Effect of d-cz-TocopherolAnalogues on Lipoxygenase-Dependent Peroxidation of Phospholipid-Bile Salt Micelles Hirofumi Arai a, Akihiko Nagao b, Junji Teraob'*, Tetsuya Suzuki a and Kozo Takama a aDepartment of Food Science and Technology, Faculty of Fisheries, Hokkaido University, Hokkaido 041, Japan and bNational Food ResearchInstitute, Ministry of Agriculture, Forestry and Fisheries, Ibaraki 305, Japan

In order to know whether or not vitamin E acts as an effective antioxidant in lipoxygenase-dependent peroxidation of phospholipids, the effect of vitamin E and vitamin E analogues, 2,2,5,7,8-pentamethyl-6-hydroxychroman (PMC) and 6-hydroxy-2,5,7,8-tetramethylchroman,2-carboxylic acid (Trolox C), was investigated in enzymatic lipid peroxidation of bile salt micelles of pig liver phosphatidylcboline (PC) using soybean lipoxygenase. 15-Hydroperoxy-5,8,11,13-eicosatetraenoic acid was exclusively produced by the reaction with the PC molecular species containing arachidonic acid moiety, indicating that the hydroperoxidation of pig liver PC entirely progresses through the enzymatic reaction. PMC suppressed the accumulation of PC-hydroperoxides (PC-OOH) more efficiently than either d-c~-tocopherol (cz-Toc)or Trolox C, and 50% inhibition concentration by PMC was close to that of quercetin, a known lipoxygenase inhibitorffomnatural origin. The antioxidant activity of PMC was also superior'to that of either oc-Tocor

ABSTRACT:

Trolox C in ferrous ion-induced nonenzymatic oxidation of PC micelles in the presence of a trace amount of PC-OOH, although the radical-scavenging activities of these compounds in solution were similar or comparable to one another. In conclusion, PMC is more effective than c~-Toc as an inhibitor of lipoxygenase reaction with phospholipids and of autoxidation in phospholipids. The phytyl chain of cc-Toc seems to be unfavorable for exerting an inhibitory effect on lipoxygenase reaction with phospholipid-bile salt micelles. Lipids 30, 135-140 (1995).

It has been suggested that lipid peroxidation is one of the significant pathophysiological factors in the oxidative stress leading to various diseases, including atherosclerosis (1,2). *To whom correspondenceshould be addressed at National Food Research Institute, MAFF, Kannondai,Tsukuba,lbaraki 305, Japan. Abbreviations:AMVN, 2,2'-azobis-(2,4-dimethylvaleronitrile);BHA, butylated hydroxyanisole[2(3)-tert-butyl-4-methoxyphenol]; 15-HPETE,15-hydroperoxy-5,8,11,13-eicosatetraenoicacid; HPLC,high-performanceliquid chromatography;IC50,50% inhibitionconcentrations;MeL-OOH,methyl linoleatehydroperoxides;PC, phosphatidylcholine;PC-OOH, phosphatidylcholine hydroperoxides;PMC, 2,2,5,7,8-pentamethyl-6-hydroxychroman; SA-PC, 1-stearoyl-2-arachidonylphosphatidylcholine;TBARS,thiobarbituric acid reactingsubstances;tx-Toc,d-oc-Tocopherol;TroloxC, 6-hydroxy2,5,7,8-tetramethylchroman-2-carboxylicacid; UV, ultraviolet. Copyright © 1995 by AOCS Press

Although a large number of studies have been carried out using a variety of model systems, the mechanism of lipid peroxidation occurring in vivo is still not entirely clarified, In particular, the mechanism of its initial reaction remains equivocal in many cases. In recent years, mammalian lipoxygenase (EC 1.13.11.12) has received much attention as a possibleini~ tiator for propagating lipid peroxidation in the process of oxidative modification of low-density lipoprotein (3) and biomembranes (4,5). Human tissues possess a well'organized defense system against oxidative stress (6), and d~cz-tocoplierol ((z-Toc) is considered toact as a primary antioxidant in the suppression of the propagation reaction of nonenzymatic lipid peroxidation by trappingchain-propagating lipid peroxyl radicals (7). On the~other-hand, it has also been~reported that o~Toc plays~ anqmportant role in the modulation of lipoxygenasepathway in arachidonic acid metabolism (8,9).Thus, it is o f much interest to know the antioxidant activity of ct-Toc against lipoxygenase-dependent lipid peroxidation, when this enzymatic reaction is involved in the initiation of propagating lipid peroxidation in vivo. We studied whether or not tx-Toc acts.as an effective inhibitor of the enzymatic peroxidation of unsaturated phospholipids using soybean lipoxygenase, because soybean lipoxygenase can oxidize phospholipids'directly in the pres.ence of bile salt (10,11). Also, phospholipids are recognized as a major target of oxidative damage, induced by lipid peroxidation. The roleof the phytyl side chain onthe antioxidant activity of o~-Toc has been frequently discussed in~peroxidation of membranous phospholipids (12-15), Thus, the inhibitory effects of tx-Toc and cz-Toc analogues [1], 2,2,5,7,8-pentamethyl-6-hydroxychroman ( P M C ) [ 2 ] and 6-hydroxy-2,5,7,8-tetramethylchroman-2-carboxylic acid (Trolox C) [3] (Scheme 1), on this enzymatic reaction system were compared to those of nonenzymatic ferrous ion-induced phospholipid peroxidation in order to understand therole of the phytyl side chain inthe inhibition. The results strongly suggest that an tx-T6c analogue, PMC, is superior to ct-Toc in the inhibition of lipoxygenase-dependent phospholipid peroxidation.

135

Lipidg, Vol. 30, no. 2 (1995)

136

H. ARAI ETAL. CH 3

which the mixture of methanol/water (95:5, vol/vol) was used as an eluting solvent. HPLC analysis of the products obtained from the lipoxyHaG" y " O " l V v v v v v OH3 genase reaction with SA-PC. SA-PC micelles without an/ OH3 GIla tioxidants were prepared and oxidized by soybean lipoxyge[1] nase for 60 min in the same condition described above. The lipid fraction was extracted from the reaction mixture and was CH3 OH 3 treated with phospholipase A2 from bee venom by the method of Hughes et al. (19). Then the extracted products were anaCH a COOH lyzed by reversed-phase HPLC with UV detection according H,C- ",ff - o H,C" "T" "O" to the method reported previously (20). Authentic 15-hyOH 3 OH 3 droperoxy-5,8,1 l, 13-eicosatetraenoic acid (15-HPETE) was prepared by enzymatic oxidation of arachidonic acid using [2] [3] soybean lipoxygenase (21). The isomeric mixture of HPETE SCHEME 1 was prepared from autoxidized arachidonic acid (20). Nonenzymatic oxidation of PC micelles induced by ferrous ion-ascorbate system. PC micelles with or without antioxidants were prepared by the same method as described in the MATERIALS A N D M E T H O D S preparation for lipoxygenase reaction, except for the addition Materials. Soybean lipoxygenase (Type I-B, 94,000 units/mg of PC-OOH before preparation of the micelles. The lipid perprotein) and 1-stearoyl-2-arachidonyl phosphatidylcholine oxidation was induced by the addition of FeSO4 dissolved in (SA-PC; 99%) were purchased from Sigma (St. Louis, MO). an aqueous solution of ascorbic acid. The assay mixture contx-Toc and PMC were supplied by Eisai Pharmaceutical Co. sisted of 1 mM PC, 0.0! mM PC-OOH, appropriate concen(Tokyo, Japan). Trolox C was purchased from Aldrich (Mil- tration of antioxidant, 0.1 mM FeSO4, 1 mM ascorbic acid waukee, WI). [3-Carotene was purchased from Merck (Darm- and 20 mM sodium deoxycholate in 20 mM HEPES buffer stadt, Germany). Butylated hydroxyanisole [2(3)-tert-butyl- (pH 7.4). The mixture was incubated with continuous shak4-methoxyphenol; BHA] and caffeic acid (3,4-dihydroxycin- ing at 37°C in the dark. At regular intervals, samples were namic acid) were purchased from Nacalai Tesque Inc. (Kyoto, subjected to thiobarbituric acid assay (22), and the amount of Japan). Quercetin was obtained from Wako Pure Chem. thiobarbituric acid reacting substances (TBARS) were ex(Osaka, Japan). Pig liver phosphatidylcholine (PC) was pur- pressed as p.M malondialdehyde. chased from Funakoshi Co. (Tokyo, Japan) and purified by Measurement of peroxyl radical-scavenging activity in soreversed-phase column chromatography (16). The fatty acid lution. Peroxyl radical-scavenging activities of ot-Toc, PMC composition of pig liver PC was determined by gas chro- and Trolox C in solution were determined according to the matography after transmethylation as follows: 16:0, 27.6%, method described previously (23). Briefly, 100 mM of methyl 18:0; 17.0%, 18:l; 27.5%, 18:2; 18.0%, 20:4; 7.7%. PC-hy- linoleate in a solution of n-hexane/2-propanol/ethanol (8:3:1, droperoxides (PC-OOH) were prepared from egg yolk PC by vol) containing each antioxidant was subjected to radical (Sigma; 99%) by photooxidation (17) and purified by re- chain oxidation by the addition of 10 mM of 2,2'-azobis-(2,4versed-phase column chromatography (16). All o/her chemi- dimethylvaleronitrile) (AMVN) at 37°C in the dark. The recals and solvents were of analytical grade. sulting methyl linoleate hydroperoxides (MeL-OOH) were Preparation of PC micelles. Pig liver PC (1 lxmol) in chlo- quantified by normal-phase HPLC technique (24). The rate of roform solution and an appropriate amount of antioxidant in oxidation during the induction period (ginh) and the rate of a methanol solution were mixed in a round-bottom test tube oxidation after the induction period, that is, the rate of propaand evaporated to a thin lipid film under nitrogen and finally gation (Rp), were obtained from the slope of the curve of in vacuo. The lipid film was rapidly dispersed in 900 [aL MeL-OOH formation. The apparent induction period (tinh) HEPES buffer (20 raM, pH 7.4) containing 20 mM sodium could be calculated from the intersecting point of the curve deoxycholate by vigorous shaking on a vortex mixer for for Rin h and that for Rp. We applied the method by Burton and 1 min and sonicated with a BRANSON B-12 ultrasonifier (55 Ingold (25) for the calculation of kinetic parameters of each kHz) for 1 min. antioxidant. Lipoxygenase reaction of PC micelles. Lipid peroxidation of PC micelles was induced by the addition of 2.8 × 104 units RESULTS of soybean lipoxygenase in 100 l.tL HEPES buffer (20 mM, pH 7.4). The mixture was incubated with continuous shaking Effect of t~-Toc and its analogues on lipoxygenase reaction. at 37°C in the dark. At regular intervals, PC-OOH in the mix- The bile salt micelles of PC were rapidly oxidized by soybean ture was determined by reversed-phase high-performance lipoxygenase, although neither its multilamellar liposomes liquid chromatography (HPLC) with ultraviolet (UV)detec- nor large unilamellar liposomes were oxidized after incubation according to the method reported previously (l 8), in tion for 2 h as previously reported by Eskola and Laakso (data

H°

A2

Lipids, Vol. 30, no. 2 (I 995)

2 AL?

INHIBITION OF LIPOXYGENASEBY VITAMIN E 0.8

137

100

_____---ff 0.6 A

8 -r

0.4

0a . 0.2

0.0

0

,

,

,

,

20

40

60

80

0

10 1

10 3

Concentration of antioxidant (vM)

Time (rain)

FIG. 1. Effect of a-Toc on lipoxygenase-dependent peroxidation of PC micelles. The concentration of PC-OOH was determined by reversedphase HPLC with ultraviolet detection (Ref. 18). The reaction system consisted of pig liver PC (1 raM), soybean lipoxygenase (2.8 x 104 units) and 20 mM sodium deoxycholate in 20 mM HEPES buffer (1.0 mL, pH 7.4). No addition (O), 0.1 mM o~-Toc (@). a-ToG d-a-tocopherol; PC, phosphatidylcholine; PC-OOH, phosphatidylcholine-hydroperoxides; HPLC, high-performance liquid chromatography.

are not shown here) (10). Figure 1 shows the time course of the accumulation of PC-OOH by the reaction of PC micelles with soybean lipoxygenase in the presence and the absence of a-Toc. In the absence of a-Toc, PC-OOH accumulation progressed rapidly in the initial stage and terminated after 40 min incubation because of the exhaustion of substrate. Both the rate of PC--OOH accumulation and its maximum amount were suppressed in the presence of 0.1 mM a-Toc. Figure 2 shows the relationship between the concentration of a-Toc or a-Toc analogues and the inhibition ratio of PC-OOH accumulation after 60 min incubation. The 50% inhibition concentrations (IC50) of a-Toc and PMC were 220 and 80 l.tM, respectively. Trolox C did not exert any inhibitory effect in the range from 80 I.tM to 1 mM. Inhibition ratio (%) of a-Toc, PMC and Trolox C were calculated from the following equation: inhibition ratio (%) = ([B] - [A])/[B] x I00

10 2

FIG. 2. The relationship between the inhibition by a-Toc (Q), 2,2,5,7,8pentamethyl-6-hydroxychroman (PMC) (A), and 6-hydroxy-2,5,7,8tetramethylchroman-2-carboxylicacid (Trolox C) (I-I) on lipoxygenase reaction of PC micelles and their concentrations. The reaction system was the same as that described in the legend in Figure 1. Abbreviations as in Figure 1.

with lipoxygenase by using SA-PC to confirm that the reaction proceeds through enzymatic mechanism. It is known that autoxidation of SA-PC produces several isomers of HPETE at the arachidonic acid moiety (26). Figure 3 shows HPLC chromatograms of authentic 15-HPETE (A), phospholipase A2-mediated hydrolyzed products from SA-PC after incubation with soybean lipoxygenase (B) and the isomeric mixture of HPETE from autoxidized arachidonic acid (C). The chromatogram of HPETE of SA-PC with lipoxygenase reaction was coincident with that of 15-HPETE. Thus, it was found that oxygen molecule exclusively attacked the 15-position of arachidonic acid moiety in the reaction of SA-PC with soybean lipoxygenase at neutral pH, similarly to the reaction at pH 9.0 reported by Brash et al. (11). Consequently, it seems that hydroperoxidation of pig liver PC in this reaction system also progresses through the enzymatic reaction.

Inhibition by ot-Toc and its analogues on ferrous ion-dependent peroxidation in micelles. We investigated the effect

[1]

where [A] and [B] are the concentration of PC--OOH after 60 min incubation in the presence of either a-Toc or a-Toc analogues (PMC and Trolox C) and that in the absence of them, respectively.

TABLE 1 The 50% Inhibition Concentrations (ICs0) of Antioxidants on Lipoxygenase Reaction of Phosphatidylcholine Micelles

Effect of lipophilic antioxidants on lipoxygenase reaction. The inhibitory effect of some lipophilic antioxidants on the lipoxygenase reaction was also measured by the same procedure shown in the case of a-Toc and its analogues (Table 1). The IC50 of butylated hydroxyanisole (BHA) and quercetin were comparable to that of PMC. The inhibition by caffeic acid was weaker than that of a-Toc, whereas B-carotene showed no inhibitory effect even at the concentration of 1 mM. Product specificity of lipoxygenase reaction. We investigated the product specificity of the reaction of PC micelles

d-a-Tocopherol PMC Trolox C

Antioxidants

ICso (IJ.M)a 220 80 no effect

[[]-Carotene no effect BHA 70 Quercetin 90 Caffeic acid 610 alCs0of each antioxidant was determined by the procedure shown in Figure 3. BHA, butylated hydroxyanisole; PMC, 2,2,5,7,8-pentamethyl-6-hydroxychroman; Trolox C, 6-hydroxy-2,5,7,8-tetramethylchroman-2-carboxylic acid.

Lipids, Vol. 30, no. 2 (1995)

138.

H. ARAI ETAL.

the inhibition ratio of TBARS accumulation and various concentrations of a-Toc or its analogues. IC50 of o~-Toc, PMC and Trolox C were 0.9, 0.3 and 7.4 ~tM, respectively. Inhibition ratio of a-Toc, PMC and Trolox C were calculated from the following equation:

E

1=

inhibition ratio (%) = ([B] - [A])I[B] x 100 0e-

[A]

where [A] and [B] are TBARS value after 36 h incubation in the presence of either a-Toc or a-Toc analogues (PMC and Trolox C) and that in the absence of them, respectively. Thus, the activity of these antioxidants during ferrous ion-dependent peroxidation of PC micelles was decreased in the order of PMC > a-Toc > Trolox C. Peroxyl radical-scavenging activity o f a-Toc and its analogues in solution. The peroxyl radical-scavenging activities of a-Toc and its analogues in solution were measured at their different concentrations. The ratio of the rate constant for the inhibition of oxidation during the induction period (kinh) to the rate constant for chain propagation (kp) can be determined by the following equation (27):

[B]

kinh/kp = [ZH]/Rin h • tinh

O

E t~ tw) (D

O_

<

E tc~

[c]

(D

o: c-

[2]

[3]

in which [LH] signifies the concentration of methyl linoleate. Figure 5 shows a plot of kinh/k p vs. concentrations of antioxidants ranging from 41.7 to 166.7 l.tM. The value of kinh/k p for each antioxidant decreased with the increase of concentration. However, the values of kinh/k p for a-Toc and PMC were n o t significantly different from each other (1500-2500) at each concentration. On the other hand, the values of kinh/kp for Trolox C were slightly lower (1000-2000) than those of a-Toc and PMC. Stoichiometric numbers of radicals (n) trapped by PMC and Trolox C, a function of tinh, were also obtained, indicating that a-Toc traps two chain-propagating lipid peroxyl radicals. The n values of PMC and Trolox C were calculated at different concentrations to be 1.8-2.1 and 1.9-2.3, respectively.

4h,.

o

100

r'l

<.

A

i

0

i

10

o 20

min FIG. 3, HPLC chromatograms of the 15-hydroperoxy-5,8,11,13eicosatetraenoic acid (HPETE) (A), phospholipase A2-mediated hydrolyzed products from 1-stearoyl-2-arachidonylphosphatidylcholine after incubation with,soybean lipoxygenase (B) and the isomeric mixture of HPETEfrom autoxidized arachidonic acid (C). Abbreviation as in Figure 1.

'-

50

.2 @ t-

0 10 -3'

of a-Toc and its analogues on the ferrous ion-dependent peroxidation of PC micelles to determine the effect of these antioxidants on~nonenzymatic lipid peroxidation in the phospholipid micelles. Figure 4 shows the relationship between

Lipids; Vol. 30, no. 2 (1995)

............. 10-2

~ -1 . . . . . . . ' 0 . . . . . . . . '1 . . . . . . . . ~ 2 . . . . . . . ~ 3 . . . . . . . 10 10 10 10 10 104

Concentration

of antioxidant

~M)

FIG. 4. The relationship between the inhibition by a-Toc (e), PMC (A) and Trolox C (E3)on ferrous ion ascorbate-dependentperoxidation of PC micelles and their concentrations. Abbreviations as in Figures 1 and 2.

INHIBITION OF LIPOXYGENASE BY VITAMIN E 3.0

,,~ -2.0

~o

:1.9

U

.

.

•

.

.

I

50

.

.

.

.

.

I

~100

"

"

•

•

I

.

.

150

.

.

290

Concentration of antioxidant (~M)FIG. 5. The concentration-dependent rate constant for the

inhibition of oxidation during the induction period and rate constant for chain propagation (kinh/kp) v a l u e s of c(-Tocand ct-Tocanalogues (PMC and Trolox C) on 2,2'-azobis-(2,4-dimethylvaleronitrile)-initiated oxidation of methyl linoleate in solution, a-Toc (0), PMC (A), TroloxC (E]).Abbreviations as in Figures 1 and 2. DISCUSSION

It has been suggested that lipoxygenase from plant and animal sources can oxidize directly fatty acids esterified in phospholipids (3-5,10,11,28-30). We selected a reaction system consisting of soybean lipoxygenase and pig liver PC in bile salt micelles to investigate the inhibitory effect of 0c-Toc and its analogues on the reaction of lipoxygenase with phospholipids. This simplified system will assist in studying the effect of 0c-Toc on lipoxygenase reaction in phospholipid bilayers. Although the Sigma enzyme preparation was possibly a mixture of isozymes, L- 1 probably predominates because it is the most stable of the isozymes. Because the 15-position of arachidonic acid moiety was exclusively oxidized from SA-PC, this demonstrates that isozyme L-1 is likely to be responsible for the hydroperoxidation because of its known specificity (21). It has been reported that cc-Toc serves as an effective inhibitor of the oxidation of free fatty acids by purified lipoxygenase from plant or animal sources (31-33). However, this study is the first to report that vitamin E inhibits lipoxygenase-dependent hydroperoxidation ofphospholipids as far as we know. In addition, the effectiveness of ~-Toc as lipoxygenase inhibitor for phospholipids has not been compared to that of PMC or Trolox C, although Panganamala et al. (34) have shown that 0t-Toc inhibits the lipoxygenase reaction more efficiently than Trotox C in the oxidation of free fatty acids in ethanol/water solution. Our results clearly show that PMC, an ~-Toc analogue possessing a methyl group instead of phytyl side chain, exerts an inhibitory effect more efficiently than 0c-Toc. The efficacy of the inhibition by PMC (IC50) was nearly the same as that of a synthetic antioxidant, BHA, and a known lipoxygenase inhibitor from natural origin, quercetin (35,36). Trolox C, another oc-Toc analogue possessing a carboxyl group instead of phytyl side chain, exerted no inhibitory effect even at

139

the same concentration as that of substrate PC. Interestingly, I~-carotenedid not act as lipoxygenase inhibitor in this reaction system although Lomniski et al. (37) pointed out that [~carotene inhibits lipoxygenase activity in the oxidation of free fatty acids. In general, the inhibition of lipoxygenase reaction seems to be derived from inactivation o f the active site of the enzyme or scavenging of free radicals at the active site. Niki et al. (13) concluded that the phytyl side daain do~s not affect the peroxyl radical-scavenging activity.hy showing that the rate constant for ~ inhibition:(kinh) by'PMC was-the san~e a~ that by o~-Toein the radical chain reaction of unsaturated lipids in solution. We.revealed thatXhe kinhlk p w~S'Ttearly the same for 0t-Toc and PMC at different concentrations. The ratios obtained from Trolox C were also comparable to those obtained from 0c-Toc and PMC. Stoichiometric numbers of radicals (n) trapped by PMC and Trolox C were also similar to that trapped by o~-Toc at different concentrations. Therefore, it is unlikely that their inherent radical-scavenging activities are the determinant for the effectiveness of inhibition on the lipoxygenase reaction of PC micelles. Thus, the localization of each compound in heterogeneous environment of the reaction system definitely affects the inhibitory effect. It seems that the localization of 0c-Toc, PMC and Trolox C in micelles is different from one another depending on the structure at their 2-positions. Fukuzawa et aL (12) and Niki et al. (13) have suggested that PMC has a higher motility than o~Toc in PC liposomes. It is also likely that PMC can move more freely than oc-Toc in the PC micelles resulting in the superiority of PMC in inhibiting the lipoxygenase reaction. On the other hand, our results demonstrated that Trolox C has no effect on the lipoxygenase reaction. Barclay and Vinqvist (38) indicated that Trolox C is negatively charged at neutral pH because the pKa of its carboxyl group was 3.89. Therefore, negatively charged Trolox C would not enter the negatively charged bile salt micelles because of the electrostatic repulsion, and little effect of Trolox C was observed on the lipoxygenase reaction. Once lipid hydroperoxides are formed by an enzymatic reaction, nonenzymatic chain oxidation may also occur through the generation of alkoxyl and peroxyl radicals from preformed hydroperoxides by the reaction with contaminant metal ions. It was observed that the antioxidant activity of PMC on the oxidation of PC micelles induced by ferrous ion in the presence of trace amount of PC-OOH was also superior to that of oc-Toc or Trolox C. It is therefore apparent that PMC is more effective than (x-Toc or Trolox C in the prevention of lipoxygenase-induced phospholipid peroxidation. Lipoxygenase-induced phospholipid peroxidation has attracted much attention in relation to the formation of oxidized low-density lipoprotein in the process of atherosclerosis (2,3) and the destruction of reticulocyte membrane structures in their maturing process (39,40). This study implies that PMC is a potent antioxidant of enzymatic peroxidation of phospholipids, although the micellar system in this experiment does not reflect directly the event of lipid peroxidation occurring in vivo.

Lipids, Vol. 30, no. 2 (1995)

140

H. ARAI ETAL.

REFERENCES 1. Halliwell, B., and Gutteridge, J.M.C. (1989) Free Radical in Biology and Medicine, 2nd edn., Clarendon Press, Oxford, pp. 299-365, 417-508. 2. Steinberg, D., Parthasarathy, S., Carew, T.E., Khoo, J.C., and Witztum, J.L. (1989) New Engl. J. Med. 320, 915-924. 3. Belkner, J., Wiesner, R., Rathman, J., Barnett, J., Sigal, E., and K0hn, H. (1993)Eur. J. Biochem. 213, 251-261. 4. Kiihn, H., Belkner, J., Wiesner, R., and Brash, A.R. (1990)J. Biol. Chem. 265, 18351-18361. 5. Takahashi, Y., Glasgow, W.C., Suzuki, H., Taketani, Y., Yamamoto, S., Anton, M., Kiihn, H., and Brash, A.R. (1993) Eur. J. Biochem. 218, 165-171. 6. Halliwell, B., and Gutteridge, J.M.C. (1989) Free Radical in Biology and Medicine, 2nd edn., Clarendon Press, Oxford, pp. 86-298. 7. Packer, L. (1992) in Lipid-Soluble Antioxidants: Biochemistry and Clinical Applications (Ong, A.S.H., and Packer, L., eds.) pp. 1-16, Birkhauser Verlag, Basel. 8. Gwebu, E.T., Trewyn, R.W., Cornwell, D.G., and Panganamala, R.V. (1980) Res. Commun. Chem. Pathol. Pharmacol. 28, 361-376. 9. Panganamala, R.V., and Cornwell, D.G. (1982)Ann. NYAcad. Sci. 393, 376-391. 10. Eskola, J., and Laakso, S. (1983) Biochim. Biophys. Acta 751, 305-311. 11. Brash, A.R., Ingram, C.D., and Harris, T.M. (1987) Biochemistry 26, 5465-5471. 12. Fukuzawa, K., Chida, H., Tokumura, A., and Tsukatani, H. (1981)Arch. Biochem. Biophys. 206, 173-180. I3. Niki, E., Kawakami, A., Saito, M., Yamamoto, Y., Tsuchiya, J., and Kamiya, Y. (1985) J. BioL Chem. 260, 2191-2196. 14. Niki, E., Takahashi, M., and Komuro, E. (1986) Chem. Lett. 1573-1576. 15. Takahashi, M., Tsuchiya, J., and Niki, E. (1989) J. Am. Chem. Soc. 111, 6350---6353. 16. Terao, J., Asano, I., and Matsushita, S. (1985) Lipids 20, 312-317. 17. Terao, J., and Matsushita, S. (1981) Agric. Biol. Chem. 45, 587-593. 18. Terao, J., Shibata, S.S., Yamada, K., and Matsushita, S. (1988) in Medical Biochemical and Chemical Aspects of Free Radicals (Hayashi, O., Niki, E., Kondo, M., and Yoshikawa, T., eds.) vol. II, pp. 781-788, Elsevier Science Publishers,B.V., Amsterdam. 19. Hughes, H., Smith, C.V., Homing, E.C., and Mitchell, J.R. (1983) Anal. Biochem. 130, 431-436.

Lipids, Vol. 30, no. 2 (1995)

20. Terao, J., Shibata, S.S., and Matsushita, S. (1988) Anal. Biochem. 169, 415-423. 21. Funk, M.O., Isaac, R., and Porter, N.A. (1976) Lipids 11, 113-117. 22. Uchiyama, M., and Mihara, M. (1978) Anal. Biochem. 86, 271-278. 23. Terao, J., Karasawa, H., Arai, H., Nagao, A., Suzuki, T., and Takama, K. (1993) Biosci. Biotechnol. Biochem. 57, 1204-1205. 24. Terao, J., and Matsushita, S. (1986) Lipids 21, 255-260. 25. Burton, G.W., and Ingold, K.U. (1981) J. Am. Chem. Soc. 103, 6472--6477. 26. Weenen, H., and Porter, N.A. (1982) J. Am. Chem. Soc. 104, 5216-5221. 27. Niki, E., Saito, T., Kawakami, A., and Kamiya, Y. (1984) J. Biol. Chem. 259, 4177-4182. 28. Jung, G., Yang, D-C., and Nakao, A. (1985) Biochem. Biophys. Res. Commun. 130, 559-566. 29. Murray, J.J., and Brash, A.R. (1988) Arch. Biochem. Biophys. 265, 514-523. 30. Schewe, T., Halangk, W., Hiebsch, Ch., and Rapoport, S.M. (1975) FEBS Lett. 60, 149-152. 31. Grossman, S., and Waksman, E.G. (1984) Int. J. Biochem. 16, 281-289. 32. Reddanna, P., Rao, M.K., and Reddy, C.C. (1985) FEBS Lett. 193, 39-43. 33. Bakalova, R.A., Nekrasov, A.S., Lankin, V.Z., Kagan, V.E., Stoichev, T.S., and Evstigneeva, R.P. (1988) Dokl. Akad. Nauk. SSSR 299, 1008-1011. 34. Panganamala, R.V., Miller, J.S., Gwebu, E.T., Sharma, H.M., and Cornwell, D.G. (1977) Prostaglandins 14, 261-271. 35. Baumann, J., Bruchhausen, F.V., and Wurm, G. (1980) Prostaglandins 20, 627-639. 36. Takahama, U. (1985)Phytochemistry24, 1443-1446. 37. Lomniski, L., Bar-Natan, R., Sklan, D., and Grossman, S. (1993) Biochim. Biophys. Acta 1167, 331-338. 38. Barclay, L.R.C., and Vinqvist, M.R. (1994) Free Radic. BioL Med. 16, 779-788. 39. Rapoport, S.M., and Schewe, T. (1986) Biochim. Biophys. Acta 864, 471-495. 40. Ktihn, H., Belkner, J., and Wiesner, R. (1990) Eur. J. Biochem. 191, 221-227.

[Received August 1, 1994, and in revised form December 8, 1994; Revision accepted December 23, 1994]