- Email: [email protected]
Hyperoxia during reperfusion is a factor in reperfusion injury
Free Radical Biology & Medicine, Vol. 6, pp. 61-62, 1989 Printed in the USA. All rights reserved.
÷
0891-5849/89 $3.00 + .00 © 1989 Pergamon Press plc
Hypothesis Paper HYPEROXIA DURING REPERFUSION IS A FACTOR IN REPERFUSION INJURY MYRON L. WOLBARSHT and IRWIN FRIDOVICH* The Departmentsof Psychology,BiomedicalEngineering, and Biochemistry,Duke University, Durham, NC 27710 (Received 3 December 1987;Revised 25 February 1988;Accepted 15 March 1988)
Abstract--Imposition of ischemia should result in accumulation of lactic acid with an attendant drop in pH. Subsequent reperfusion would result in hyperoxia, in the affected tissue, due to the Bohr Effect. 02 should therefore be produced in greater than normal amounts, due to this transient hyperoxia, and may contribute to reperfusion injury. Tissue acidification, during extreme exercise or in diabetes mellitus, may similarly lead to hyperoxia and to tissue damage by 02 •
INTRODUCTION
Ischemia and its attendant hypoxia necessitates reliance upon anaerobic glycolysis. In addition to the
CO2 that remains after aerobic metabolism has decreased, lactic acid then accumulates in the hypoxic tissue and further lowers the pH. Mammalian hemoglobins exhibit a positive Bohr effect, which is to say that oxyhemoglobin is more acidic than deoxyhemoglobin, so that oxygenation is accompanied by release of protons. The oxygen-binding coefficient for mammalian hemoglobin is therefore lessened by a drop in pH. Blood entering the acidic environment of previously ischemic tissue would unload oxygen to a greater degree and establish a higher pO2 than would be the case at the pH of normal tissue. Figure 1 gives some data at selected pH levels to indicate the magnitude of this effect. For example, if the normal pH is 7.4 and the ischemic tissue pH is taken to be 6.4 which is the value in normal tissue capillaries during heavy exercise,~4 the oxygen tension could be doubled or tripled; the maximum 02 tension reached would be increased further by the reactive hyperemia which is seen during reperfusion by giving a higher initial 02 tension. Any CO2 remaining in the tissue at the time of reperfusion would add to the magnitude of the Bohr shift by lowering the pH still more. H. B. Wagnert has shown that imposition of ischemia by clamping of the carotid arteries causes the pO2 in gerbil brain tissue to drop precipitously from 18-23 mm Hg to zero. Reperfusion after 5 min of ischemia resulted in rapid reestablishment of normal blood flow but the tissue pO2 rose to 100 mm Hg and remained at this level for 5 minutes
*Author to whom correspondenceshould be addressed.
"['PersonalCommunication, Dr. H. B. Wagner, National Institute of Neurological and CommunicativeDisorders, Bethesda, Maryland.
The possibility that the tissue injury, which follows temporary hypoxia, might really occur duringreperfusion and might reflect generation of oxygen radicals was presented almost a decade ago 1 and has since received ample support. 2 The protective effects of superoxide dismutase 3 and of allopurinol 4 have led to a scenario in which hypoxanthine, accumulated during hypoxia, is oxidized by xanthine oxidase when oxygen is again available during reperfusion with production of O2-.2'5 Recently, spin trapping has been used to demonstrate the production of free radicals during reperfusion of ischemic myocardium. 6-8 There is another factor which has not yet been considered in relationship to the phenomenon of reperfusion and that is the hyperoxia which would occur during reperfusion of the affected tissues. This hyperoxia, resulting from acidification of the hypoxic tissues and from the Bohr effect on the hemoglobin of the entering blood, would increase superoxide radical production. Thus, in the case of xanthine oxidase, increasing pO2 favors the univalent reduction of 02 to 02- over its divalent reduction to H202 .9 Raising the pO2 also increased 02- production by lung homogenates ~° and by mitochondria. ~j,~2 Reperfusion hyperoxia
61
62
WOLBARSHT
M.L.
Bohr coeff. 255
E
.325
215
,375
/ / /
E
• •
.425
~
/
/
.
Acknowledgements--We wish to thank J. Fordice for providing Table I and Sara DeWitt for much assistance with the refcrences. This work was supported by grants from the Council for Tobacco Research-U.S.A., Inc., the U. S. Army Research Office, The National Institutes of Health, and the National Science Foundation.
l-t
175
CN
- '~
.466
~
~
l-t
O
CL
FRIDOVI['it
chemia. All the above suggests that reperfusion injury may derive in part from excess oxygen tensions during reperfusion, and that similar mechanisms for injury exist in many other situations in which the tissue pH may become lower than normal.~'-~°
295
O3 -r'-
and I.
135
REFERENCES 7.4
7.2
7.0
6.8
6.6
6.4
pH Fig. 1. pH-Induced Shifts in pO2 for Hemoglobin at Several Values of the Bohr Coefficient. All values were calculated assuming that pO2 of the entering arterial blood is 95 mm Hg and using the equation log pO, = Bohr Coef. (7.4 - pHi The highest Bohr coefficient used in these calculations, that is, 0.466, is probably closest to the one which would apply in ischemic tissue, yet the others may apply when the initial O~ saturation of the hemoglobin is lower. All of these curves illustrate the exponentially increasing pO2 with decrease in pH.
before declining towards the normal range of 18-23 mm Hg. It is thus clear that the reperfusion hyperoxia we are proposing does occur and does persist for several minutes. The use of the Bohr Shift induced by lactic acid and a low tissue pH to increase blood 02 tension is well known in at least one type of normal physiology, the filling of the fish swim bladder. ~5.,6 As in the fish swim bladder case, further increases in the 02 tension can occur through salting out and similar affects.X6 These have all been neglected by the calculations above but would increase 02 tension to some degree, at least. Some indication that the Bohr shift may be important in the absence of accumulation of hypoxanthine is suggested by a similar type of capillary injury following heavy exercise. ~7 In this case, there is an accumulation of lactic acid without ischemia, and, thus, the hypoxia would be unlikely to be accompanied by the accumulation of hypoxanthine. Another case is diabetes in which capillary leakage and damage occur chronically, possibly due to the constant elevation of capillary oxygen tension resulting from the lactic acid accumulation induced by elevated glucose levels. ~8 Simple elevation of oxygen in the breathing mixture will increase free radical formation and is accompanied by vascular pathology that closely resembles that found in reperfusion injury. ~0 The increased oxygen tensions in these cases are similar to those used to calculate the data in Figure 1 and are based on tissue lactic acid accumulation probably below that found during is-
1. Fridovich, I. Hypoxia and oxygen toxicity. Adv. Neurol. 26:255 259; 1979. 2. McCord, J. M. Oxygen-derived radicals: a link between reperfusion injury and inflammation. Fed. Proc. 46:2402-2406: 1987. 3. Granger, D. N.; Rutili, G.; McCord, J. M. Superoxide radicals in feline intestinal ischemia. Gastroenterology 81:22-29; 1981. 4. Parks, D. A.; Bulkley, G. B.; Granger, D. N.: Hamilton, S. R.; McCord, J. M. Ischemic injury in the cat small intestine: role of superoxide radicals. Gastroenterology 82:9-15:1982. 5. Smith, S. M.; Grisham, M. B.; Manci, E. A.; Granger, D. N.; Kvietys, P. R. Gastric mucosal injury in the rat. Role of iron and xanthine oxidase. Gastroenterology 92:950-956; 1987. 6. Blasig, I. E.; Ebert, B.; Lowe, H. Identification of free radicals trapped during myocardial ischemia In vitro by ESR. Studia Biophysica 116:35-42: 1986. 7. Zweier, J. L.; Flaherty, J. T.: Weisfeldt, M. L. Direct measurement of free radicals generated following reperfusion of ischemic myocardium. Pro('. Natl. Acad. Sci. U.S.A. 84:1404 1407; 1987. 8. Kramer, J. H.; Arroyo, C. M.; Dickens, B. F.; Weglicki, W. B. Spin trapping evidence that graded myocardial ischemia alters post-ischemic superoxide production. Free Radio. Biol. Med. 3:153--159; 1987. 9. Fridovich, I. Quantativc aspects of the production of superoxide anion radical by xanthine oxidasc. J. Biol. Chem. 245:40534057; 1970. 10. Freeman, B. A.; Topolsky, M. K.; Crapo, J. D. Hyperoxia increases oxygen radical production in rat lung homogenates. Arch. Biochem. Biophys. 216:477-489; 1982. I 1. Boveris, A.: Chance, B. The mitochondrial generation of hydrogen peroxide. General properties and effect of hyperbaric oxygen. Biochem. J. 134:707-716; 1973. 12. Boveris, A. Mitochondrial production of superoxide radical and hydrogen peroxide. Advan. Exptl. Biol. Med. 78:67 82; 1977. 13. Bohr, C.; Hasselbach, K.; Krogh, A. Ueber eninen in biologischer beziehung wichtigen einfluss, den die kohlensaurespannung des blutes auf dessen sauerstoffbindung ubt. Skand. Arch. Physiol. 16:402-412; 1904. 14. Hermansen, L.; Osnes, J. Blood and muscle pH after maximal exercisc in man. J. Appl. Physiol. 32:304-308; 1972. 15. Scholander, P. F.; Van Dam, L. Secretion of gases against high pressure in the swim bladder of deep sea fishes. 1. Oxygen dissociation in blood. Biol. Bull. 207:247-259; 1954. 16. Gerth, W. A.; Hemmingson, E. A. Hints of gas secretion by the salting out effect in the fish swim bladder. J. Comp. Physiol. 146:129-136; 1982. 17. Davies, K. J. A.: Quintanilha, A. T.; Brooks, G. A.; Packer. L. Free radicals and tissue damage produced by exercise. Biochem. Biophys, Res. Comm. 107:1198-1205; 1982. 18. Kelton, J. G.; Collins, F. F. Exceptionally high arterial oxygen tension in diabetic ketoacidosis. Southern Med. J. 72:11271128; 1979. 19. Lumley, J.; Wood, C. Unexpected oxygen tensions in fetal acidosis. J. Perinat. Med. 1:166 173; 1973. 20. Halliwell, B. Oxidants and human disease: some new concepts. Fed. Amer. Soc. Exp. Biol. J. 1:358-364; 1987.
÷
0891-5849/89 $3.00 + .00 © 1989 Pergamon Press plc
Hypothesis Paper HYPEROXIA DURING REPERFUSION IS A FACTOR IN REPERFUSION INJURY MYRON L. WOLBARSHT and IRWIN FRIDOVICH* The Departmentsof Psychology,BiomedicalEngineering, and Biochemistry,Duke University, Durham, NC 27710 (Received 3 December 1987;Revised 25 February 1988;Accepted 15 March 1988)
Abstract--Imposition of ischemia should result in accumulation of lactic acid with an attendant drop in pH. Subsequent reperfusion would result in hyperoxia, in the affected tissue, due to the Bohr Effect. 02 should therefore be produced in greater than normal amounts, due to this transient hyperoxia, and may contribute to reperfusion injury. Tissue acidification, during extreme exercise or in diabetes mellitus, may similarly lead to hyperoxia and to tissue damage by 02 •
INTRODUCTION
Ischemia and its attendant hypoxia necessitates reliance upon anaerobic glycolysis. In addition to the
CO2 that remains after aerobic metabolism has decreased, lactic acid then accumulates in the hypoxic tissue and further lowers the pH. Mammalian hemoglobins exhibit a positive Bohr effect, which is to say that oxyhemoglobin is more acidic than deoxyhemoglobin, so that oxygenation is accompanied by release of protons. The oxygen-binding coefficient for mammalian hemoglobin is therefore lessened by a drop in pH. Blood entering the acidic environment of previously ischemic tissue would unload oxygen to a greater degree and establish a higher pO2 than would be the case at the pH of normal tissue. Figure 1 gives some data at selected pH levels to indicate the magnitude of this effect. For example, if the normal pH is 7.4 and the ischemic tissue pH is taken to be 6.4 which is the value in normal tissue capillaries during heavy exercise,~4 the oxygen tension could be doubled or tripled; the maximum 02 tension reached would be increased further by the reactive hyperemia which is seen during reperfusion by giving a higher initial 02 tension. Any CO2 remaining in the tissue at the time of reperfusion would add to the magnitude of the Bohr shift by lowering the pH still more. H. B. Wagnert has shown that imposition of ischemia by clamping of the carotid arteries causes the pO2 in gerbil brain tissue to drop precipitously from 18-23 mm Hg to zero. Reperfusion after 5 min of ischemia resulted in rapid reestablishment of normal blood flow but the tissue pO2 rose to 100 mm Hg and remained at this level for 5 minutes
*Author to whom correspondenceshould be addressed.
"['PersonalCommunication, Dr. H. B. Wagner, National Institute of Neurological and CommunicativeDisorders, Bethesda, Maryland.
The possibility that the tissue injury, which follows temporary hypoxia, might really occur duringreperfusion and might reflect generation of oxygen radicals was presented almost a decade ago 1 and has since received ample support. 2 The protective effects of superoxide dismutase 3 and of allopurinol 4 have led to a scenario in which hypoxanthine, accumulated during hypoxia, is oxidized by xanthine oxidase when oxygen is again available during reperfusion with production of O2-.2'5 Recently, spin trapping has been used to demonstrate the production of free radicals during reperfusion of ischemic myocardium. 6-8 There is another factor which has not yet been considered in relationship to the phenomenon of reperfusion and that is the hyperoxia which would occur during reperfusion of the affected tissues. This hyperoxia, resulting from acidification of the hypoxic tissues and from the Bohr effect on the hemoglobin of the entering blood, would increase superoxide radical production. Thus, in the case of xanthine oxidase, increasing pO2 favors the univalent reduction of 02 to 02- over its divalent reduction to H202 .9 Raising the pO2 also increased 02- production by lung homogenates ~° and by mitochondria. ~j,~2 Reperfusion hyperoxia
61
62
WOLBARSHT
M.L.
Bohr coeff. 255
E
.325
215
,375
/ / /
E
• •
.425
~
/
/
.
Acknowledgements--We wish to thank J. Fordice for providing Table I and Sara DeWitt for much assistance with the refcrences. This work was supported by grants from the Council for Tobacco Research-U.S.A., Inc., the U. S. Army Research Office, The National Institutes of Health, and the National Science Foundation.
l-t
175
CN
- '~
.466
~
~
l-t
O
CL
FRIDOVI['it
chemia. All the above suggests that reperfusion injury may derive in part from excess oxygen tensions during reperfusion, and that similar mechanisms for injury exist in many other situations in which the tissue pH may become lower than normal.~'-~°
295
O3 -r'-
and I.
135
REFERENCES 7.4
7.2
7.0
6.8
6.6
6.4
pH Fig. 1. pH-Induced Shifts in pO2 for Hemoglobin at Several Values of the Bohr Coefficient. All values were calculated assuming that pO2 of the entering arterial blood is 95 mm Hg and using the equation log pO, = Bohr Coef. (7.4 - pHi The highest Bohr coefficient used in these calculations, that is, 0.466, is probably closest to the one which would apply in ischemic tissue, yet the others may apply when the initial O~ saturation of the hemoglobin is lower. All of these curves illustrate the exponentially increasing pO2 with decrease in pH.
before declining towards the normal range of 18-23 mm Hg. It is thus clear that the reperfusion hyperoxia we are proposing does occur and does persist for several minutes. The use of the Bohr Shift induced by lactic acid and a low tissue pH to increase blood 02 tension is well known in at least one type of normal physiology, the filling of the fish swim bladder. ~5.,6 As in the fish swim bladder case, further increases in the 02 tension can occur through salting out and similar affects.X6 These have all been neglected by the calculations above but would increase 02 tension to some degree, at least. Some indication that the Bohr shift may be important in the absence of accumulation of hypoxanthine is suggested by a similar type of capillary injury following heavy exercise. ~7 In this case, there is an accumulation of lactic acid without ischemia, and, thus, the hypoxia would be unlikely to be accompanied by the accumulation of hypoxanthine. Another case is diabetes in which capillary leakage and damage occur chronically, possibly due to the constant elevation of capillary oxygen tension resulting from the lactic acid accumulation induced by elevated glucose levels. ~8 Simple elevation of oxygen in the breathing mixture will increase free radical formation and is accompanied by vascular pathology that closely resembles that found in reperfusion injury. ~0 The increased oxygen tensions in these cases are similar to those used to calculate the data in Figure 1 and are based on tissue lactic acid accumulation probably below that found during is-
1. Fridovich, I. Hypoxia and oxygen toxicity. Adv. Neurol. 26:255 259; 1979. 2. McCord, J. M. Oxygen-derived radicals: a link between reperfusion injury and inflammation. Fed. Proc. 46:2402-2406: 1987. 3. Granger, D. N.; Rutili, G.; McCord, J. M. Superoxide radicals in feline intestinal ischemia. Gastroenterology 81:22-29; 1981. 4. Parks, D. A.; Bulkley, G. B.; Granger, D. N.: Hamilton, S. R.; McCord, J. M. Ischemic injury in the cat small intestine: role of superoxide radicals. Gastroenterology 82:9-15:1982. 5. Smith, S. M.; Grisham, M. B.; Manci, E. A.; Granger, D. N.; Kvietys, P. R. Gastric mucosal injury in the rat. Role of iron and xanthine oxidase. Gastroenterology 92:950-956; 1987. 6. Blasig, I. E.; Ebert, B.; Lowe, H. Identification of free radicals trapped during myocardial ischemia In vitro by ESR. Studia Biophysica 116:35-42: 1986. 7. Zweier, J. L.; Flaherty, J. T.: Weisfeldt, M. L. Direct measurement of free radicals generated following reperfusion of ischemic myocardium. Pro('. Natl. Acad. Sci. U.S.A. 84:1404 1407; 1987. 8. Kramer, J. H.; Arroyo, C. M.; Dickens, B. F.; Weglicki, W. B. Spin trapping evidence that graded myocardial ischemia alters post-ischemic superoxide production. Free Radio. Biol. Med. 3:153--159; 1987. 9. Fridovich, I. Quantativc aspects of the production of superoxide anion radical by xanthine oxidasc. J. Biol. Chem. 245:40534057; 1970. 10. Freeman, B. A.; Topolsky, M. K.; Crapo, J. D. Hyperoxia increases oxygen radical production in rat lung homogenates. Arch. Biochem. Biophys. 216:477-489; 1982. I 1. Boveris, A.: Chance, B. The mitochondrial generation of hydrogen peroxide. General properties and effect of hyperbaric oxygen. Biochem. J. 134:707-716; 1973. 12. Boveris, A. Mitochondrial production of superoxide radical and hydrogen peroxide. Advan. Exptl. Biol. Med. 78:67 82; 1977. 13. Bohr, C.; Hasselbach, K.; Krogh, A. Ueber eninen in biologischer beziehung wichtigen einfluss, den die kohlensaurespannung des blutes auf dessen sauerstoffbindung ubt. Skand. Arch. Physiol. 16:402-412; 1904. 14. Hermansen, L.; Osnes, J. Blood and muscle pH after maximal exercisc in man. J. Appl. Physiol. 32:304-308; 1972. 15. Scholander, P. F.; Van Dam, L. Secretion of gases against high pressure in the swim bladder of deep sea fishes. 1. Oxygen dissociation in blood. Biol. Bull. 207:247-259; 1954. 16. Gerth, W. A.; Hemmingson, E. A. Hints of gas secretion by the salting out effect in the fish swim bladder. J. Comp. Physiol. 146:129-136; 1982. 17. Davies, K. J. A.: Quintanilha, A. T.; Brooks, G. A.; Packer. L. Free radicals and tissue damage produced by exercise. Biochem. Biophys, Res. Comm. 107:1198-1205; 1982. 18. Kelton, J. G.; Collins, F. F. Exceptionally high arterial oxygen tension in diabetic ketoacidosis. Southern Med. J. 72:11271128; 1979. 19. Lumley, J.; Wood, C. Unexpected oxygen tensions in fetal acidosis. J. Perinat. Med. 1:166 173; 1973. 20. Halliwell, B. Oxidants and human disease: some new concepts. Fed. Amer. Soc. Exp. Biol. J. 1:358-364; 1987.