ARCHIVES
Vol.
OF BIOCHEMISTRY
285, No. 1, February
AND
BIOPHYSICS
15, pp. 83-89,
1991
Role of Iron, Hydrogen Species in Microsomal to Formaldehyde Liviu
A. Clejan
Department
Received
and Arthur
of Biochemistry,
August
Peroxide and Reactive Oxygen Oxidation of Glycerol
I. Cederbauml
The Mount
6, 1990, and in revised
form
Sinai School of Medicine
October
should Sinai 10029.
0003.9861/91 $3.00 Copyright 0 1991 by Academic Press, All rights of reproduction in any form
New York, New York 10029
29, 1990
Rat liver microsomes can oxidize glycerol to formaldehyde. This oxidation is sensitive to catalase and glutathione plus glutathione peroxidase, suggesting a requirement for HzOz in the overall pathway of glycerol oxidation. Hydrogen peroxide can not replace NADPH in supporting glycerol oxidation; however, added HzOz increased the NADPH-dependent rate. Ferric chloride or ferric-ATP had no effect on glycerol oxidation, whereas ferric-EDTA was inhibitory. Certain iron chelators such as desferrioxamine, EDTA or diethylenetriaminepentaacetic acid, but not others such as ADP or citrate, inhibited glycerol oxidation. The inhibition by desferrioxamine could be overcome by added iron. Neither superoxide dismutase nor hydroxyl radical scavengers had any effect on glycerol oxidation. With the exception of propyl gallate, several antioxidants which inhibit lipid peroxidation had no effect on formaldehyde production from glycerol. The inhibition by propyl gallate could be overcome by added iron. In contrast to glycerol, formaldehyde production from dimethylnitrosamine was not sensitive to catalase or iron chelators, thus disassociating the overall pathway of glycerol oxidation from typical mixed-function oxidase activity of cytochrome P450. These studies indicate that HzOz and nonheme iron are required for glycerol oxidation to formaldehyde. The responsible oxidant is not superoxide, HzOz, or hydroxyl radical. Cytochrome P450 may function to generate the HzOz and reduce the nonheme iron. There may be additional roles for P450 since rates of formaldehyde production by microsomes exceed rates found with model chemical systems. Elevated rates of HzOz production by certain P450 isozymes, e.g., P450 IIEl, may contribute to enhanced rates of glycerol oxidation. 0 1991 Academic Press. Inc.
‘To whom correspondence Biochemistry, Box 1020, Mount L. Levy Place, New York, NY
(CUNY),
be addressed at Department of School of Medicine, One Gustave
Glycerol was recently shown to inhibit the oxidation of substrates such as dimethylnitrosamine (1) and pyrazole (2) by rat liver microsomes. Microsomes were found to oxidize glycerol to a Nash-reactive product in a NADPH-dependent, carbon monoxide-sensitive reaction (3, 4). That the Nash-reactive material was indeed formaldehyde was validated by a specific glutathione-dependent formaldehyde dehydrogenase reaction. By utilizing [14C]glycerol, and coupling the formaldehyde and formate dehydrogenase reactions, 14COZcould be detected, indicating that the formaldehyde produced was derived from the added glycerol (4). Microsomes isolated from rats treated with inducers of cytochrome P450IIE1, e.g., pyrazole, ethanol, and acetone, oxidized glycerol to formaldehyde at rates that were two- to threefold elevated as compared to controls (4). These results suggested the possibility that glycerol may be an effective, although not a specific, substrate for cytochrome P450IIEl. Since rats treated with inducers of P450IIEl displayed elevated rates of oxygen radical generation (&-lo), we evaluated whether glycerol was oxidized to formaldehyde by hydroxyl radical ( * OH)-generating systems (4). Very low amounts of formaldehyde were produced from glycerol by the chemical and enzymatic . OH-generating systems (4). Experiments were carried out in the presence report to continue to investigate the possible role of reactive oxygen species in the pathway of microsomal oxidation of glycerol. Special attention was paid to the possible role of iron, since iron is an effective catalyst of the production of potent oxidizing species by microsomes (11-13). MATERIALS
AND
METHODS
Liver microsomes were isolated from male Sprague-Dawley rats weighing about 150 g and which were treated with pyrazole, 200 mg per kg body weight per day for 2 days. The rats were starved overnight prior to being killed. Microsomes were prepared by differential centrifugation, washed twice, resuspended in 0.125 M KCl-0.01 M potassium phosphate, pH 7.4, and stored at -70°C at a protein concentration of about 20 mg per milliliter. 83
Inc. reserved.
84
CLEJAN TABLE Effect
AND
I
of Catalase, Azide, and Glutathione on Microsomal Oxidation of Glycerol to Formaldehyde Rate
Addition
Concentration (units or mM)
A. None Catalase Catalase Catalase Catalase Catalase B. None Catalaseboiled Aside Catalase plus azide Catalase plus aside Glutathione Glutathione plus glutathione peroxidase
lu 10 u
25u 5ou 250~
of formaldehyde production (nmol/min/nmol P450)
6.09 5.67 3.52 2.43 1.36 0.43
+ i 2 + + iz
0.64 0.68 0.86 0.52 0.56 0.25
Effect of addition (%)
-7 -42" -60* -7a* -93*
1mM
5.47 5.47 7.38
0 +35
25 u, 1 mM
7.31
+34
50u, 1 mM 2mM
6.56 3.17
+20 -42
5u,2mM
0.27
-95
-
Note. The oxidation of 100 mM glycerol to formaldehyde was determined as described under Materials and Methods in the presence of the indicated additions. Azide was omitted unless specifically indicated. Results for experiment A are from four microsomal preparations, while results for experiment B are from two preparations. a P < 0.01
*P < 0.001
The oxidation of glycerol was determined at 37°C in a reaction system containing 0.1 M potassium phosphate, pH 7.4, 0.1 to 0.2 M glycerol, and microsomes equivalent to 1 nmol cytochrome P450 per milliliter of reaction in a final volume of 0.25 ml. Sodium azide (1 mM) was present in most experiments, except where catalase was added. Reactions were initiated by the addition of NADPH to a final concentration of 1.0 mM, and were terminated usually after 20 min by the addition of trichloroacetic acid (final concentration of 6% w/v). Formaldehyde was determined by the Nash procedure (14). All values were corrected for zerotime controls in which TCA was added prior to the NADPH. In some experiments, 4 mM dimethylnitrosamine replaced glycerol as substrate. The content of cytochrome P450 was determined by the method of Omura and Sato (15). All buffers and the water used to prepare solutions were passed through columns of Chelex 100 to remove iron. Results refer to means + SD and the number of microsomal preparations is indicated in the table legends. Statistical evaluation was carried out by Student’s t test. Some experiments were carried out with microsomes from two preparations; variability between the two experiments did not exceed 15%.
RESULTS
Isolated rat liver microsomes oxidized glycerol to formaldehyde as described previously (3, 4). The addition of catalase produced a concentration-dependent inhibition of glycerol oxidation (Table I). Sodium azide, an inhibitor of catalase, produced a small increase in glycerol oxidation
CEDERBAUM
by microsomes, suggesting the presence of some catalase in the isolated microsomes. By following the disappearance of HzOz at 240 nm, microsomes were found to contain catalase at a concentration of about 3 units per milligram protein. In the presence of sodium azide, concentrations of catalase which nearly completely inhibited the oxidation of glycerol were without any effect (Table I), indicating that the inhibition by catalase required enzymatic activity. Boiled catalase had no effect on glycerol oxidation. To inhibit the endogenous catalase, 1 mM azide was used in all subsequent experiments. Hydrogen peroxide generated by microsomes is primarily removed by the glutathione peroxidase system (16,17). Glutathione itself produced some inhibition of glycerol oxidation, whereas the combination of glutathione plus glutathione peroxidase was almost totally inhibitory (Table I). Glutathione peroxidase alone did not inhibit glycerol oxidation to formaldehyde. The inhibition by catalase and by glutathione plus glutathione peroxidase suggestsa role for H202 in the glycerol oxidation pathway. However, Hz02 in the absence of NADPH did not result in production of formaldehyde from glycerol (Table II, experiment A). In the presence of NADPH, added HzOz increased the rate of production of formaldehyde (Table II, experiment B), suggesting that HzOz may be a rate-limiting component for the overall TABLE
II
Effect of H,Oz and CumeneHydroperoxide on Microsomal Oxidation of Glycerol to Formaldehyde
NADPH (mM) A.
H,O, bM) 0
0 0 0 0 B.l 1 1 1
0.25 0.50 1.0 0 0
1 1 1 1
0 0 0
c.1
0 0 0 0.5 1.0
0 0.001 0.01
0.05 0.10 0.20 0.50
1
Cumene hydroperoxide (mM)
Rate of formaldehyde production (nmol/min/ nmol P450) 0 0 0 0 0.79
6.53 f 0.88
-
7.19 f 0.69
+11
8.83 + 0.80
0.5 1.0
+35
14.94 + 1.68
+129”
16.77 f 2.45
+157a +1450
15.99 f 3.69 18.23 + 4.12 0
Effect of HzOz or cumene hydroperoxide (%)
5.58 6.54 7.22
+1790 +17 +29
Note. The oxidation of glycerol by microsomes was determined in the presence of the indicated concentrations of HzOz or cumene hydroperoxide either in the absence (A) or in the presence (B and C) of NADPH. The incubation time for these experiments was reduced to 6 min. Results for experiment B are from four microsomal preparations, while results for experiments A and C are from two preparations. e P < 0.01.
ROLE
OF
OXYGEN
RADICALS
IN
MICROSOMAL
oxidation of glycerol. HzOz, at these concentrations, did not decrease the content of P450 nor convert P450 to P420 over a 6-min incubation period. NADPH did cause some loss (about 20%) of the P450 spectrum, most likely as a result of lipid peroxidation. The addition of 0.5 mM H202 had no effect on the decrease of total P450 produced by NADPH. Organic hydroperoxides such as cumene hydroperoxide react directly with microsomal cytochrome P450 to produce oxygenated complexes which are capable of oxidizing certain drug substrates or causing lipid peroxidation (18-20). Cumene hydroperoxide, in the absence of NADPH (Table II, experiment A), did not effectively catalyze the oxidation of glycerol to formaldehyde. In the presence of NADPH, cumene hydroperoxide caused a small increase in formaldehyde production (Table II, experiment C); the extent of increase was less than that produced by H202 (Table II). Isolated microsomes contain tightly bound nonheme iron (21-24). To calculate the nonheme iron concentration in our preparations, the microsomes were extracted with perchloric acid, the extract was neutralized, 2,2’-bipyridyl was added followed by dithionite, and the absorbance of the reduced ferrous bipyridyl complex determined at 520 nm. We obtained a range of 1 to 4 nmol per milligram microsomal protein for the nonheme content of our preparations. Desferrioxamine is a potent iron chelating agent which is very effective in inhibiting microsomal generation of reactive oxygen intermediates; at concentrations of less
TABLE
III
Effect of Chelators of Iron on Microsomal of Glycerol to Formaldehyde
Addition A. None Desferrioxamine Desferrioxamine Desferrioxamine Desferrioxamine Desferrioxamine B. None EDTA EDTA EDTA DTPA Citrate ADP ATP
Concentration (PM)
1 2 3 6 10 3 5 10 10 100 100 100
Oxidation
Rate of formaldehyde production (nmol/min/ nmol P450) 6.18 5.09 3.48 3.09 1.87 0.30 6.50 3.83 2.09 0.89 0.80 5.31 6.26 6.77
f + + ?I + + k f k k f f f 5
0.24 0.40 0.35 0.42 0.38 0.11 0.23 1.06 0.72 0.41 0.46 0.19 0.54 0.04
Note. The oxidation of 100 mM glycerol to formaldehyde mined in the presence of the indicated chelating agents. from four to six microsomal preparations. = P < 0.01. *P < 0.001.
Effect of addition (%) -18 -44” -50’ -70* -95* -41” -68* -86* -88b -18 -4 +4 was deterResults are
OXIDATION
OF
85
GLYCEROL TABLE
IV
Effect of Iron on Microsomal Oxidation to Formaldehyde
Concentration (PM)
Addition None Fe& FeCl, FeEDTA FeATP Desferrioxamine Desferrioxamine Desferrioxamine Desferrioxamine Desferrioxamine
+ + + +
FeCI, FeCl, FeCl, FeCl,
6 12 25 25 3 3 plus 3 3 plus 6 3 plus 12 3 plus 15
of Glycerol
Rate of formaldehyde production (nmol/min/ nmol P450) 10.3 12.8 13.2 2.3 9.4 1.0 2.6 6.9 8.9 9.8
Effect of addition (So)
+24 +28 -78 -9 -90 -75 -33 -14 -5
Note. The oxidation of glycerol was determined in the presence of the indicated additions. Ferric-EDTA was utilized as a 1:2 iron:chelator complex, while ferric-ATP was used in a 1:20 ratio. Results are from two experiments.
than 10 PM, desferrioxamine has been shown to inhibit oxidation of *OH scavengers by microsomes, lipid peroxidation, and microsomal chemiluminescence (13, 25). Desferrioxamine was very effective in inhibiting microsomal oxidation of glycerol to formaldehyde, with significant inhibition occurring at desferrioxamine concentrations of 2 PM (Table III). The strong inhibition by desferrioxamine suggests a role for iron in microsomal oxidation of glycerol. In microsomal systems, certain iron chelators have been shown to be effective in inhibiting one oxygen radicaldependent process, while stimulating a different process. For example, EDTA and diethylenetriaminepentaacetic acid (DTPA)2 inhibit microsomal lipid peroxidation and chemiluminescence but stimulate microsomal generation of . OH-like species (11-13). EDTA and DTPA at concentrations of 10 PM nearly completely inhibited the oxidation of glycerol (Table III). Other iron chelators such as citrate, ATP, or ADP did not inhibit glycerol oxidation (Table III). The experiments with different chelating agents suggested that although iron appears to be required in the overall pathway for glycerol oxidation, the chelated form of the iron is very important. Ferric-EDTA produced strong inhibition at a concentration which markedly increases microsomal * OH generation (Table IV). Similar inhibition was observed with ferric-DTPA (data not shown). However, ferric chloride or ferric-ATP did not inhibit glycerol oxidation. Indeed, the inhibition of glyc* Abbreviations SOD, superoxide
used: DTPA, dismutase.
diethylenetriaminepenta-acetic
acid:
86
CLEJAN TABLE
AND
V
Effect of SOD, Hydroxyl Radical Scavengers, and Antioxidants on Microsomal Oxidation of Glycerol to Formaldehyde
Addition None SOD SOD DMSO DMSO Mannitol Mannitol Propyl gallate Propyl gallate Propyl gallate Trolox Trolox Trolox BHT Vitamin E
Concentration (units or mM)
10 u 100 u 10 mM
50mM 10 mM 25 mM 0.001
mM
0.005mM 0.010
mM
0.01 mM
0.05mM 0.10 mM 0.10 mM
0.033mM
Rate of formaldehyde production (nmol/min/ nmol P450)
5.5 5.3 5.6 5.5 6.3 5.7 4.7 4.2 1.5 0.5 5.5 5.5 5.7 5.5 5.5
Effect of addition (%)
-3 +3 0 +14 +3 -15 -24 -73 -91 0 0
+3 0 0
Note. The oxidation of glycerol to formaldehyde was determined in the presence of the indicated additions. Results are from two experiments. The BHT and vitamin E were dissolved in ethanol, which by itself (100 mM) had no effect on oxidation of 200 mM glycerol.
erol oxidation by desferrioxamine could be completely overcome by ferric chloride, suggesting that the inhibition was related to chelation of iron present in the reaction system.
The inhibition of glycerol oxidation by EDTA, DTPA, and ferric-EDTA suggests that . OH does not play a significant role in glycerol oxidation. This was further substantiated by experiments showing that potent . OH scavengers such as mannitol and dimethylsulfoxide had little effect (
Vol.
OF BIOCHEMISTRY
285, No. 1, February
AND
BIOPHYSICS
15, pp. 83-89,
1991
Role of Iron, Hydrogen Species in Microsomal to Formaldehyde Liviu
A. Clejan
Department
Received
and Arthur
of Biochemistry,
August
Peroxide and Reactive Oxygen Oxidation of Glycerol
I. Cederbauml
The Mount
6, 1990, and in revised
form
Sinai School of Medicine
October
should Sinai 10029.
0003.9861/91 $3.00 Copyright 0 1991 by Academic Press, All rights of reproduction in any form
New York, New York 10029
29, 1990
Rat liver microsomes can oxidize glycerol to formaldehyde. This oxidation is sensitive to catalase and glutathione plus glutathione peroxidase, suggesting a requirement for HzOz in the overall pathway of glycerol oxidation. Hydrogen peroxide can not replace NADPH in supporting glycerol oxidation; however, added HzOz increased the NADPH-dependent rate. Ferric chloride or ferric-ATP had no effect on glycerol oxidation, whereas ferric-EDTA was inhibitory. Certain iron chelators such as desferrioxamine, EDTA or diethylenetriaminepentaacetic acid, but not others such as ADP or citrate, inhibited glycerol oxidation. The inhibition by desferrioxamine could be overcome by added iron. Neither superoxide dismutase nor hydroxyl radical scavengers had any effect on glycerol oxidation. With the exception of propyl gallate, several antioxidants which inhibit lipid peroxidation had no effect on formaldehyde production from glycerol. The inhibition by propyl gallate could be overcome by added iron. In contrast to glycerol, formaldehyde production from dimethylnitrosamine was not sensitive to catalase or iron chelators, thus disassociating the overall pathway of glycerol oxidation from typical mixed-function oxidase activity of cytochrome P450. These studies indicate that HzOz and nonheme iron are required for glycerol oxidation to formaldehyde. The responsible oxidant is not superoxide, HzOz, or hydroxyl radical. Cytochrome P450 may function to generate the HzOz and reduce the nonheme iron. There may be additional roles for P450 since rates of formaldehyde production by microsomes exceed rates found with model chemical systems. Elevated rates of HzOz production by certain P450 isozymes, e.g., P450 IIEl, may contribute to enhanced rates of glycerol oxidation. 0 1991 Academic Press. Inc.
‘To whom correspondence Biochemistry, Box 1020, Mount L. Levy Place, New York, NY
(CUNY),
be addressed at Department of School of Medicine, One Gustave
Glycerol was recently shown to inhibit the oxidation of substrates such as dimethylnitrosamine (1) and pyrazole (2) by rat liver microsomes. Microsomes were found to oxidize glycerol to a Nash-reactive product in a NADPH-dependent, carbon monoxide-sensitive reaction (3, 4). That the Nash-reactive material was indeed formaldehyde was validated by a specific glutathione-dependent formaldehyde dehydrogenase reaction. By utilizing [14C]glycerol, and coupling the formaldehyde and formate dehydrogenase reactions, 14COZcould be detected, indicating that the formaldehyde produced was derived from the added glycerol (4). Microsomes isolated from rats treated with inducers of cytochrome P450IIE1, e.g., pyrazole, ethanol, and acetone, oxidized glycerol to formaldehyde at rates that were two- to threefold elevated as compared to controls (4). These results suggested the possibility that glycerol may be an effective, although not a specific, substrate for cytochrome P450IIEl. Since rats treated with inducers of P450IIEl displayed elevated rates of oxygen radical generation (&-lo), we evaluated whether glycerol was oxidized to formaldehyde by hydroxyl radical ( * OH)-generating systems (4). Very low amounts of formaldehyde were produced from glycerol by the chemical and enzymatic . OH-generating systems (4). Experiments were carried out in the presence report to continue to investigate the possible role of reactive oxygen species in the pathway of microsomal oxidation of glycerol. Special attention was paid to the possible role of iron, since iron is an effective catalyst of the production of potent oxidizing species by microsomes (11-13). MATERIALS
AND
METHODS
Liver microsomes were isolated from male Sprague-Dawley rats weighing about 150 g and which were treated with pyrazole, 200 mg per kg body weight per day for 2 days. The rats were starved overnight prior to being killed. Microsomes were prepared by differential centrifugation, washed twice, resuspended in 0.125 M KCl-0.01 M potassium phosphate, pH 7.4, and stored at -70°C at a protein concentration of about 20 mg per milliliter. 83
Inc. reserved.
84
CLEJAN TABLE Effect
AND
I
of Catalase, Azide, and Glutathione on Microsomal Oxidation of Glycerol to Formaldehyde Rate
Addition
Concentration (units or mM)
A. None Catalase Catalase Catalase Catalase Catalase B. None Catalaseboiled Aside Catalase plus azide Catalase plus aside Glutathione Glutathione plus glutathione peroxidase
lu 10 u
25u 5ou 250~
of formaldehyde production (nmol/min/nmol P450)
6.09 5.67 3.52 2.43 1.36 0.43
+ i 2 + + iz
0.64 0.68 0.86 0.52 0.56 0.25
Effect of addition (%)
-7 -42" -60* -7a* -93*
1mM
5.47 5.47 7.38
0 +35
25 u, 1 mM
7.31
+34
50u, 1 mM 2mM
6.56 3.17
+20 -42
5u,2mM
0.27
-95
-
Note. The oxidation of 100 mM glycerol to formaldehyde was determined as described under Materials and Methods in the presence of the indicated additions. Azide was omitted unless specifically indicated. Results for experiment A are from four microsomal preparations, while results for experiment B are from two preparations. a P < 0.01
*P < 0.001
The oxidation of glycerol was determined at 37°C in a reaction system containing 0.1 M potassium phosphate, pH 7.4, 0.1 to 0.2 M glycerol, and microsomes equivalent to 1 nmol cytochrome P450 per milliliter of reaction in a final volume of 0.25 ml. Sodium azide (1 mM) was present in most experiments, except where catalase was added. Reactions were initiated by the addition of NADPH to a final concentration of 1.0 mM, and were terminated usually after 20 min by the addition of trichloroacetic acid (final concentration of 6% w/v). Formaldehyde was determined by the Nash procedure (14). All values were corrected for zerotime controls in which TCA was added prior to the NADPH. In some experiments, 4 mM dimethylnitrosamine replaced glycerol as substrate. The content of cytochrome P450 was determined by the method of Omura and Sato (15). All buffers and the water used to prepare solutions were passed through columns of Chelex 100 to remove iron. Results refer to means + SD and the number of microsomal preparations is indicated in the table legends. Statistical evaluation was carried out by Student’s t test. Some experiments were carried out with microsomes from two preparations; variability between the two experiments did not exceed 15%.
RESULTS
Isolated rat liver microsomes oxidized glycerol to formaldehyde as described previously (3, 4). The addition of catalase produced a concentration-dependent inhibition of glycerol oxidation (Table I). Sodium azide, an inhibitor of catalase, produced a small increase in glycerol oxidation
CEDERBAUM
by microsomes, suggesting the presence of some catalase in the isolated microsomes. By following the disappearance of HzOz at 240 nm, microsomes were found to contain catalase at a concentration of about 3 units per milligram protein. In the presence of sodium azide, concentrations of catalase which nearly completely inhibited the oxidation of glycerol were without any effect (Table I), indicating that the inhibition by catalase required enzymatic activity. Boiled catalase had no effect on glycerol oxidation. To inhibit the endogenous catalase, 1 mM azide was used in all subsequent experiments. Hydrogen peroxide generated by microsomes is primarily removed by the glutathione peroxidase system (16,17). Glutathione itself produced some inhibition of glycerol oxidation, whereas the combination of glutathione plus glutathione peroxidase was almost totally inhibitory (Table I). Glutathione peroxidase alone did not inhibit glycerol oxidation to formaldehyde. The inhibition by catalase and by glutathione plus glutathione peroxidase suggestsa role for H202 in the glycerol oxidation pathway. However, Hz02 in the absence of NADPH did not result in production of formaldehyde from glycerol (Table II, experiment A). In the presence of NADPH, added HzOz increased the rate of production of formaldehyde (Table II, experiment B), suggesting that HzOz may be a rate-limiting component for the overall TABLE
II
Effect of H,Oz and CumeneHydroperoxide on Microsomal Oxidation of Glycerol to Formaldehyde
NADPH (mM) A.
H,O, bM) 0
0 0 0 0 B.l 1 1 1
0.25 0.50 1.0 0 0
1 1 1 1
0 0 0
c.1
0 0 0 0.5 1.0
0 0.001 0.01
0.05 0.10 0.20 0.50
1
Cumene hydroperoxide (mM)
Rate of formaldehyde production (nmol/min/ nmol P450) 0 0 0 0 0.79
6.53 f 0.88
-
7.19 f 0.69
+11
8.83 + 0.80
0.5 1.0
+35
14.94 + 1.68
+129”
16.77 f 2.45
+157a +1450
15.99 f 3.69 18.23 + 4.12 0
Effect of HzOz or cumene hydroperoxide (%)
5.58 6.54 7.22
+1790 +17 +29
Note. The oxidation of glycerol by microsomes was determined in the presence of the indicated concentrations of HzOz or cumene hydroperoxide either in the absence (A) or in the presence (B and C) of NADPH. The incubation time for these experiments was reduced to 6 min. Results for experiment B are from four microsomal preparations, while results for experiments A and C are from two preparations. e P < 0.01.
ROLE
OF
OXYGEN
RADICALS
IN
MICROSOMAL
oxidation of glycerol. HzOz, at these concentrations, did not decrease the content of P450 nor convert P450 to P420 over a 6-min incubation period. NADPH did cause some loss (about 20%) of the P450 spectrum, most likely as a result of lipid peroxidation. The addition of 0.5 mM H202 had no effect on the decrease of total P450 produced by NADPH. Organic hydroperoxides such as cumene hydroperoxide react directly with microsomal cytochrome P450 to produce oxygenated complexes which are capable of oxidizing certain drug substrates or causing lipid peroxidation (18-20). Cumene hydroperoxide, in the absence of NADPH (Table II, experiment A), did not effectively catalyze the oxidation of glycerol to formaldehyde. In the presence of NADPH, cumene hydroperoxide caused a small increase in formaldehyde production (Table II, experiment C); the extent of increase was less than that produced by H202 (Table II). Isolated microsomes contain tightly bound nonheme iron (21-24). To calculate the nonheme iron concentration in our preparations, the microsomes were extracted with perchloric acid, the extract was neutralized, 2,2’-bipyridyl was added followed by dithionite, and the absorbance of the reduced ferrous bipyridyl complex determined at 520 nm. We obtained a range of 1 to 4 nmol per milligram microsomal protein for the nonheme content of our preparations. Desferrioxamine is a potent iron chelating agent which is very effective in inhibiting microsomal generation of reactive oxygen intermediates; at concentrations of less
TABLE
III
Effect of Chelators of Iron on Microsomal of Glycerol to Formaldehyde
Addition A. None Desferrioxamine Desferrioxamine Desferrioxamine Desferrioxamine Desferrioxamine B. None EDTA EDTA EDTA DTPA Citrate ADP ATP
Concentration (PM)
1 2 3 6 10 3 5 10 10 100 100 100
Oxidation
Rate of formaldehyde production (nmol/min/ nmol P450) 6.18 5.09 3.48 3.09 1.87 0.30 6.50 3.83 2.09 0.89 0.80 5.31 6.26 6.77
f + + ?I + + k f k k f f f 5
0.24 0.40 0.35 0.42 0.38 0.11 0.23 1.06 0.72 0.41 0.46 0.19 0.54 0.04
Note. The oxidation of 100 mM glycerol to formaldehyde mined in the presence of the indicated chelating agents. from four to six microsomal preparations. = P < 0.01. *P < 0.001.
Effect of addition (%) -18 -44” -50’ -70* -95* -41” -68* -86* -88b -18 -4 +4 was deterResults are
OXIDATION
OF
85
GLYCEROL TABLE
IV
Effect of Iron on Microsomal Oxidation to Formaldehyde
Concentration (PM)
Addition None Fe& FeCl, FeEDTA FeATP Desferrioxamine Desferrioxamine Desferrioxamine Desferrioxamine Desferrioxamine
+ + + +
FeCI, FeCl, FeCl, FeCl,
6 12 25 25 3 3 plus 3 3 plus 6 3 plus 12 3 plus 15
of Glycerol
Rate of formaldehyde production (nmol/min/ nmol P450) 10.3 12.8 13.2 2.3 9.4 1.0 2.6 6.9 8.9 9.8
Effect of addition (So)
+24 +28 -78 -9 -90 -75 -33 -14 -5
Note. The oxidation of glycerol was determined in the presence of the indicated additions. Ferric-EDTA was utilized as a 1:2 iron:chelator complex, while ferric-ATP was used in a 1:20 ratio. Results are from two experiments.
than 10 PM, desferrioxamine has been shown to inhibit oxidation of *OH scavengers by microsomes, lipid peroxidation, and microsomal chemiluminescence (13, 25). Desferrioxamine was very effective in inhibiting microsomal oxidation of glycerol to formaldehyde, with significant inhibition occurring at desferrioxamine concentrations of 2 PM (Table III). The strong inhibition by desferrioxamine suggests a role for iron in microsomal oxidation of glycerol. In microsomal systems, certain iron chelators have been shown to be effective in inhibiting one oxygen radicaldependent process, while stimulating a different process. For example, EDTA and diethylenetriaminepentaacetic acid (DTPA)2 inhibit microsomal lipid peroxidation and chemiluminescence but stimulate microsomal generation of . OH-like species (11-13). EDTA and DTPA at concentrations of 10 PM nearly completely inhibited the oxidation of glycerol (Table III). Other iron chelators such as citrate, ATP, or ADP did not inhibit glycerol oxidation (Table III). The experiments with different chelating agents suggested that although iron appears to be required in the overall pathway for glycerol oxidation, the chelated form of the iron is very important. Ferric-EDTA produced strong inhibition at a concentration which markedly increases microsomal * OH generation (Table IV). Similar inhibition was observed with ferric-DTPA (data not shown). However, ferric chloride or ferric-ATP did not inhibit glycerol oxidation. Indeed, the inhibition of glyc* Abbreviations SOD, superoxide
used: DTPA, dismutase.
diethylenetriaminepenta-acetic
acid:
86
CLEJAN TABLE
AND
V
Effect of SOD, Hydroxyl Radical Scavengers, and Antioxidants on Microsomal Oxidation of Glycerol to Formaldehyde
Addition None SOD SOD DMSO DMSO Mannitol Mannitol Propyl gallate Propyl gallate Propyl gallate Trolox Trolox Trolox BHT Vitamin E
Concentration (units or mM)
10 u 100 u 10 mM
50mM 10 mM 25 mM 0.001
mM
0.005mM 0.010
mM
0.01 mM
0.05mM 0.10 mM 0.10 mM
0.033mM
Rate of formaldehyde production (nmol/min/ nmol P450)
5.5 5.3 5.6 5.5 6.3 5.7 4.7 4.2 1.5 0.5 5.5 5.5 5.7 5.5 5.5
Effect of addition (%)
-3 +3 0 +14 +3 -15 -24 -73 -91 0 0
+3 0 0
Note. The oxidation of glycerol to formaldehyde was determined in the presence of the indicated additions. Results are from two experiments. The BHT and vitamin E were dissolved in ethanol, which by itself (100 mM) had no effect on oxidation of 200 mM glycerol.
erol oxidation by desferrioxamine could be completely overcome by ferric chloride, suggesting that the inhibition was related to chelation of iron present in the reaction system.
The inhibition of glycerol oxidation by EDTA, DTPA, and ferric-EDTA suggests that . OH does not play a significant role in glycerol oxidation. This was further substantiated by experiments showing that potent . OH scavengers such as mannitol and dimethylsulfoxide had little effect (
