), independent of G6PD status.
To our knowledge, this is the first investigation of the effect of exhaustive exercise on the redox status of individuals with G6PD deficiency. Our hypothesis was that exercise would result in greater oxidative stress in the G6PD-deficient group. However, the present results revealed that exercise similarly modified the levels of the selected oxidative stress markers and did not oxidize hemoglobin (as the absence of Heinz bodies indicated) in persons with or without G6PD deficiency.
Redox status of the control and G6PD-deficient individuals at rest.
GSH is the most abundant low-molecular weight thiol, and GSH/GSSG is the major redox couple in animal cells (22). We found about twofold lower resting GSH levels in subjects with G6PD deficiency compared with non-G6PD-deficient individuals. Similar findings have been reported by other relevant studies (13,18), even though Bilmen et al. (4) have not found such differences. This discrepancy may be due to the difference in the age of the participants among the studies. Namely, Bilmen et al. (4) used individuals aged 0-12 yr, whereas the other studies used adults. Nevertheless, the GSSG levels were also about twofold lower in G6PD-deficient individuals, and, as a result, the GSH/GSSG ratio was similar between deficient subjects and controls.
Lipid peroxidation was not found to be significantly different in G6PD-deficient individuals, based on serum TBARS measured. This agrees with some findings (4), whereas others have reported increased levels in G6PD-deficient individuals (9). Because we found less than half of GSH and no differences in lipid peroxidation in G6PD-deficient individuals, it is probable that these individuals have developed alternative protective mechanisms from oxidation. In addition, TAC (an indicator of the overall antioxidant capacity of serum) did not differ between G6PD-deficient and non-G6PD-deficient individuals. Although the blood antioxidant defenses were similar between the two groups, the levels of antioxidants in muscle may be higher in G6PD-deficient individuals.
Redox status of the control and G6PD-deficient individuals after exercise.
The main finding of the present study is that G6PD-deficient individuals responded in a similar manner with non-G6PD-deficient individuals after acute exhaustive exercise. G6PD-deficient individuals, despite their much lower levels of GSH, are able to exercise at a high intensity until exhaustion without experiencing higher blood oxidative stress (or oxidation of hemoglobin) than their non-G6PD-deficient counterparts. Except for the fact that the levels of antioxidants in muscle may be higher in G6PD-deficient individuals, it is also likely that the activity of some antioxidant molecules crucial for maintaining redox status in erythrocytes are increased in G6PD-deficient persons. Alternatively, the composition of erythrocyte membranes may be different between the groups (e.g., fatty acid profile more resistant to peroxidation in G6PD-deficient individuals compared with nondeficient).
To the best of our knowledge, the only studies relevant to the effect of exercise on G6PD-deficient subjects are three case studies (5,14,23). The case studies by Bresolin et al. (5) and Ninfali et al. (23) reported that one G6PD-deficient individual (in each study) was hospitalized for myalgia and myoglobinuria after intense exercise. Nevertheless, it is not clear whether these clinical symptoms appeared due to the G6PD deficiency or were just a result of a preceding severe exercise bout. Indeed, myalgia and myoglobinuria can also be found in non-G6PD-deficient individuals after intense, particularly eccentric, exercise (11,20). It is therefore probable, in our opinion, that the symptoms reported in these studies were due to preceding exercise bout and were not caused by the G6PD deficiency itself. The third case study, from our laboratory (14), generally showed that 30 min of treadmill exercise at 70-75% V̇O2max did not affect the levels of several oxidative stress indices in one individual with G6PD deficiency. However, it should be noted that the subject in this study was well trained (his V̇O2max was 57.1 mL·kg−1·min−1) and may have been protected by an enhanced antioxidant defense system (26).
Effect of exercise duration on redox status.
The effect of exercise on GSH status (in all three parameters) was dependent on exercise duration. Within this context, it must be mentioned that the exercise intensities of the two protocols were different (i.e., the short one was incremental and the long one was steady, for the most part); therefore, discussing the effects of exercise duration on redox status should be done with some caution. Specifically, both exercise protocols significantly decreased the levels of GSH, though the decreases were much higher after the long exercise protocol (12.4-24.9% after the short exercise vs 45.4-47.5% after the long exercise). Regarding GSSG, only the long exercise was capable to increase its levels (43.3-56.2%). As a result, GSH/GSSG ratio did not change significantly after the short exercise bout, whereas it decreased significantly by 61.9% in controls and 66.4% in G6PD-deficient individuals after the long exercise bout. These results support the idea that exercise duration is an important determinant of the magnitude of the exercise-induced changes in GSH status. It is probable that during high-intensity exercise, such as the one used in the present study, hepatic GSH supply may not be sufficient to match the enhanced peripheral utilization, and this results in a net reduction of blood GSH. Except for a possible insufficient hepatic GSH release, there might be an increase in the blood clearance of GSH (e.g., increased consumption in muscle) during prolonged strenuous exercise, which would also explain a decrease in blood GSH concentration. The majority of the human literature generally agrees that acute exhaustive exercise decreases GSH, increases GSSG, and decreases their ratio (reviewed in Sen and Packer (29)). However, to the best of our knowledge, there are no studies that have compared the response of GSH after exercise bouts of different durations.
Acute exercise increased the levels of TBARS, protein carbonyls, catalase, and TAC independent of exercise duration. We are not aware of any study that has investigated the effect of exercise duration on these oxidative stress indices. On the other hand, several studies have examined the effect of a single duration exercise on these indices and have generally reported similar increases to ours (21) (regarding TBARS (8), protein carbonyls (7), and TAC), with the exception of catalase activity, where some researchers have reported increases after exercise (12), and others no differences (21). Summarizing the results related to the effect of exercise duration on redox status, we report that exercise duration is an important determinant of the magnitude of exercise-induced changes for GSH, GSSG, and GSH/GSSG, but not for TBARS, protein carbonyls, catalase activity, and TAC.
Why was only the response of the GSH-related parameters dependent on exercise duration? This is a difficult question to answer through the present, essentially descriptive, investigation. However, this different response among oxidative biomarkers probably has to do with the involvement of different tissues that control their homeostasis (the liver in GSH and skeletal muscle in the other indices). What would probably be the most sensitive measure is to take pre- and postexercise muscle biopsy samples and blood samples to determine the shift in GSH movement from the organs to the blood. The duration-dependent GSH response implies that GSH, GSSG, and GSH/GSSG "sense" exercise duration and probably are better blood indices for describing the effects of exercise on blood redox status than the other indices measured in the present work.
The small (from 2.1 to 5.1%) but consistent increases in TAC after exercise in all groups suggest that acute exercise activates antioxidant defenses of the body. Mobilization of tissue antioxidant stores into plasma is probably one of the mechanisms behind the significant increase (and not decrease, as might be intuitively expected) of TAC that was observed in the present study. This is a widely accepted phenomenon that would help maintain or even increase antioxidant status of plasma in times of need (3). This increase in TAC is primarily due to increasing uric acid and vitamin C in blood during exercise (3).
G6PD-deficient individuals are able to perform short- and long-duration acute exercise without experiencing greater oxidative stress than non-G6PD-deficient individuals (as indicated by the oxidative stress biomarkers used in the present study). The present findings illustrate that despite the theoretically lower capacity of G6PD-deficient individuals to resist perturbations in their redox status, they are not more susceptible to oxidative stress, probably because they have developed alternative protective mechanisms. On the other hand, the effect of exercise on GSH status was dependent on duration, whereas exercise increased the levels of TBARS, protein carbonyls, catalase activity, and TAC independently of duration. Future studies should aim to delineate the mechanisms that enable G6PD-deficient individuals to adequately respond to an oxidant challenge, such as that imposed by an exhaustive exercise.
The first author was supported by a postdoctoral scholarship from the Greek State Scholarships Foundation. We thank the two anonymous reviewers for their helpful comments.
1. Aebi, H. Catalase in vitro. Methods Enzymol.
2. American College of Sports Medicine. ACSM's Guidelines for Exercise Testing and Prescription
, Philadelphia: Lippincott Williams & Wilkins, 2000, p. 77.
3. Balakrishnan, S. D., and C. V. Anuradha. Exercise, depletion of antioxidants and antioxidant manipulation. Cell Biochem. Funct.
4. Bilmen, S., T. A. Aksu, S. Gumuslu, D. K. Korgun, and D. Canatan. Antioxidant capacity of G-6-PD-deficient erythrocytes. Clin. Chim. Acta
5. Bresolin, N., L. Bet, M. Moggio, et al. Muscle glucose-6-phosphate dehydrogenase deficiency. J. Neurol.
6. Dill, D. B., and D. L. Costill. Calculation of percentage changes in volumes of blood, plasma, and red cells in dehydration. J. Appl. Physiol.
7. Fatouros, I. G., A. Z. Jamurtas, V. Villiotou, et al. Oxidative stress responses in older men during endurance training and detraining. Med. Sci. Sports Exerc.
8. Goldfarb, A. H., S. W. Patrick, S. Bryer, and T. You. Vitamin C supplementation affects oxidative-stress blood markers in response to a 30-minute run at 75% V̇O2max
. Int. J. Sport Nutr. Exerc. Metab.
9. Gurbuz, N., O. Yalcin, T. A. Aksu, and O. K. Baskurt. The relationship between the enzyme activity, lipid peroxidation and red blood cells deformability in hemizygous and heterozygous glucose-6-phosphate dehydrogenase deficient individuals. Clin. Hemorrheol. Microcirc.
10. Halliwell, B. Free radicals and other reactive species in disease. Encyclopedia of Life Sciences
, 2001, pp. 11-17.
11. Hamer, R. When exercise goes awry: exertional rhabdomyolysis. South Med. J.
12. Inal, M., F. Akyuz, A. Turgut, and W. M. Getsfrid. Effect of aerobic and anaerobic metabolism on free radical generation swimmers. Med. Sci. Sports Exerc.
13. Jain, S. K. Glutathione and glucose-6-phosphate dehydrogenase deficiency can increase protein glycosylation. Free Radic. Biol. Med.
14. Jamurtas, A. Z., I. G. Fatouros, V. S. Alexiou, S. C. Chung, and A. H. Goldfarb. Exercise-induced oxidative damage in a person with glucose-6-phosphate dehydrogenase deficiency. J. Hum. Mov. Stud.
15. Janaszewska, A., and G. Bartosz. Assay of total antioxidant capacity: comparison of four methods as applied to human blood plasma. Scand. J. Clin. Lab. Invest.
16. Ji, L. L. Antioxidants and oxidative stress in exercise. Proc. Soc. Exp. Biol. Med.
17. Keles, M. S., S. Taysi, N. Sen, H. Aksoy, and F. Akcay. Effect of corticosteroid therapy on serum and CSF malondialdehyde and antioxidant proteins in multiple sclerosis. Can. J. Neurol. Sci.
18. Magnani, M., V. Stocchi, F. Canestrari, et al. Redox and energetic state of red blood cells in G6PD deficiency, heterozygous beta-thalassemia and the combination of both. Acta Haematol.
19. Mehta, A., P. J. Mason, and T. J. Vulliamy. Glucose-6-phosphate dehydrogenase deficiency. Baillieres Best Pract. Res. Clin. Haematol.
20. Milne, C. J. Rhabdomyolysis, myoglobinuria and exercise. Sports Med.
21. Miyazaki, H., S. Oh-ishi, T. Ookawara, et al. Strenuous endurance training in humans reduces oxidative stress following exhausting exercise. Eur. J. Appl. Physiol.
22. Mylonas, C., and D. Kouretas. Lipid peroxidation and tissue damage. In Vivo
23. Ninfali, P., N. Bresolin, L. Baronciani, et al. Glucose-6-phosphate dehydrogenase Lodi844C: a study on its expression in blood cells and muscle. Enzyme
24. Oztasan, N., S. Taysi, K. Gumustekin, et al. Endurance training attenuates exercise-induced oxidative stress in erythrocytes in rat. Eur. J. Appl. Physiol.
25. Patsoukis, N., G. Zervoudakis, N. T. Panagopoulos, C. D. Georgiou, F. Angelatou, and N. A. Matsokis. Thiol redox state (TRS) and oxidative stress in the mouse hippocampus after pentylenetetrazol-induced epileptic seizure. Neurosci. Lett.
26. Powers, S. K., L. L. Ji, and C. Leeuwenburgh. Exercise training-induced alterations in skeletal muscle antioxidant capacity: a brief review. Med. Sci. Sports Exerc.
27. Radak, Z., A. W. Taylor, H. Ohno, and S. Goto. Adaptation to exercise-induced oxidative stress: from muscle to brain. Exerc. Immun. Rev.
28. Reddy, Y. N., S. V. Murthy, D. R. Krishna, and M. C. Prabhakar. Role of free radicals and antioxidants in tuberculosis patients. Indian J. Tuberc.
29. Sen, C. K., and L. Packer. Thiol homeostasis and supplements in physical exercise. Am. J. Clin. Nutr.
30. Tietze, F. Enzymic method for quantitative determination of nanogram amounts of total and oxidized glutathione: applications to mammalian blood and other tissues. Anal. Biochem.