- Email: [email protected]
Effect of glutathione peroxidase activity on lipid peroxidation in biological membranes
459
Biochimica et Biophysics Acta, 431 (1976) 459-468 @ Elsevier Scientific Publishing Company, Amsterdam
- Printed
in The Netherlands
BBA 56795
EFFECT OF GLUTATHIONE PEROXIDASE ACTIVITY PEROXIDATION IN BIOLOGICAL MEMBRANES
a PAUL B. h&CAY, HORNBROOK
a DONALD
D. GIBSON,
a KUO-LAN
ON LIPID
FONG and b K. ROGER
a Biomembrane Research Laboratory, Oklahoma Medical Research Foundation, Oklahoma City, Okla. 73104 and b department of Pharmacology, University of Oklahoma Health Sciences &enter, Oklahoma City, Okfa. 73190 (U.S.A.) (Received September (Revised manuscript
8th, 1975) received March 4th, 1976)
Summary Results are presented indicating that, although glutathione peroxidase activity inhibits lipid peroxidation in membranes, it does not appear to do so by reducing membrane lipid peroxides to lipid alcohols, as has been shown by others to be the case for free fatty acid peroxides in solution. Lipid peroxidation was studied in an enzymic system (microsomal NADPH oxidase) and in a non-enzymic system (mitochondria plus ascorbate). A study of the fatty acids in -the phospholipids of microsomes and mitochondria demonstrated that detectable amounts of hydroxy fatty acids were not formed in the membranes when the latter were incubated in the presence of the glutathione peroxidase system even under conditions known to have generated significant levels of lipid peroxides in the membrane. Fatty acid analyses of the microsomal and mitochondrial particles indicated that glutathione peroxidase activity inhibited loss of polyun~turated fatty acids when these organelles were exposed to peroxidizing conditions. If glutathione peroxidase activity were inhibiting the formation of malondialdehyde (a product of lipid peroxidation) by converting peroxide groups to alcohols, the loss of the constitutive polyunsaturated fatty acids in the membrane should not have been appreciably affected by addition of the peroxidase system. The protective effect cannot be due to quenching of an autocatalytic type of lipid peroxidation (at least in the microsomal system) since it has been established that the microsomal enzyme system (NADPH oxidase) catalyzes a continuous attack on microsomal polyunsaturated fatty acyl groups during the reaction and that the peroxidative process is not autocatalytic in nature. It appears, therefore, that glutathione peroxidase activity must exert its effect on this system by preventing free radical attack on the polyunsaturated membrane lipids in the first place. A possible mechanism for the interruption of a free radical attack on the lipids is proposed.
460
Introduction The function of glutathione peroxidase in animal tissues has been suggested to include the reduction of endogenously formed peroxides of unsaturated fatty acids present in membranous portions of subcellular organelles [1,2], thereby preventing oxidative degradation of phospholipids. Christophersen has shown conclusively that hydroperoxides of polyunsaturated fatty acids in solution as salts are reduced to hydroxy derivatives by the glutathione peroxidase system [3] . Organic peroxides are also reduced [4]. Because the glutathione peroxidase system inhibits malondialdehyde formation in microsomal and mitochondrial preparations when the latter are subjected to conditions known to promote peroxidation of membrane lipids [ 2,5] , it has been considered that the inhibition occurred by conversion of the lipid hydroperoxide precursors of malondialdehyde to hydroxy fatty acids. This point, however, has not been tested directly. We have studied the effect of glutathione peroxidase system on both enzymic and non-enzymic lipid peroxidation in biological membranes. The results indicate that, while this system does indeed protect membrane unsaturated phospholipids from oxidative degradation, it does not appear to do so by reduction of lipid peroxides to hydroxy functions. Rather, the data indicate that the glutathione peroxidase system inhibits the initiation of peroxidative attacks on membrane lipids. In addition, the glutathione peroxidase system was shown not to catalyze the reduction of peroxidized unsaturated fatty acyl groups in phospholipid emulsions, suggesting either that phase differences interfere with the ability of the enzyme to act on the peroxide groups, or that steric hinderance may be involved. Materials and Methods Enzyme
preparations
Rat liver microsomes were prepared as described previously [6] . Some preparations of liver microsomes contained significant amounts of glutathione peroxidase activity, presumably in cytosolic material trapped within the vesicles since repeated washing does not remove it. The data shown were obtained with preparations which contained low levels of glutathione peroxidase activity. A study of the effect of preparation procedures on liver microsome glutathione peroxidase content is in progress. Rat liver mitochondria were prepared as follows. Livers from young male rats (180-250 g) were homogenized in 0.15 M potassium phosphate buffer, pH 7.5 containing 0.25 M sucrose (8.0 ml/g liver, wet weight). The homogenate was centrifuged at 350 X g for 10 min in a Sorvall RC 2-B centrifuge. The supernatant fraction was then collected and centrifuged at 8200 X g for 10 min. The pellet obtained was resuspended in the sucrose/phosphate buffer and centrifuged at 8200 X g for 10 min. The supernatant fraction was decanted and the tubes containing the mitochondrial pellets were resuspended in sucrose/ phosphate buffer as described below and used immediately. Dialyzed liver cell sap (used as a source of glutathione peroxidase) was prepared by homogenizing rat livers in 3 ~01s. of 0.1 M Tris * HCl buffer pH
461
7.5. The homogenate was centrifuged at 9500 X g for 15 min. The supernatant fraction was recovered and centrifuged at 105 000 X g for 90 min. The supernatant fraction from this step was then dialyzed overnight at 4°C against the Tris * HCl buffer. Purified glutathione peroxidase was a gift from Dr. H.E. Ganther of the University of Wisconsin, Madison, Wisconsin. The enzyme had been purified through step 4 according to the method of Oh et al. [7]. Incubation systems (1) Enzymatic lipid peroxidation. Microsomes were suspended in 0.15 M Tris buffer (pH 7.5) to give a suspension in which 1.0 ml contained particles derived from 1 g of liver. The incubation systems (10 ml) contained microsomes from 1 g of liver. Other additions are given as the concentration of that component in the final incubation mixture; 4 mM ADP, 1.2 * lo-’ M Fe3+, 5 mM glucose 6-phosphate, 0.3 mM NADP and 0.5 Kornberg units of glucose-6-phosphate dehydrogenaselml of reaction system. Where indicated, 10 mM glutathione and/or 4.0 ml of the cell sap preparation were added. The incubations were carried out under air in 50-ml glass beakers at 37°C for 30 min in a Dubnoff metabolic shaker. In some cases, the microsomal NADPH oxidase system was incubated for 10 min to allow maximum peroxide formation to occur. Then the glutathione and glutathione peroxidase preparations were added and incubated for an additional 45 min. (2) Non-enzymatic lipid peroxidation. Fresh mitochondrial pellets were resuspended in the sucrose/phosphate buffer, using a volume of buffer calculated to provide a suspension containing the mitochondria derived from 1.0 g of liver in 1.0 ml. The incubation system contained 3 ml of mitochondrial suspension (approximately 36 mg protein) and, where indicated, ascorbate (0.66 mM) and glutathione (3.6 mM). Final volume of all systems was 6 ml. The systems were incubated under air at 37°C for 60 min in a Dubnoff shaker. (3) Enzymic preparation of linoleic acid hydroperoxide. The preparation of purified linoleic acid hydroperoxide and incubation conditions and isolation of the reaction products for the enzymatic oxidation of GSH by hydroperoxide were performed as described by Christophersen [3]. Reaction products were analyzed by thin-layer chromatography as described in Fig. 2. Analytical procedures The protein content of the subcellular preparations was determined by the method of Lowry et al. [8]. At the end of each incubation 1.0 ml samples of the systems were taken for measurement of malondialdehyde formation [9]. For fatty acid analyses, the remainder of the incubation mixture was then placed in 20 ~01s. of chloroform/methanol (2 : 1, v/v) according to the method of Folch et al. [lo]. The lipids obtained were methylated by the boron trifluoride procedure and then chromatographed at 185°C on a Tracer 222 gas-liquid chromatograph using a 10% SP-222-PS column, with Nz as carrier at a flow rate of 40 ml/min. In some studies, the phospholipid fraction of the total lipid extract was isolated by silicic acid column chromatography. The total lipid extracts were placed on the columns (2.0 X 6.0 cm), in chloroform and washed with 40 ml of
chloroform to remove neutral lipids and then the phospholipids were eluted with 15 ml of methanol. Snake venom phospholipase was then employed to release P-position fatty acids from the phospholipid molecules (which contain most of the polyunsaturated fatty acids in these organelles) for detection of hydroxy fatty acids on thin-layer chromatographic plates. 20 mg of Baja naja cobra venom was dissolved in 0.5 ml of 0.005 M CaCl,. The venom solution (0.1 ml) was mixed with 1.5 ml anhydrous ether and approximately 15 pmol of phospholipids. This solution was then left standing overnight. 1.5 ml of absolute ethanol was then added to denature the enzyme and to allow the lysophospholipids to become soluble. The snake venom was removed by centrifugation and the lysophospholipid and @-position fatty acids were decanted, the solvent evaporated, and the lipids resuspended in a small amount of chloroform for transfer to thin-layer plates. The P-position fatty acids released (approx. 4.0 mg) were analysed for the presence of hydroxy fatty acids by thin-layer chromatography as described in the figure legends. Control studies with known hydroxy fatty acids indicated that the snake venom treatment in this manner caused no loss of these fatty acids derivatives. Results and grunion Glutathione peroxidase appears to play an important role in the protection of various cells from damage promoted by intracellular processes and, being a selenoenzyme [ 111, explains at least part of the dietary requirement for selenium. The relationship between o-tocopherol and selenium in protecting various laboratory animals from different types of tissue degeneration [ 12-151 led to suggestions that the role of the glutathione peroxidase system may be to remove lipid peroxides as they form and thus prevent lipid degradation and membrane disordering [12--141. The studies of Christophersen, and of Flohe and colleagues, mentioned above, have been cited in support of this concept. If glutathione peroxidase activity does protect biological membranes against peroxidative loss of unsaturated lipids by catalyzing the reduction of the peroxide functions to hydroxyl groups before fatty acid chain cleavage reactions occur, an accumulation of unsaturated hydroxy fatty acids should be observed in the membrane lipids subjected to conditions causing a continuous peroxidizing attack on the lipids. We have shown previously that significant losses of polyunsaturated fatty acids occur during NADPH oxidation by microsomes [6] and mitochondria [ 161, apparently as a result of peroxidation catalyzed by a free radical process initiated by the enzymic oxidation of NADPH. This type of peroxidative activity is not autocatalytic as was clearly shown by the observations that peroxidation stops immediately (1) on inactivating the microsomal NADPH oxidase system by warming of the microsomes to 65°C for 1 min, (2) by addition of ~-chloromercuribenzoate to the incubation system, or (3) on complete oxidation of all the NADPH when no NADPHgenerating system is added [ 61. None of these procedures inhibits autocatalytic lipid peroxidation. Furthermore, other studies in this laboratory have shown that significant amounts of lipid peroxides are generated in the microsomal membrane during at least the first 10-15 min oxidation of NADPH under the conditions described [ 1’71. It is very unlikely, therefore, that the action of
463
glutathione peroxidase in the microsomal system is due to interruption of an autocatalytic chain propagated process. That peroxide is formed continuously during the activity of this enzyme system was shown earlier by determination of the lipid peroxide content of liver microsomes at various times during NADPH oxidation. It was demonstrated that the peroxide content of the flposition fatty acids of microsomal phospholipid increases progressively to approx. 0.35 pmol/mg microsomal protein during the first 10 min of incubation and remained at that level for at least another 10 min while malondialdehyde accumulates [ 171. This indicates that lipid peroxides are forming and undergoing chain cleavage rapidly; a process that continues until the primary substrate (microsomal arachidonate) becomes limiting. Therefore, if the action of glutathione peroxidase resulted in effective conversion of peroxide groups to hydroxyl groups, changes in the unsaturated fatty acid composition of microsomal lipids undergoing peroxidation in the presence of the glutathione system should still be observed even though formation of fatty acid chain-cleavage products (such as malondialdehyde) would be inhibited. Table I shows that when microsomes are incubated in a reaction system that promotes peroxidation of the microsomal lipids, substantial loss of polyunsaturated fatty acids occurs. This was accompanied by significant malondialdehyde formation. The addition of either glutathione or dialyzed cell sap preparation alone had little effect on either fatty acid loss or malondialdehyde formation by these microsomal preparatiqns. However, when glutathione and cell sap are added to the microsomal system, both fatty acid loss and malondialdehyde formation were markedly inhibited. Since the fatty acid loss was inhibited in the presence of the glutathione peroxidase system, the data indicate that the latter system was preventing peroxidative attack rather than converting peroxides to hydroxy groups. This finding was supported by analyzing the P-position fatty acid groups of phospholipids of peroxidizing microsomes incubated with the glutathione peroxidase system. No trace of hydroxy fatty acids was detected on thin-layer analysis (Fig. 1, compare lanes 3, 5 and 6). If the unsaturated fatty acids (which would have been peroxidized in the reaction without the glutathione peroxidase system) had been converted to hydroxy fatty acids in this particular reaction system, one-fourth to one-third of the P-position fatty acids should have appeared as the hydroxy derivatives. These derivatives would have been easily detectable by the thin-layer analyses. The fact that they were not observed correlates with the observation that the glutathione peroxidase system inhibited the loss of polyunsaturated fatty acids from the membrane (Table I). It should be pointed out that incubation of microsomes with glutathione and ADP * Fe3’ in the absence of an NADPH-generating system produced no lipid peroxidation (data not shown). The experiment was also performed in another way. The microsomal NADPH oxidase system (5.0 ml final volume) described above was allowed to incubate for a period of 10 min to allow maximum formation of lipid peroxides in the membrane. At 10 min, both glutathione and cell sap were added to the system and incubation was continued for an additional 45 min. The total lipids were extracted and separated from the neutral lipids as described under Analytical procedures. The &position fatty acids were isolated and chromatographed on thin-layer plates (Fig. 1, lane 8). Only free fatty acid and lysophospholipid
ADP . Fe3+ ADP * Fe3+ ADP . Fe3+ ADP f Fe3+ ADP * Fe3”
+ + + +
NADPH-GS * * NADPH-GS + GSH NADPH-GS + GSH + CS *** NADPH-GS + CS -_---
EFFECT
OF GLUTATHIONE
k 57 i 79 + 63 + 64 i 107 .___~_..
1399 1185 1226 1399 117li
18 : 0
.-. i 81 * 144 ?1 93 i 82 81 -..
: 1
in pg.
i 53 i: 51 ‘_ 61 i 44 -t 68 i 64 ic 57 ?- 44 i: 44
i 25 +. 33 + 46 k 39 t 38 t 38 “I 36 i- 27 ?: 38
437 446 356 367 409 313 343 331 305
920 877 735 888 927 811 884 732 743
696 705 600 650 695 556 611 554 527 i 18 +_23 +_23 t 36 k 47 f 48 -r 57 + 19 +I41
:1
18
--.
18 : 0
16 : 0
ACID *
FATTY
426 ?: 48 399 f: 51 417% 26 607 + 39 692 ‘c 100
18
*
+ * t t +
:2 56 51 33 40 75
648 + 35 655 + 47 511+ 55 552 +_44 634 i 66 473 + 71 526 + 62 465t44 414 k 50
18 : 2
:4
:
4
1165t 1197 776 997 1178 899 1029 535 518
20
k 144 ‘- 137 * 83 k 95 i 101
: 6
3512 54 187 f. 48 169 zk29 344 ” 31 172 -i. 30
22
---__
56 + 63 +_ 50 i- 58 i 111 + 112 i 92 + 14 f 72
_-.-.
286 i- 24 279 k 24 127234 283 i 30 267 i 35 218 i. 38 248 i 18 169 + 33 94 + 26
:6 _-_22
+ ?: + i t
0.72 4.44 7.24 3.74 7.53
3.49 5.58 27.08 7.21 4.12 5.73 7.00 30.62 30.24
?: 0.81 f. 0.67 i_ 0.99 * 0.52 IO.94 +_0.72 + 1.36 + 1.08
c 0.69
Malondialdehyde formation (nmol)
5.12 81.89 58.64 20.97 66.35
M~ondi~dehyde formed (nmol)
POLYUNSATURATED
IN MITOCHONDRIA _.
1542 610 938 1498 937
20
OF MICROSOMAL
PEROXIDATION
534 373 454 688 687
18
.__ ~~__--.
PEROXIDATION
Mitochondrial fatty acid composition ____
ON POLYUNSATURATED
in M.
907 819 818 1060 1063 -_-
_ -
* Total mitochondriai fatty acids per incubation system (7 experiments),
Unincubated mitochondria Mitochondria alone Mitochondria + ascorbate Mitochondria + ascorbate + GSH Mitochondria + GSH Heated mitochondria Heated mitochondria + GSH Heated mitochondria + ascorbate Heated mitochondria + ascorbate + GSH
System
ASCORBATE-CATALYZED ---. -_
TABLE II
**
ENZYME-CALTALYZED
Microsomal fatty acid composition
ON
16 : 0 _-_-. ~.______
_.~ ~~~
SYSTEM
* Total microsomal fatty acids per incubation system. NADPH-generating system. * * * CeII sap.
+ + + + +
_ ._ ._
PEROXIDASE
~________~
GLUTATHIONE
-. _ _.--
OF THE
ACIDS
Microsomes Microsomes Microsomes Microsomes Microsomes ---_.-
System
FATTY
EFFECT
TABLE I
465
Fig. 1. Incubation systems (5 ml) were composed as described for enzymatic lipid peroxidation under Materials and Methods. 0.5-ml aliquots were removed for malondialdehyde determination at the end of a 55 min incubation period. Lipids were extracted from the remainder of the system and the phospholipi~ were treated with snake venom phospholipase A. The lipid products of the phospholipase A reaction were analyzed by thin-layer chromatography along with control systems and lipid markers as indicated. Solvent systems for thin-layer chromatograph: diethylether/hexane/glacial acetic acid (60 : 40 : 1, v/v/v). Lane 1. cholesterol, arachidonic acid; lane 2, lysolecithin, cholesterol paimitate;lane 3, microsomes + ADP *Fe’+: lane 4, ricinoleic acid; lane 5, microsomes + ADP . Fe3+ + NADPH-generating system; lane 6, microsomes + ADP . Fe 3+ + NADPH-generating system + GSH + cell sap; lane 7, S,lO-dihydroxystearic acid, ricinoleic acid ; lane 8, microsomes f ADP Fe3+ f NADPH-generating system incubated for 10 min then GSH + cell sap were added. The incubation was continued for an additionat 45 min. Lane 9, cell sap.
components were detected by either iodine vapor or rhodamine G spray. Since at 10 min the peroxide content of the 5.0 ml systems (containing 5 mg microsomal protein) was never less than about 1.7 pmol [ 171, the amount of hydroxy fatty acid which theoretically could have formed would be approx. 510 pg (based on the molecular weight of arachidonic acid, which is the primary substrate of this reaction [6] ). Even allowing for the possibility that each polyunsaturated fatty acid lost may have undergone two peroxidative alterations, approx. 255 pg of dihydroxy derivatives could be expected. The thin-layer analytical procedure employed would have detected far lesser amounts. As mentioned above, we have previously shown that the microsomal NADPH
466
oxidase system consistently forms at least 0.35 E.tmol of phospholipid peroxide/ mg microsomal protein (or more if an oxygenated system is used) in about 10 min [ 171. We extracted the total lipid from the microsomal system after 10 min of reaction time, suspended it as an emulsion, and incubated it with the glutathione peroxidase system according to the procedure of Christophersen [ 31. Then the unsaturated P-fatty acids were hydrolyzed and analyzed by thinlayer chromatography for the appearance of components not observed in flposition fatty acids of phospholipids from control systems in which no peroxidation occurred (no NADPH added). These studies employed two dimensional thin-layer chromatography to reduce the possibility that a hydroxy fatty acid might be masked by the small amount of cholesterol present. No difference in chromatographic patterns was observed indicating that the enzymatic glutathione peroxidase system apparently does not reduce peroxides of fatty acid acylated to a phospholipid. In view of this finding, it is not surprising that production of hydroxy fatty acids in phospholipids in peroxidizing membranes could not be observed. The source of glutathione peroxidase in these studies was dialyzed liver cell sap. It was shown that this preparation metabolized free fatty acid hydroperoxides in the same manner as a purified preparation of glutathione peroxidase (Fig. 2). Since all previous studies done by others on the glutathione peroxidase-dependent reduction of fatty acid peroxides to corresponding alcohols were done with liver cell sap preparations, these studies were carried out with the same type of preparation. To determine if the failure of the glutathione peroxidase system to form hydroxy fatty acids during enzyme-catalyzed peroxidation of membrane lipids was due to some interference with the activity of NADPH oxjdase system which promotes the peroxidation, we tested the effect of glutathione peroxidase activity on non-enzymic lipid peroxidation in mitochondria. In this system, ascorbate was utilized to promote the non-enzymic peroxidation. Table II shows that addition of glutathione resulted in inhibition of fatty acid loss and malondialdehyde formation. That this effect must require mitochondrial glutathione peroxidase activity was shown by the fact that addition of glutathione to systems containing ascorbate and heated mitochondria resulted in no inhibition of either fatty acid loss or malondialdehyde formation (Table II). The inhibition of non-enzymic lipid peroxidation in mitochondria required, therefore, a glutathione-dependent heat-labile component, presumably glutathione peroxidase, which is known to be present in this organelle [ 51. Earlier studies in this laboratory have indicated that initiation of measureable peroxidation of polyunsaturated fatty acids in biological membranes apparently involves the generation of free radicals (probably hydroxyl) via an interaction of HzOz and superoxide anion (both enzymically produced) with traces of inorganic iron [ 181. Removal of any one of these components is sufficient to inhibit radical formation. Inasmuch as the glutathione peroxidase system probably does not have access to the hydrophobic interior of biological membranes where the peroxidized fatty acyl chains would be situated and, in any case, does not appear to catalyze reduction of peroxidized phospholipids extracted from the microsomal membrane, we view the most likely mechanism of inhibition as being the reduction by the glutathione peroxidase system of
467
0
0
0
00
0
0
0
0
0
0
0
0
0
0
0
.
.
A
B
.
CC
.
D
.
.
.
.
.
E
F
G
H
I
Fig. 2. Thin-layer chroma~gr~s of the products formed by incubation of linoleic acid hydroperoxide with the glutathione peroxidase system. The incubation system and isolation of the reaction products are under Materials and Methods. The solvent system was heptane/diethyl ether/acetic acid (60 : 40 : 1. v/v/v). (A) linoleic acid; (B) linoleic hydroperoxide ; (C) linoleic hydroperoxide + glutathione; (D) linoleic hydroperoxide + cell sap; (E) linoleic hydroperoxide + glutathione + cell sap; (F) cell sap + glutathione; (G) linoleic hydroperoxide f purified glutathione peroxidase f glutathione; (H) linoleic hydroperoxide 4 purified glutathione peroxidase; (I) ricinoleic acid. The components migrating ahead of the free fatty acid soots are cholesterol ester and triacylglycerol present in the lipid extract of systems containing cell sap.
H202 formed in these reaction systems; via the enzymic oxidation of NADPH by microsomes in one case, and via the non-enzymic oxidation of ascorbate in the other. If H,Oz is the critical factor involved in enzyme catalyzed lipid peroxidation, catalase addition might be expected to inhibit lipid peroxidation. CataIase addition, even at high concentration does not affect lipid peroxidation in either of the systems described in this study. However, the K, of catalase for Hz02 is very high (1.10 M) 1191 and according to Misra is not an efficient scavenger of Hz02 [20] and would not be expected to affect the low steady state levels of H,Oz which may be involved. On the other hand, Flohc and Brand have reported an apparent I& of 1 PM for H,Oz in the glutathione peroxidase and the reaction would not become rate-limited until the concentration of HzOz falls to very low levels [21] . The results indicate that prevention of initiation of peroxidation of membrane lipids is a biological function of the glutathione peroxidase system rather than the conversion of peroxyl groups formed on polyunsaturated fatty acids to hydroxyl groups. Electron spin
468
resonance studies now being completed using spin-trapping agents have demonstrated that the formation of hydroxyl radicals by the NADH oxidase system is inhibited by the glutathione peroxidase system (Gibson, D.D. and McKay, P.B., unpublished). Acknowledgements The authors wish to thank Kristi Herring for her help in the preparation of this manuscript. This study was supported by grants AM 06978 and AM 08397 from the National Institutes of Health, USPHS. References 1 2 3 4 5 6 7 8 9 10 11
Christophersen, B.O. (1969) Biochim. Biophys. Acta 176.463-470 Flohe, L. and Zimmermann, R. (1970) Bioehim. Biophys. Acta 223.210-213 Christophersen, B.O. (1968) Biochim. Biophys. Acta 164.35-46 Little, C., OIinescu. R., Reid, K.G. and O’Brien, P.J. (1970) J. Biol. Chem. 245.3632-3636 Christophersen, B.O. (1968) Biochem. J. 106, 515-522 May. H.E. and McCay, P.B. (1968) J. Biol. Chem. 243, 2288-2295 Oh, S.-H., Ganther, H.E. and Hoekstra, W.G. (1974) Biochemistry 13,1825-1829 Lowry, O.H., Rosebrough, NJ., Farr, A.L. and Randall, R.J. (1951) J. Biol. Chem. lS3,265-275 Bernheim. F., Bernheim, ML. and Wilbur, K.M. (1948) J. Biol. Chem. 174.257-264 Folch, J., Lees, M. and Sloane-StanIey, G.H. (1967) J. Biol. Chem. 226,497-509 Rotrwk. J.T.. Pope, A.L., Ganther, H.E., Swanson, A.B., Hafeman, D.G. and Hoekstra, W.G. (1973)
12 13
Science 179. 588590 Noguchi, T.. Cantor. A.H. and Scott. M.L. (1973) J. Nutr. 103,1502-1511 Hoekstra, W.G. (1974) in Trace Element Metabolism in Animals-2 (Hoekatra,
W.G.,
Ganther, H.E. and Merty, W., eds.), P. 61-77, University Park Press, Baltimore 14 Tappel, A.L. (1974) Am. J. Clin. Nutr. 27.960-965 15 piccardO, M.G. and Schwarz, K. (1958) Symposium on Liver Function, Am. Inst. Biof. Washington, D.C. 16 Pfeiffer, P.M. and McCay, P.B. (1972) J. Biol. Chem. 247.6763-676s 17 Tam, B.K. and McCay, P.B. (1970) J. Biol. Chem. 245, 2295-2300 18 Fong.K.-L..McCay,P.B.,Poyer,J.L.,Keele. B.B. and Misra, H. (1973) 19 20 21
Ogura, Y. (1955) Arch. Biochem. Biophys. 57,288-300 Misra. H. (1974) J. Biol. Chem. 249.2151-2155 FlohB, L. and Brand, I. (1969) Biochim. Biophys. Acta 191,
541-549
Suttie,
si.
J.W.,
P. 528,
J. Biol. Chem. 248,7792-7797
Biochimica et Biophysics Acta, 431 (1976) 459-468 @ Elsevier Scientific Publishing Company, Amsterdam
- Printed
in The Netherlands
BBA 56795
EFFECT OF GLUTATHIONE PEROXIDASE ACTIVITY PEROXIDATION IN BIOLOGICAL MEMBRANES
a PAUL B. h&CAY, HORNBROOK
a DONALD
D. GIBSON,
a KUO-LAN
ON LIPID
FONG and b K. ROGER
a Biomembrane Research Laboratory, Oklahoma Medical Research Foundation, Oklahoma City, Okla. 73104 and b department of Pharmacology, University of Oklahoma Health Sciences &enter, Oklahoma City, Okfa. 73190 (U.S.A.) (Received September (Revised manuscript
8th, 1975) received March 4th, 1976)
Summary Results are presented indicating that, although glutathione peroxidase activity inhibits lipid peroxidation in membranes, it does not appear to do so by reducing membrane lipid peroxides to lipid alcohols, as has been shown by others to be the case for free fatty acid peroxides in solution. Lipid peroxidation was studied in an enzymic system (microsomal NADPH oxidase) and in a non-enzymic system (mitochondria plus ascorbate). A study of the fatty acids in -the phospholipids of microsomes and mitochondria demonstrated that detectable amounts of hydroxy fatty acids were not formed in the membranes when the latter were incubated in the presence of the glutathione peroxidase system even under conditions known to have generated significant levels of lipid peroxides in the membrane. Fatty acid analyses of the microsomal and mitochondrial particles indicated that glutathione peroxidase activity inhibited loss of polyun~turated fatty acids when these organelles were exposed to peroxidizing conditions. If glutathione peroxidase activity were inhibiting the formation of malondialdehyde (a product of lipid peroxidation) by converting peroxide groups to alcohols, the loss of the constitutive polyunsaturated fatty acids in the membrane should not have been appreciably affected by addition of the peroxidase system. The protective effect cannot be due to quenching of an autocatalytic type of lipid peroxidation (at least in the microsomal system) since it has been established that the microsomal enzyme system (NADPH oxidase) catalyzes a continuous attack on microsomal polyunsaturated fatty acyl groups during the reaction and that the peroxidative process is not autocatalytic in nature. It appears, therefore, that glutathione peroxidase activity must exert its effect on this system by preventing free radical attack on the polyunsaturated membrane lipids in the first place. A possible mechanism for the interruption of a free radical attack on the lipids is proposed.
460
Introduction The function of glutathione peroxidase in animal tissues has been suggested to include the reduction of endogenously formed peroxides of unsaturated fatty acids present in membranous portions of subcellular organelles [1,2], thereby preventing oxidative degradation of phospholipids. Christophersen has shown conclusively that hydroperoxides of polyunsaturated fatty acids in solution as salts are reduced to hydroxy derivatives by the glutathione peroxidase system [3] . Organic peroxides are also reduced [4]. Because the glutathione peroxidase system inhibits malondialdehyde formation in microsomal and mitochondrial preparations when the latter are subjected to conditions known to promote peroxidation of membrane lipids [ 2,5] , it has been considered that the inhibition occurred by conversion of the lipid hydroperoxide precursors of malondialdehyde to hydroxy fatty acids. This point, however, has not been tested directly. We have studied the effect of glutathione peroxidase system on both enzymic and non-enzymic lipid peroxidation in biological membranes. The results indicate that, while this system does indeed protect membrane unsaturated phospholipids from oxidative degradation, it does not appear to do so by reduction of lipid peroxides to hydroxy functions. Rather, the data indicate that the glutathione peroxidase system inhibits the initiation of peroxidative attacks on membrane lipids. In addition, the glutathione peroxidase system was shown not to catalyze the reduction of peroxidized unsaturated fatty acyl groups in phospholipid emulsions, suggesting either that phase differences interfere with the ability of the enzyme to act on the peroxide groups, or that steric hinderance may be involved. Materials and Methods Enzyme
preparations
Rat liver microsomes were prepared as described previously [6] . Some preparations of liver microsomes contained significant amounts of glutathione peroxidase activity, presumably in cytosolic material trapped within the vesicles since repeated washing does not remove it. The data shown were obtained with preparations which contained low levels of glutathione peroxidase activity. A study of the effect of preparation procedures on liver microsome glutathione peroxidase content is in progress. Rat liver mitochondria were prepared as follows. Livers from young male rats (180-250 g) were homogenized in 0.15 M potassium phosphate buffer, pH 7.5 containing 0.25 M sucrose (8.0 ml/g liver, wet weight). The homogenate was centrifuged at 350 X g for 10 min in a Sorvall RC 2-B centrifuge. The supernatant fraction was then collected and centrifuged at 8200 X g for 10 min. The pellet obtained was resuspended in the sucrose/phosphate buffer and centrifuged at 8200 X g for 10 min. The supernatant fraction was decanted and the tubes containing the mitochondrial pellets were resuspended in sucrose/ phosphate buffer as described below and used immediately. Dialyzed liver cell sap (used as a source of glutathione peroxidase) was prepared by homogenizing rat livers in 3 ~01s. of 0.1 M Tris * HCl buffer pH
461
7.5. The homogenate was centrifuged at 9500 X g for 15 min. The supernatant fraction was recovered and centrifuged at 105 000 X g for 90 min. The supernatant fraction from this step was then dialyzed overnight at 4°C against the Tris * HCl buffer. Purified glutathione peroxidase was a gift from Dr. H.E. Ganther of the University of Wisconsin, Madison, Wisconsin. The enzyme had been purified through step 4 according to the method of Oh et al. [7]. Incubation systems (1) Enzymatic lipid peroxidation. Microsomes were suspended in 0.15 M Tris buffer (pH 7.5) to give a suspension in which 1.0 ml contained particles derived from 1 g of liver. The incubation systems (10 ml) contained microsomes from 1 g of liver. Other additions are given as the concentration of that component in the final incubation mixture; 4 mM ADP, 1.2 * lo-’ M Fe3+, 5 mM glucose 6-phosphate, 0.3 mM NADP and 0.5 Kornberg units of glucose-6-phosphate dehydrogenaselml of reaction system. Where indicated, 10 mM glutathione and/or 4.0 ml of the cell sap preparation were added. The incubations were carried out under air in 50-ml glass beakers at 37°C for 30 min in a Dubnoff metabolic shaker. In some cases, the microsomal NADPH oxidase system was incubated for 10 min to allow maximum peroxide formation to occur. Then the glutathione and glutathione peroxidase preparations were added and incubated for an additional 45 min. (2) Non-enzymatic lipid peroxidation. Fresh mitochondrial pellets were resuspended in the sucrose/phosphate buffer, using a volume of buffer calculated to provide a suspension containing the mitochondria derived from 1.0 g of liver in 1.0 ml. The incubation system contained 3 ml of mitochondrial suspension (approximately 36 mg protein) and, where indicated, ascorbate (0.66 mM) and glutathione (3.6 mM). Final volume of all systems was 6 ml. The systems were incubated under air at 37°C for 60 min in a Dubnoff shaker. (3) Enzymic preparation of linoleic acid hydroperoxide. The preparation of purified linoleic acid hydroperoxide and incubation conditions and isolation of the reaction products for the enzymatic oxidation of GSH by hydroperoxide were performed as described by Christophersen [3]. Reaction products were analyzed by thin-layer chromatography as described in Fig. 2. Analytical procedures The protein content of the subcellular preparations was determined by the method of Lowry et al. [8]. At the end of each incubation 1.0 ml samples of the systems were taken for measurement of malondialdehyde formation [9]. For fatty acid analyses, the remainder of the incubation mixture was then placed in 20 ~01s. of chloroform/methanol (2 : 1, v/v) according to the method of Folch et al. [lo]. The lipids obtained were methylated by the boron trifluoride procedure and then chromatographed at 185°C on a Tracer 222 gas-liquid chromatograph using a 10% SP-222-PS column, with Nz as carrier at a flow rate of 40 ml/min. In some studies, the phospholipid fraction of the total lipid extract was isolated by silicic acid column chromatography. The total lipid extracts were placed on the columns (2.0 X 6.0 cm), in chloroform and washed with 40 ml of
chloroform to remove neutral lipids and then the phospholipids were eluted with 15 ml of methanol. Snake venom phospholipase was then employed to release P-position fatty acids from the phospholipid molecules (which contain most of the polyunsaturated fatty acids in these organelles) for detection of hydroxy fatty acids on thin-layer chromatographic plates. 20 mg of Baja naja cobra venom was dissolved in 0.5 ml of 0.005 M CaCl,. The venom solution (0.1 ml) was mixed with 1.5 ml anhydrous ether and approximately 15 pmol of phospholipids. This solution was then left standing overnight. 1.5 ml of absolute ethanol was then added to denature the enzyme and to allow the lysophospholipids to become soluble. The snake venom was removed by centrifugation and the lysophospholipid and @-position fatty acids were decanted, the solvent evaporated, and the lipids resuspended in a small amount of chloroform for transfer to thin-layer plates. The P-position fatty acids released (approx. 4.0 mg) were analysed for the presence of hydroxy fatty acids by thin-layer chromatography as described in the figure legends. Control studies with known hydroxy fatty acids indicated that the snake venom treatment in this manner caused no loss of these fatty acids derivatives. Results and grunion Glutathione peroxidase appears to play an important role in the protection of various cells from damage promoted by intracellular processes and, being a selenoenzyme [ 111, explains at least part of the dietary requirement for selenium. The relationship between o-tocopherol and selenium in protecting various laboratory animals from different types of tissue degeneration [ 12-151 led to suggestions that the role of the glutathione peroxidase system may be to remove lipid peroxides as they form and thus prevent lipid degradation and membrane disordering [12--141. The studies of Christophersen, and of Flohe and colleagues, mentioned above, have been cited in support of this concept. If glutathione peroxidase activity does protect biological membranes against peroxidative loss of unsaturated lipids by catalyzing the reduction of the peroxide functions to hydroxyl groups before fatty acid chain cleavage reactions occur, an accumulation of unsaturated hydroxy fatty acids should be observed in the membrane lipids subjected to conditions causing a continuous peroxidizing attack on the lipids. We have shown previously that significant losses of polyunsaturated fatty acids occur during NADPH oxidation by microsomes [6] and mitochondria [ 161, apparently as a result of peroxidation catalyzed by a free radical process initiated by the enzymic oxidation of NADPH. This type of peroxidative activity is not autocatalytic as was clearly shown by the observations that peroxidation stops immediately (1) on inactivating the microsomal NADPH oxidase system by warming of the microsomes to 65°C for 1 min, (2) by addition of ~-chloromercuribenzoate to the incubation system, or (3) on complete oxidation of all the NADPH when no NADPHgenerating system is added [ 61. None of these procedures inhibits autocatalytic lipid peroxidation. Furthermore, other studies in this laboratory have shown that significant amounts of lipid peroxides are generated in the microsomal membrane during at least the first 10-15 min oxidation of NADPH under the conditions described [ 1’71. It is very unlikely, therefore, that the action of
463
glutathione peroxidase in the microsomal system is due to interruption of an autocatalytic chain propagated process. That peroxide is formed continuously during the activity of this enzyme system was shown earlier by determination of the lipid peroxide content of liver microsomes at various times during NADPH oxidation. It was demonstrated that the peroxide content of the flposition fatty acids of microsomal phospholipid increases progressively to approx. 0.35 pmol/mg microsomal protein during the first 10 min of incubation and remained at that level for at least another 10 min while malondialdehyde accumulates [ 171. This indicates that lipid peroxides are forming and undergoing chain cleavage rapidly; a process that continues until the primary substrate (microsomal arachidonate) becomes limiting. Therefore, if the action of glutathione peroxidase resulted in effective conversion of peroxide groups to hydroxyl groups, changes in the unsaturated fatty acid composition of microsomal lipids undergoing peroxidation in the presence of the glutathione system should still be observed even though formation of fatty acid chain-cleavage products (such as malondialdehyde) would be inhibited. Table I shows that when microsomes are incubated in a reaction system that promotes peroxidation of the microsomal lipids, substantial loss of polyunsaturated fatty acids occurs. This was accompanied by significant malondialdehyde formation. The addition of either glutathione or dialyzed cell sap preparation alone had little effect on either fatty acid loss or malondialdehyde formation by these microsomal preparatiqns. However, when glutathione and cell sap are added to the microsomal system, both fatty acid loss and malondialdehyde formation were markedly inhibited. Since the fatty acid loss was inhibited in the presence of the glutathione peroxidase system, the data indicate that the latter system was preventing peroxidative attack rather than converting peroxides to hydroxy groups. This finding was supported by analyzing the P-position fatty acid groups of phospholipids of peroxidizing microsomes incubated with the glutathione peroxidase system. No trace of hydroxy fatty acids was detected on thin-layer analysis (Fig. 1, compare lanes 3, 5 and 6). If the unsaturated fatty acids (which would have been peroxidized in the reaction without the glutathione peroxidase system) had been converted to hydroxy fatty acids in this particular reaction system, one-fourth to one-third of the P-position fatty acids should have appeared as the hydroxy derivatives. These derivatives would have been easily detectable by the thin-layer analyses. The fact that they were not observed correlates with the observation that the glutathione peroxidase system inhibited the loss of polyunsaturated fatty acids from the membrane (Table I). It should be pointed out that incubation of microsomes with glutathione and ADP * Fe3’ in the absence of an NADPH-generating system produced no lipid peroxidation (data not shown). The experiment was also performed in another way. The microsomal NADPH oxidase system (5.0 ml final volume) described above was allowed to incubate for a period of 10 min to allow maximum formation of lipid peroxides in the membrane. At 10 min, both glutathione and cell sap were added to the system and incubation was continued for an additional 45 min. The total lipids were extracted and separated from the neutral lipids as described under Analytical procedures. The &position fatty acids were isolated and chromatographed on thin-layer plates (Fig. 1, lane 8). Only free fatty acid and lysophospholipid
ADP . Fe3+ ADP * Fe3+ ADP . Fe3+ ADP f Fe3+ ADP * Fe3”
+ + + +
NADPH-GS * * NADPH-GS + GSH NADPH-GS + GSH + CS *** NADPH-GS + CS -_---
EFFECT
OF GLUTATHIONE
k 57 i 79 + 63 + 64 i 107 .___~_..
1399 1185 1226 1399 117li
18 : 0
.-. i 81 * 144 ?1 93 i 82 81 -..
: 1
in pg.
i 53 i: 51 ‘_ 61 i 44 -t 68 i 64 ic 57 ?- 44 i: 44
i 25 +. 33 + 46 k 39 t 38 t 38 “I 36 i- 27 ?: 38
437 446 356 367 409 313 343 331 305
920 877 735 888 927 811 884 732 743
696 705 600 650 695 556 611 554 527 i 18 +_23 +_23 t 36 k 47 f 48 -r 57 + 19 +I41
:1
18
--.
18 : 0
16 : 0
ACID *
FATTY
426 ?: 48 399 f: 51 417% 26 607 + 39 692 ‘c 100
18
*
+ * t t +
:2 56 51 33 40 75
648 + 35 655 + 47 511+ 55 552 +_44 634 i 66 473 + 71 526 + 62 465t44 414 k 50
18 : 2
:4
:
4
1165t 1197 776 997 1178 899 1029 535 518
20
k 144 ‘- 137 * 83 k 95 i 101
: 6
3512 54 187 f. 48 169 zk29 344 ” 31 172 -i. 30
22
---__
56 + 63 +_ 50 i- 58 i 111 + 112 i 92 + 14 f 72
_-.-.
286 i- 24 279 k 24 127234 283 i 30 267 i 35 218 i. 38 248 i 18 169 + 33 94 + 26
:6 _-_22
+ ?: + i t
0.72 4.44 7.24 3.74 7.53
3.49 5.58 27.08 7.21 4.12 5.73 7.00 30.62 30.24
?: 0.81 f. 0.67 i_ 0.99 * 0.52 IO.94 +_0.72 + 1.36 + 1.08
c 0.69
Malondialdehyde formation (nmol)
5.12 81.89 58.64 20.97 66.35
M~ondi~dehyde formed (nmol)
POLYUNSATURATED
IN MITOCHONDRIA _.
1542 610 938 1498 937
20
OF MICROSOMAL
PEROXIDATION
534 373 454 688 687
18
.__ ~~__--.
PEROXIDATION
Mitochondrial fatty acid composition ____
ON POLYUNSATURATED
in M.
907 819 818 1060 1063 -_-
_ -
* Total mitochondriai fatty acids per incubation system (7 experiments),
Unincubated mitochondria Mitochondria alone Mitochondria + ascorbate Mitochondria + ascorbate + GSH Mitochondria + GSH Heated mitochondria Heated mitochondria + GSH Heated mitochondria + ascorbate Heated mitochondria + ascorbate + GSH
System
ASCORBATE-CATALYZED ---. -_
TABLE II
**
ENZYME-CALTALYZED
Microsomal fatty acid composition
ON
16 : 0 _-_-. ~.______
_.~ ~~~
SYSTEM
* Total microsomal fatty acids per incubation system. NADPH-generating system. * * * CeII sap.
+ + + + +
_ ._ ._
PEROXIDASE
~________~
GLUTATHIONE
-. _ _.--
OF THE
ACIDS
Microsomes Microsomes Microsomes Microsomes Microsomes ---_.-
System
FATTY
EFFECT
TABLE I
465
Fig. 1. Incubation systems (5 ml) were composed as described for enzymatic lipid peroxidation under Materials and Methods. 0.5-ml aliquots were removed for malondialdehyde determination at the end of a 55 min incubation period. Lipids were extracted from the remainder of the system and the phospholipi~ were treated with snake venom phospholipase A. The lipid products of the phospholipase A reaction were analyzed by thin-layer chromatography along with control systems and lipid markers as indicated. Solvent systems for thin-layer chromatograph: diethylether/hexane/glacial acetic acid (60 : 40 : 1, v/v/v). Lane 1. cholesterol, arachidonic acid; lane 2, lysolecithin, cholesterol paimitate;lane 3, microsomes + ADP *Fe’+: lane 4, ricinoleic acid; lane 5, microsomes + ADP . Fe3+ + NADPH-generating system; lane 6, microsomes + ADP . Fe 3+ + NADPH-generating system + GSH + cell sap; lane 7, S,lO-dihydroxystearic acid, ricinoleic acid ; lane 8, microsomes f ADP Fe3+ f NADPH-generating system incubated for 10 min then GSH + cell sap were added. The incubation was continued for an additionat 45 min. Lane 9, cell sap.
components were detected by either iodine vapor or rhodamine G spray. Since at 10 min the peroxide content of the 5.0 ml systems (containing 5 mg microsomal protein) was never less than about 1.7 pmol [ 171, the amount of hydroxy fatty acid which theoretically could have formed would be approx. 510 pg (based on the molecular weight of arachidonic acid, which is the primary substrate of this reaction [6] ). Even allowing for the possibility that each polyunsaturated fatty acid lost may have undergone two peroxidative alterations, approx. 255 pg of dihydroxy derivatives could be expected. The thin-layer analytical procedure employed would have detected far lesser amounts. As mentioned above, we have previously shown that the microsomal NADPH
466
oxidase system consistently forms at least 0.35 E.tmol of phospholipid peroxide/ mg microsomal protein (or more if an oxygenated system is used) in about 10 min [ 171. We extracted the total lipid from the microsomal system after 10 min of reaction time, suspended it as an emulsion, and incubated it with the glutathione peroxidase system according to the procedure of Christophersen [ 31. Then the unsaturated P-fatty acids were hydrolyzed and analyzed by thinlayer chromatography for the appearance of components not observed in flposition fatty acids of phospholipids from control systems in which no peroxidation occurred (no NADPH added). These studies employed two dimensional thin-layer chromatography to reduce the possibility that a hydroxy fatty acid might be masked by the small amount of cholesterol present. No difference in chromatographic patterns was observed indicating that the enzymatic glutathione peroxidase system apparently does not reduce peroxides of fatty acid acylated to a phospholipid. In view of this finding, it is not surprising that production of hydroxy fatty acids in phospholipids in peroxidizing membranes could not be observed. The source of glutathione peroxidase in these studies was dialyzed liver cell sap. It was shown that this preparation metabolized free fatty acid hydroperoxides in the same manner as a purified preparation of glutathione peroxidase (Fig. 2). Since all previous studies done by others on the glutathione peroxidase-dependent reduction of fatty acid peroxides to corresponding alcohols were done with liver cell sap preparations, these studies were carried out with the same type of preparation. To determine if the failure of the glutathione peroxidase system to form hydroxy fatty acids during enzyme-catalyzed peroxidation of membrane lipids was due to some interference with the activity of NADPH oxjdase system which promotes the peroxidation, we tested the effect of glutathione peroxidase activity on non-enzymic lipid peroxidation in mitochondria. In this system, ascorbate was utilized to promote the non-enzymic peroxidation. Table II shows that addition of glutathione resulted in inhibition of fatty acid loss and malondialdehyde formation. That this effect must require mitochondrial glutathione peroxidase activity was shown by the fact that addition of glutathione to systems containing ascorbate and heated mitochondria resulted in no inhibition of either fatty acid loss or malondialdehyde formation (Table II). The inhibition of non-enzymic lipid peroxidation in mitochondria required, therefore, a glutathione-dependent heat-labile component, presumably glutathione peroxidase, which is known to be present in this organelle [ 51. Earlier studies in this laboratory have indicated that initiation of measureable peroxidation of polyunsaturated fatty acids in biological membranes apparently involves the generation of free radicals (probably hydroxyl) via an interaction of HzOz and superoxide anion (both enzymically produced) with traces of inorganic iron [ 181. Removal of any one of these components is sufficient to inhibit radical formation. Inasmuch as the glutathione peroxidase system probably does not have access to the hydrophobic interior of biological membranes where the peroxidized fatty acyl chains would be situated and, in any case, does not appear to catalyze reduction of peroxidized phospholipids extracted from the microsomal membrane, we view the most likely mechanism of inhibition as being the reduction by the glutathione peroxidase system of
467
0
0
0
00
0
0
0
0
0
0
0
0
0
0
0
.
.
A
B
.
CC
.
D
.
.
.
.
.
E
F
G
H
I
Fig. 2. Thin-layer chroma~gr~s of the products formed by incubation of linoleic acid hydroperoxide with the glutathione peroxidase system. The incubation system and isolation of the reaction products are under Materials and Methods. The solvent system was heptane/diethyl ether/acetic acid (60 : 40 : 1. v/v/v). (A) linoleic acid; (B) linoleic hydroperoxide ; (C) linoleic hydroperoxide + glutathione; (D) linoleic hydroperoxide + cell sap; (E) linoleic hydroperoxide + glutathione + cell sap; (F) cell sap + glutathione; (G) linoleic hydroperoxide f purified glutathione peroxidase f glutathione; (H) linoleic hydroperoxide 4 purified glutathione peroxidase; (I) ricinoleic acid. The components migrating ahead of the free fatty acid soots are cholesterol ester and triacylglycerol present in the lipid extract of systems containing cell sap.
H202 formed in these reaction systems; via the enzymic oxidation of NADPH by microsomes in one case, and via the non-enzymic oxidation of ascorbate in the other. If H,Oz is the critical factor involved in enzyme catalyzed lipid peroxidation, catalase addition might be expected to inhibit lipid peroxidation. CataIase addition, even at high concentration does not affect lipid peroxidation in either of the systems described in this study. However, the K, of catalase for Hz02 is very high (1.10 M) 1191 and according to Misra is not an efficient scavenger of Hz02 [20] and would not be expected to affect the low steady state levels of H,Oz which may be involved. On the other hand, Flohc and Brand have reported an apparent I& of 1 PM for H,Oz in the glutathione peroxidase and the reaction would not become rate-limited until the concentration of HzOz falls to very low levels [21] . The results indicate that prevention of initiation of peroxidation of membrane lipids is a biological function of the glutathione peroxidase system rather than the conversion of peroxyl groups formed on polyunsaturated fatty acids to hydroxyl groups. Electron spin
468
resonance studies now being completed using spin-trapping agents have demonstrated that the formation of hydroxyl radicals by the NADH oxidase system is inhibited by the glutathione peroxidase system (Gibson, D.D. and McKay, P.B., unpublished). Acknowledgements The authors wish to thank Kristi Herring for her help in the preparation of this manuscript. This study was supported by grants AM 06978 and AM 08397 from the National Institutes of Health, USPHS. References 1 2 3 4 5 6 7 8 9 10 11
Christophersen, B.O. (1969) Biochim. Biophys. Acta 176.463-470 Flohe, L. and Zimmermann, R. (1970) Bioehim. Biophys. Acta 223.210-213 Christophersen, B.O. (1968) Biochim. Biophys. Acta 164.35-46 Little, C., OIinescu. R., Reid, K.G. and O’Brien, P.J. (1970) J. Biol. Chem. 245.3632-3636 Christophersen, B.O. (1968) Biochem. J. 106, 515-522 May. H.E. and McCay, P.B. (1968) J. Biol. Chem. 243, 2288-2295 Oh, S.-H., Ganther, H.E. and Hoekstra, W.G. (1974) Biochemistry 13,1825-1829 Lowry, O.H., Rosebrough, NJ., Farr, A.L. and Randall, R.J. (1951) J. Biol. Chem. lS3,265-275 Bernheim. F., Bernheim, ML. and Wilbur, K.M. (1948) J. Biol. Chem. 174.257-264 Folch, J., Lees, M. and Sloane-StanIey, G.H. (1967) J. Biol. Chem. 226,497-509 Rotrwk. J.T.. Pope, A.L., Ganther, H.E., Swanson, A.B., Hafeman, D.G. and Hoekstra, W.G. (1973)
12 13
Science 179. 588590 Noguchi, T.. Cantor. A.H. and Scott. M.L. (1973) J. Nutr. 103,1502-1511 Hoekstra, W.G. (1974) in Trace Element Metabolism in Animals-2 (Hoekatra,
W.G.,
Ganther, H.E. and Merty, W., eds.), P. 61-77, University Park Press, Baltimore 14 Tappel, A.L. (1974) Am. J. Clin. Nutr. 27.960-965 15 piccardO, M.G. and Schwarz, K. (1958) Symposium on Liver Function, Am. Inst. Biof. Washington, D.C. 16 Pfeiffer, P.M. and McCay, P.B. (1972) J. Biol. Chem. 247.6763-676s 17 Tam, B.K. and McCay, P.B. (1970) J. Biol. Chem. 245, 2295-2300 18 Fong.K.-L..McCay,P.B.,Poyer,J.L.,Keele. B.B. and Misra, H. (1973) 19 20 21
Ogura, Y. (1955) Arch. Biochem. Biophys. 57,288-300 Misra. H. (1974) J. Biol. Chem. 249.2151-2155 FlohB, L. and Brand, I. (1969) Biochim. Biophys. Acta 191,
541-549
Suttie,
si.
J.W.,
P. 528,
J. Biol. Chem. 248,7792-7797