The Failure of Aerosolized Superoxide Dismutase to Modify Pulmonary Oxygen Toxicityl-3 JAMES D. CRAPO, DOUGLAS M. DeLONG, KAREN SJOSTROM, GALEN R. HASLER, and ROBERT T. DREW
SUMMARY _________________________________________________________ Superoxide (0 2 o) is a highly toxic free radical that may he an important component of pulmonary 0 2 toxicity. The primary defense against this free radical is superoxide dismutase. Rats were exposed to aerosolized superoxide dismutase, and it failed to modify either the time course or the cumulative toxicity of 100 per cent 0 2 . Because the aerosolized enzyme can be expected to be delivered only to the extracellular space of the lung, it is suggested that the primary site of production and of damage due to 0 2 -induced free radicals must be within the intracellular space.
Introduction Biologic production of increased numbers of free radicals may be an important component of pulmonary 0 2 toxicity. Gerschman and aswciates (I) proposed this as early as 1954 after noting the close correlation between the lung damage caused by radiation and that caused by 0 2 ; but mechanisms to study the specific free radicals that were believed to be contributing to 0 2 toxicity have not been available until recently. Oxygen acts primarily as an electron acceptor and not uncommonly is reduced by a series of univalent steps. The transfer of only one additional electron to 0 2 produces the superoxide anion (Oi) (2). A number of biologic systems can produce either univalently reduced (0 2 •) or divalently reduced (H 2 0 2) oxygen as one of their products, and the presence of high Po 2 can shift some of these reactions (Received in original form March I, 1976 and in revised form March 25, 1977)
From The Department of Medicine, Duke University, Durham, N. C. 27710, and the National Institute of Environmental Health Sciences, Research Triangle Park, N.C. 27709. 2 Supported in part by National Institutes of Health Grant HLI7603-0l. 3 Requests for reprints should be addressed to James D. Crapo, The Department of Medicine, Duke University, Durham, N.C. 27710. 1
toward a marked increase in 0 2• production (2). Supcroxide is highly reactive and can initiate destructive chain reactions. It now appears that superoxide is capable of reacting with H 2 0 2 to produce both the hydroxyl radical and singlet oxygen: 02-: + H202~ OH- + OH· +02*
(3).
These species are also extremely reactive and may be causing many of the uncontrolled oxidations associated with 0 2 toxicity. Catalase inactivates H 20 2 by a dismutation reaction and has been postulated for many years to be important in protecting against 0 2 toxicity (4). The discovery of the function of superoxide dismutase in 1969 identified another of the primary defense systems against the oxygen free radicals (5). Superoxide dismutase (SOD), a metalloprotein with a molecular weight of 32,000 daltons, catalyzes the dismutation of o2; according to the reaction: Superoxide dismutase has been postulated to Superoxide dismutase has been postulated to be essential for survival in 0 2 atmospheres (6). All aerobic plant and animal tissues studied so far have been found to contain superoxide dismutases, whereas strict anaerobic bacteria do not contain this enzyme (7). In 1974, Crapo and Tierney (8) reported the first studies in which superoxide dismutase
AMERICAt\ REVIEW OF RESPIRATORY DISEASE, VOLUME 115, 1977
1027
1028
CRAPO, DELO:>IG, SJOSTROM, HASLER, AND DREW
was shown to have a probable role in protecting a mammal against the toxic effects of exposure to high concentrations of 0 2 . Rats exposed to 85 per cent 0 2 for 5 to 7 days were shown to become tolerant to 100 per cent 0 2 and at the same time, to acquire a 50 per cent increase in superoxide dismutase in their lungs. Control animals died in I 00 per cent 0 2 after 60 to 72 hours of exposure (8). These studies led us to postulate that if sufficient amounts of superoxide dismutase could be delivered to the lung parenchyma during the initial 72 hours of exposure to 0 2 , the action of the enzyme might be able to modify the toxic effects of 0 2 enough to permit survival of those animals.
Materials and Methods Materials and equipment. Animals and exposures. The animals used were CD strain specific pathogenfree male rats obtained from Charles River Laboratories at 3 weeks of age. They were barrier sustained and fed Wayne Sterilizable Lab Blox in our animal facilities until the exposures were started. At the time of exposure, the rats were 10 weeks old and weighed 300 to 350 g. All exposures were continuous and were performed at 23 to 24° C in polystyrene chambers according to the method previously described (8). Superoxide dismutase administration. The first group of rats was treated with intraperitoneal superoxide dismutase using a dose of I mg per rat at the start of exposure to 100 per cent 0 2 and every 8 hours thereafter. This dose is approximately 225 times the average human clinical dose (0.04 rng per kg of body weight per day) being tested at the present time (9). The superoxide dismutase was prepared in 0.01 M potassium phosphate buffer (pH 7.4), and all control animals received intraperitoneal injections of the buffer at the same time intervals. The delivery of superoxide dismutase by aerosol was accomplished by dissolving I g of superoxide dismutase in 800 ml of 0.01 M potassium phosphate buffer (pH 7.4), and then generating an aerosol using a Laskin nebulizer and 100 per cent 0 2 at a flow of 7 liter per min. The resulting particles were sized aerodynamically with an Anderson Sampler. More than 80 per cent of the particles were found to be less than 0.5 ,urn in diameter. The average superoxide dismutase concentration in the chamber was 23 ,ug per liter of 0 2 , and the 0 2 concentration was maintained at more than 98 per cent during the entire exposure. Control animals were exposed to an aerosol containing only the phosphate buffer. The control aerosol was generated using the same Laskin nebulizer and 100 per cent 0 2 at a flow of 7 liter per min. Each exposure to either aerosolized superoxide dismutase or buffer alone was repeated twice. On the second set of exposures, the aerosol was started using air instead of 0 2 for 6 hours before the exposure to
0 2 was started. This was done so that bovine superoxide dismutase would be in the animals' lungs at the initiation of the exposure to 0 2 . Materials and assays. Spectrophotometric and agarose gel assays for superoxide dismutase activity were performed as previously described (10). Premade agarose gels were obtained from Corning ACI. After electrophoresis of crude homogenates on these gels, superoxide dismutase was identified using a stain containing nitroblue tetrazolium and a photochemical system for producing superoxide (10). In this stain, superoxide dismutase signals its presence by dismuting the superoxide radical and thereby preventing the reduction of nitroblue tetrazolium to blue formazan. This leaves a white band on an otherwise uniform blue background. The activity of superoxide dismutase can be roughly quantified on these gels by com paring each unknown to a series of standards electrophoresed and stained under identical conditions. Purified bovine superoxide dismutase (Orgatein®) with a specific activity of 3,000 units per mg (pH 7.8 cytochrome c assay) (10) was obtained from Diag· nostic Data, Inc., Mountain View, Ca. Tissue preparation. After appropriate exposures, rats were immediately anesthetized with intraperitoneal injections of 30 mg of pentobarbital, after which various organs were removed and homogenized; the supernatant fraction of each tissue specimen was prepared for enzyme assays as previously described (8).
Results The cumulative mortality of 20 rats exposed to 100 per cent 0 2 while being treated with intraperitoneal injections of either potassium phosphate buffer or superoxide dismutase in the potassium phosphate buffer is shown in table l. The cumulative mortality rates among 20 TABLE 1 EFFECTS OF INTRAPERITONEAL SUPEROXIDE DISMUTASE ON THE MORTALITY OF RATS EXPOSED TO 100 PER CENT 02 %Dead*
Time
(hours) 0 12 24 36 48 60 72
Control Ratst 0 0 0 0 0 30 100
Treated Rats • • 0 0 0 0 0 40 100
• N = 10 for each group. tTreated with 1 ml of 0.01 M 'potassium phosphate buffer (pH 7 .4) intraperitoneally, every 8 hours. **Treated with 1 mgofsuperoxidedismutase in 0.01 M potassium phosphate buffer (pH 7.4) intraperitoneally, every 8 hours.
1029
SUPEROXIDE DISMUTASE AND 0 2 TOXICITY
rats exposed to aerosolized superoxide dismutase in 100 per cent 0 2 and among 20 control rats exposed to aerosolized phosphate buffer in 100 per cent 0 2 are shown in figure I. There was no significant difference in either the time of death or the cumulative mortality between any of the exposed and control groupsBecause the bovine cuprozinc superoxide dismutase used in the treated group migrates differently on electrophoretic gels than does the endogenous rat cuprozinc superoxide dismutase, the activity of the endogenous enzyme and the aerosolized enzyme can be separately evaluated. Spectrophotometric assays and agarose gel assays of the lungs of rats treated with intraperitoneal injections of superoxide dismutase failed to demonstrate the presence of the bovine enzyme; however, aerosoliza tion of the enzyme was effective in delivering large amounts of the enzyme to the lung. Animals were sampled after 0.5, 1, 2, 6, and 24 hours of exposure, and the results are shown in figure 2- The bovine superoxide dismutase remained active after aerosolization, and its activity could be identified in the lung after 30 min, but did not reach equilibrium until after more than 6 hours of exposure (figure 2)- The multiple achromatic
Vl Q)
c
>
0
-'
e
Q)
L
Q.
..._
Q)
..._
0
__£
..._ Q)
0...
Fig. 2. The accumulation of aerosolized bovine superoxide dismutase in rat lung. The rats were exposed to 100 per cent 02 containing 23 !Jg of bovine superoxide dismutase per liter. All gels were stained for superoxide dismutase activity. The first 2 gels contained 0.01 /Jg of the pure rat and bovine enzymes. The remainder of the gels contained 1/10,000 of the total lung supernatant from each animal. The most peripheral and the perihi!ar portions of one rat lung were compared on a separate set of gels, and both the rat and bovine superoxide dismutase can be seen, although they did not migrate as far from the origin during the electrophoresis.
1030
CRAPO, DELO:-IG, SjOSTROM, HASLER, A:-ID DREW
TABLE 2 SPECTROPHOTOMETRIC ASSAYS OF SUPEROXIDE DISMUTASE IN RAT LUNGS AFTER EXPOSURE TO AEROSOLIZED BOVINE SUPEROXIDE DISMUTASE FOR 24 HOURS
Superoxide Dismutase, unitstt
Control Rats•
Control Ratst
Treated Rats • •
2,612±134
3,245 ± 122
5,369 ± 140* ••
For each group, n = 4. All data are mean ± SD. • Untreated control rats kept in air. tcontrol rats exposed to aerosol of 0.01 M potassium phosphate buffer (pH 7.4)
kept in 100 per cent Oz . • • Rats treated with an aerosol of superoxide dismutase in 0 .01 M potassium phos· phate buffer (pH 7.4) and kept in 100 per cent Oz. t t Units determined using 3-ml cytochrome c assay at pH 1 0.0 ( 10). ***P
SUMMARY _________________________________________________________ Superoxide (0 2 o) is a highly toxic free radical that may he an important component of pulmonary 0 2 toxicity. The primary defense against this free radical is superoxide dismutase. Rats were exposed to aerosolized superoxide dismutase, and it failed to modify either the time course or the cumulative toxicity of 100 per cent 0 2 . Because the aerosolized enzyme can be expected to be delivered only to the extracellular space of the lung, it is suggested that the primary site of production and of damage due to 0 2 -induced free radicals must be within the intracellular space.
Introduction Biologic production of increased numbers of free radicals may be an important component of pulmonary 0 2 toxicity. Gerschman and aswciates (I) proposed this as early as 1954 after noting the close correlation between the lung damage caused by radiation and that caused by 0 2 ; but mechanisms to study the specific free radicals that were believed to be contributing to 0 2 toxicity have not been available until recently. Oxygen acts primarily as an electron acceptor and not uncommonly is reduced by a series of univalent steps. The transfer of only one additional electron to 0 2 produces the superoxide anion (Oi) (2). A number of biologic systems can produce either univalently reduced (0 2 •) or divalently reduced (H 2 0 2) oxygen as one of their products, and the presence of high Po 2 can shift some of these reactions (Received in original form March I, 1976 and in revised form March 25, 1977)
From The Department of Medicine, Duke University, Durham, N. C. 27710, and the National Institute of Environmental Health Sciences, Research Triangle Park, N.C. 27709. 2 Supported in part by National Institutes of Health Grant HLI7603-0l. 3 Requests for reprints should be addressed to James D. Crapo, The Department of Medicine, Duke University, Durham, N.C. 27710. 1
toward a marked increase in 0 2• production (2). Supcroxide is highly reactive and can initiate destructive chain reactions. It now appears that superoxide is capable of reacting with H 2 0 2 to produce both the hydroxyl radical and singlet oxygen: 02-: + H202~ OH- + OH· +02*
(3).
These species are also extremely reactive and may be causing many of the uncontrolled oxidations associated with 0 2 toxicity. Catalase inactivates H 20 2 by a dismutation reaction and has been postulated for many years to be important in protecting against 0 2 toxicity (4). The discovery of the function of superoxide dismutase in 1969 identified another of the primary defense systems against the oxygen free radicals (5). Superoxide dismutase (SOD), a metalloprotein with a molecular weight of 32,000 daltons, catalyzes the dismutation of o2; according to the reaction: Superoxide dismutase has been postulated to Superoxide dismutase has been postulated to be essential for survival in 0 2 atmospheres (6). All aerobic plant and animal tissues studied so far have been found to contain superoxide dismutases, whereas strict anaerobic bacteria do not contain this enzyme (7). In 1974, Crapo and Tierney (8) reported the first studies in which superoxide dismutase
AMERICAt\ REVIEW OF RESPIRATORY DISEASE, VOLUME 115, 1977
1027
1028
CRAPO, DELO:>IG, SJOSTROM, HASLER, AND DREW
was shown to have a probable role in protecting a mammal against the toxic effects of exposure to high concentrations of 0 2 . Rats exposed to 85 per cent 0 2 for 5 to 7 days were shown to become tolerant to 100 per cent 0 2 and at the same time, to acquire a 50 per cent increase in superoxide dismutase in their lungs. Control animals died in I 00 per cent 0 2 after 60 to 72 hours of exposure (8). These studies led us to postulate that if sufficient amounts of superoxide dismutase could be delivered to the lung parenchyma during the initial 72 hours of exposure to 0 2 , the action of the enzyme might be able to modify the toxic effects of 0 2 enough to permit survival of those animals.
Materials and Methods Materials and equipment. Animals and exposures. The animals used were CD strain specific pathogenfree male rats obtained from Charles River Laboratories at 3 weeks of age. They were barrier sustained and fed Wayne Sterilizable Lab Blox in our animal facilities until the exposures were started. At the time of exposure, the rats were 10 weeks old and weighed 300 to 350 g. All exposures were continuous and were performed at 23 to 24° C in polystyrene chambers according to the method previously described (8). Superoxide dismutase administration. The first group of rats was treated with intraperitoneal superoxide dismutase using a dose of I mg per rat at the start of exposure to 100 per cent 0 2 and every 8 hours thereafter. This dose is approximately 225 times the average human clinical dose (0.04 rng per kg of body weight per day) being tested at the present time (9). The superoxide dismutase was prepared in 0.01 M potassium phosphate buffer (pH 7.4), and all control animals received intraperitoneal injections of the buffer at the same time intervals. The delivery of superoxide dismutase by aerosol was accomplished by dissolving I g of superoxide dismutase in 800 ml of 0.01 M potassium phosphate buffer (pH 7.4), and then generating an aerosol using a Laskin nebulizer and 100 per cent 0 2 at a flow of 7 liter per min. The resulting particles were sized aerodynamically with an Anderson Sampler. More than 80 per cent of the particles were found to be less than 0.5 ,urn in diameter. The average superoxide dismutase concentration in the chamber was 23 ,ug per liter of 0 2 , and the 0 2 concentration was maintained at more than 98 per cent during the entire exposure. Control animals were exposed to an aerosol containing only the phosphate buffer. The control aerosol was generated using the same Laskin nebulizer and 100 per cent 0 2 at a flow of 7 liter per min. Each exposure to either aerosolized superoxide dismutase or buffer alone was repeated twice. On the second set of exposures, the aerosol was started using air instead of 0 2 for 6 hours before the exposure to
0 2 was started. This was done so that bovine superoxide dismutase would be in the animals' lungs at the initiation of the exposure to 0 2 . Materials and assays. Spectrophotometric and agarose gel assays for superoxide dismutase activity were performed as previously described (10). Premade agarose gels were obtained from Corning ACI. After electrophoresis of crude homogenates on these gels, superoxide dismutase was identified using a stain containing nitroblue tetrazolium and a photochemical system for producing superoxide (10). In this stain, superoxide dismutase signals its presence by dismuting the superoxide radical and thereby preventing the reduction of nitroblue tetrazolium to blue formazan. This leaves a white band on an otherwise uniform blue background. The activity of superoxide dismutase can be roughly quantified on these gels by com paring each unknown to a series of standards electrophoresed and stained under identical conditions. Purified bovine superoxide dismutase (Orgatein®) with a specific activity of 3,000 units per mg (pH 7.8 cytochrome c assay) (10) was obtained from Diag· nostic Data, Inc., Mountain View, Ca. Tissue preparation. After appropriate exposures, rats were immediately anesthetized with intraperitoneal injections of 30 mg of pentobarbital, after which various organs were removed and homogenized; the supernatant fraction of each tissue specimen was prepared for enzyme assays as previously described (8).
Results The cumulative mortality of 20 rats exposed to 100 per cent 0 2 while being treated with intraperitoneal injections of either potassium phosphate buffer or superoxide dismutase in the potassium phosphate buffer is shown in table l. The cumulative mortality rates among 20 TABLE 1 EFFECTS OF INTRAPERITONEAL SUPEROXIDE DISMUTASE ON THE MORTALITY OF RATS EXPOSED TO 100 PER CENT 02 %Dead*
Time
(hours) 0 12 24 36 48 60 72
Control Ratst 0 0 0 0 0 30 100
Treated Rats • • 0 0 0 0 0 40 100
• N = 10 for each group. tTreated with 1 ml of 0.01 M 'potassium phosphate buffer (pH 7 .4) intraperitoneally, every 8 hours. **Treated with 1 mgofsuperoxidedismutase in 0.01 M potassium phosphate buffer (pH 7.4) intraperitoneally, every 8 hours.
1029
SUPEROXIDE DISMUTASE AND 0 2 TOXICITY
rats exposed to aerosolized superoxide dismutase in 100 per cent 0 2 and among 20 control rats exposed to aerosolized phosphate buffer in 100 per cent 0 2 are shown in figure I. There was no significant difference in either the time of death or the cumulative mortality between any of the exposed and control groupsBecause the bovine cuprozinc superoxide dismutase used in the treated group migrates differently on electrophoretic gels than does the endogenous rat cuprozinc superoxide dismutase, the activity of the endogenous enzyme and the aerosolized enzyme can be separately evaluated. Spectrophotometric assays and agarose gel assays of the lungs of rats treated with intraperitoneal injections of superoxide dismutase failed to demonstrate the presence of the bovine enzyme; however, aerosoliza tion of the enzyme was effective in delivering large amounts of the enzyme to the lung. Animals were sampled after 0.5, 1, 2, 6, and 24 hours of exposure, and the results are shown in figure 2- The bovine superoxide dismutase remained active after aerosolization, and its activity could be identified in the lung after 30 min, but did not reach equilibrium until after more than 6 hours of exposure (figure 2)- The multiple achromatic
Vl Q)
c
>
0
-'
e
Q)
L
Q.
..._
Q)
..._
0
__£
..._ Q)
0...
Fig. 2. The accumulation of aerosolized bovine superoxide dismutase in rat lung. The rats were exposed to 100 per cent 02 containing 23 !Jg of bovine superoxide dismutase per liter. All gels were stained for superoxide dismutase activity. The first 2 gels contained 0.01 /Jg of the pure rat and bovine enzymes. The remainder of the gels contained 1/10,000 of the total lung supernatant from each animal. The most peripheral and the perihi!ar portions of one rat lung were compared on a separate set of gels, and both the rat and bovine superoxide dismutase can be seen, although they did not migrate as far from the origin during the electrophoresis.
1030
CRAPO, DELO:-IG, SjOSTROM, HASLER, A:-ID DREW
TABLE 2 SPECTROPHOTOMETRIC ASSAYS OF SUPEROXIDE DISMUTASE IN RAT LUNGS AFTER EXPOSURE TO AEROSOLIZED BOVINE SUPEROXIDE DISMUTASE FOR 24 HOURS
Superoxide Dismutase, unitstt
Control Rats•
Control Ratst
Treated Rats • •
2,612±134
3,245 ± 122
5,369 ± 140* ••
For each group, n = 4. All data are mean ± SD. • Untreated control rats kept in air. tcontrol rats exposed to aerosol of 0.01 M potassium phosphate buffer (pH 7.4)
kept in 100 per cent Oz . • • Rats treated with an aerosol of superoxide dismutase in 0 .01 M potassium phos· phate buffer (pH 7.4) and kept in 100 per cent Oz. t t Units determined using 3-ml cytochrome c assay at pH 1 0.0 ( 10). ***P
