- Email: [email protected]
Intermembrane transfer and antioxidant action of α-tocopherol in liposomes
ARCHIVES
OF BIOCHEMISTRY
AND
BIOPHYSICS
Vol. 280, No. 1, July, pp. 147-152,199O
Intermembrane Transfer and Antioxidant of a-Tocopherol in Liposomes
Action
V. E. Kagan,” R. A. Bakalova,” Zh. Zh. Zhelev,* D. S. Rangelova,* V. A. Tyurin,? N. K. Denisova,? and L. Packe#
E. A. Serbinova,*
*Institute of Physiology, Bulgarian Academy of Sciences, Sofia 1113, Bulgaria; tlnstitute of Evolutionary and Physiology, USSR Academy of Sciences, Leningrad 194223, USSR; and SDepartment of Molecular Cell Biology, 2.51 LSA, University of California, Berkeley, California 94720
Biochemistry
and
Received December 4,1989, and in revised form March 2,1SS0
Intermembrane transfer and exchange of tocopherol are not well understood. To study this we tested the ability of a-tocopherol containing unilamellar donor liposomes to inhibit the accumulation of lipid peroxidation products in acceptor liposomes. With molar ratios of cz-tocopherol:phospholipids from 1:lOO to 1:lOOO in donor liposomes prepared by sonication of lipid dispersions, a-tocopherol was incorporated into both monolayers and was homogenously distributed in monomeric form without forming clusters in the liposomes. Concentrations of a-tocopherol which completely prevented the peroxidation of lipids were chosen for donor liposomes. Hence inhibition of lipid peroxidation in mixtures of donor and acceptor liposomes was determined by the antioxidant effect of a-tocopherol in acceptor liposomes which resulted from intermembrane transfer and exchange of a-tocopherol. Evidence was obtained that this was not due to fusion of donor with acceptor liposomes. The efficiency of the “intermembrane” antioxidant action of tocopherol was more pronounced when donor liposomes contained unsaturated phospholipids, indicating that the presence of unsaturated fatty acids in the outer monolayer phospholipids 8 1990 facilitates intermembrane tocopherol exchange. Academic
Press,
Inc.
The regulation of free radical lipid peroxidation in eucaryotic membranes is predominantly dependent upon vitamin E, which is mainly in the form of a-tocopherol. w-Tocopherol interacts with lipid alkoxy and peroxy radicals, acting as a chain-breaking antioxidant (1, 2). In the lipid bilayer of membranes a-tocopherol is not uniformly distributed, but forms clusters and is prefer1 To whom correspondence
should be addressed.
0003.9861/90 $3.00 Copyright 0 1990 by Academic Press, All rights of reproduction in any form
entially concentrated in domains rich in polyenoic phospholipids (3,4). Oxidative stress is usually accompanied by a loss of vitamin E (5-7). Hence this may cause a localized antioxidant deficiency in the membrane. Thus domains rich in enzymatic generators of active oxygen species and lipid radicals might become tocopherol-deficient microenvironments. This “local E-hypovitaminosis” in the membrane might be overcome by the migration of tocopherols from regions with a high content of vitamin E to those poor in tocopherois. The efficiency of trans-bilayer tocopherol migration (“flip-flop”) is very low (8,9). Replenishment of E-deficient membrane domains might be brought about by either lateral diffusion or intermembrane transfer of tocopherols. Since the rate of the lateral diffusion of tocopherols is sufficiently high (10) this process may not be limiting. However opinions differ concerning the intermembrane transfer and exchange of tocopherols. It has been suggested that intermembrane exchange is only efficient in the presence of tocopherol-binding proteins (11-13). Existing information about lipid peroxidation inhibition in the case of intermembrane transfer of tocopher01s requires further elucidation. Niki et al. (14) and Nakagawa et al. (15) have found that donor liposomes containing unsaturated phospholipids and a-tocopherol do not influence the rate of lipidperoxidation in acceptor liposomes. However, Fukuzawa et al. (16) have demon-
strated a protective effect under similar conditions when the donor liposomes are obtained from saturated phospholipids. The present study was undertaken to clarify this discrepancy and to obtain further evidence for intermembrane exchange of tocopherols. To study this we used tocopherol-loaded (donor) and tocopherol-free (acceptor) liposomes under conditions where no membrane fu147
Inc. reserved.
148
KAGAN
ET AL. ascorbate
sion occurs. The antioxidant effects of cY-tocopherol were examined by incorporating it into unilamellar donor liposomes prepared from either saturated or unsaturated phospholipids, and monitoring its inhibition of lipid peroxidation in acceptor liposomes prepared from rat cerebral cortex lipids. EXPERIMENTAL
PROCEDURES
Liposomes. Total lipids from the cerebral cortex of male Wistar rats (150-180 g) were isolated by the method of Folch (17). Lipid fatty acid composition was determined by gas-liquid chromatography (Pye104) using a flame-ionization detector. Unilamellar liposomes were obtained bysonication (10 min, 22 kHz at 25°C for dimiristoylphosphatidylcholine and at 4°C for unsaturated lipids) of lipid dispersions (0.6 buffer (pH 7.4 at 25°C) until mg of lipids/ml) in 0.1 M K,Na-phosphate the suspension became clear. Sonication was performed under nitrogen stream. The amount of lipid peroxidation products in unsaturated lipids under conditions used did not exceed 0.9 nmol MDA*/mg lipids. Negatively stained electron micrographs showed that the average diameter of unilammelar liposomes was 30-40 nm. Incorporation of (Ytocopherol into liposomes was accomplished either by addition of the ethanol solution of ol-tocopherol to the liposome suspension or by dissolving lipids and o-tocopherol in chloroform, evaporating to dryness, and subsequent dispersion and sonication in K,Na-phosphate buffer, pH 7.4, as described above. Tocopherol distribution in liposomes. The incorporation and monomeric distribution of a-tocopherol in liposomes were estimated using a fluorescence method (3). This method is based on an increase of fluorescence intensity of a-tocopherol (excitation = 292 nm, emission = 325 nm) which occurs when there is a decrease in local a-tocopherol concentration. This increase of fluorescence intensity is due to elimination of tocopherol concentration-dependent fluorescence selfquenching. The maximum incorporation and the fully monomeric distribution of a-tocopherol in liposome bilayers was determined by the fluorescence intensity obtained in the presence of the detergent, sodium desoxycholate (at a concentration of 25 mM, which exceeded the critical micellar concentration). In the presence of detergent, a-tocopherol is distributed uniformly in mixed detergent-lipid micelles. Lipid peroxidation in liposomes was Lipid peroxidation products. induced by the Fe*+ + ascorbate oxidation system. Incubation medium contained: 0.1 M K,Na-phosphate buffer, pH 7.4 at 37°C 80 pM Fe’+, 0.5 mM ascorbate, 0.6 mg lipid/ml. The reaction was started by simultaneous addition of Fe’+ and ascorbate. The reaction mixture was shaken in air at 37°C. Lipid peroxidation products were determined by reaction with 2-thiobarbituric acid (TBA) and measured spectrophotometrically at 535 nm (E = 1.56 X lo5 M-’ cm-‘) (18). Lipid peroxidation induction was performed in mixtures of two types of liposomes: donor liposomes containing a-tocopherol and acceptor liposomes containing no oc-tocopherol. Donor and acceptor liposomes were preincubated for the time periods described in the legends to the figures from 1 to 60 min at 37°C. Test for fusion of donor and acceptor liposomes. To assess fusion of donor and acceptor liposomes, we obtained donor liposomes containing a-tocopherol only in the inner membrane monolayer and acceptor liposomes with ferricyanide trapped inside. For these experiments liposomes containing ferricyanide were prepared by sonication (10 min at 22 kHz) of lipid dispersions (2.5 mg/ ml) in the presence of ferricyanide (0.5 M). The suspension was then applied onto a column filled with Sephadex G-25, which was presaturated by a mixture of lipids and ferricyanide, and was chromatographed in 0.1 M K,Na-phosphate buffer at pH 7.4. The spectrophoto2 Abbreviations used: TBA, 2-thiobarbituric acid; DMPC, toylphosphatidylcholine; MDA, malondialdehyde.
dimiris-
0
1
2
Time ecurs)
FIG. 1. Changes in the absorbance of potassium ferricyanide in unilamellar liposomes prepared from cerebral cortex lipids after addition of ascorbate and subsequent sonication (22 kHz, 10 s). Incubation medium: liposomes, 0.65 mg lipids/ml; potassium ferricyanide inside liposomes or absorbed on the surface of liposomes, 0.33 pmol/mg lipids; ascorbate, 1.0 mM; 0.1 M K,Na-phosphate buffer, pH 7.4, at 37°C. There were four replicates of each measurement. metric determination of ferricyanide at 418 nm (3) revealed two fractions: liposome-bound ferricyanide and free ferricyanide. The amount of liposome-bound ferricyanide was measured spectrophotometrically in the above mentioned buffer after extraction of lipids by a chloroform-methanol mixture (2:l v/v). This amount of ferricyanide in liposomes was equal to 0.33 pmol/mg lipids. The absorption of ferricyanide on the outer surface of liposomes and the leakage of ferricyanide from the inside of liposomes were examined by the changes in the absorbance of ferricyanide after the addition of 1.0 mM ascorbate as the reductant to the incubation medium. Figure 1 shows that the absorbance of the liposome-bound ferricyanide decreased lo-12% immediately after ascorbate was added, and thereafter remained unchanged during the 2 h of incubation at 37°C. Upon sonication all ferricyanide proved to be accessible to ascorbate. Liposomes in which a-tocopherol in its reduced form was incorporated only into the inner monolayer were obtained by treatment of cu-tocopherol-containing liposomes with ferricyanide (5 X 10m4M) to oxidize the ol-tocopherol in the outer monolayer. After that the suspension was applied onto a Sephadex G-25 column to separate liposomes from ferri- and ferrocyanide. The liposomes obtained contained nonoxidized o-tocopherol in the inner monolayer and oxidized a-tocopherol (which does not possess characteristic fluorescence) in the outer monolayer. Reagents. d&Tocopherol, dimiristoylphosphatidylcholine (DMPC), P-oleoyl-y-palmitoyl-phosphatidylcholine (Serva), egg-yolk lecithin (Sigma), KzHPOl, NaH2P04, trichloroacetic acid (Merck), ascorbate (Reanal), sodium desoxycholate, potassium ferricyanide (Fluka), 2thiobarbituric acid (Sigma), chloroform, methanol (Reachim), Sephadex G-25 (Pharmacia).
RESULTS
Distribution of a-tocopherol in liposomes. It is known that tocopherols undergo diffusion in the lipid bilayer in monomeric form and also associate in domains or form clusters (3). Nonuniform distribution may influence the efficiency of the intermembrane transfer of a-tocoph-
a-TOCOPHEROL TABLE
I
Effect of Sodium Desoxycholate on a-Tocopherol Fluorescence in Liposomes Fluorescence
intensity
a-Tocopherol in the outer monolayer* Liposomal
lipids
Rat cerebral cortex lipids Egg-yolk lecithin DMPCd
Control
2.8 f 0.3 2.1 t- 0.3 1.8 t 0.2
+Detergent
4.0 + 0.4 4.0 f 0.3 3.9 f 0.4
(arbitrary
units) a
a-Tocopherol in the outer and inner monolayer’ Control
4.5 + 0.3 4.2 f 0.3 4.4 + 0.4
+Detergent
4.3 * 0.5 3.9 f 0.4 4.1 i 0.3
’ There were four replicates of each measurement. The values are given as mean values f SD. ’ Ethanol solution of a-tocopherol was added to liposomes to a final concentration 6.66 nmol/mg lipids. ’ Lipids sonicated (22 kHz, 10 min) in the presence of 6.66 nmolo1tocopherol/mg lipid. d Dimiristoylphosphatidylcholine.
erol. Thus we examined the uniformity of a-tocopherol distribution in liposome monolayers by comparing the fluorescence intensity of a-tocopherol in liposomes in the presence and absence of the detergent sodium desoxycholate. Table I shows results of fluorescence measurements of a-tocopherol incorporated into one or both monolayers of liposomes. When a-tocopherol was incorporated into both monolayers of liposomes from saturated (DMPC) or unsaturated lipids (egg-yolk lecithin or rat brain lipids), the addition of the detergent did not cause any significant changes in fluorescence intensity. When a-tocopherol was incorporated only into the outer monolayer of liposomes (by the addition of cy-tocopherol in ethanol to the liposome suspension), in the presence of detergent the fluorescence intensity was much higher than that in the absence of the detergent. This indicates that, at the concentrations used, a-tocopherol does not form clusters when incorporated into both monolayers, rather it, is present in monomeric form. In the experiments described below, we used the procedure where a-tocopherol is incorporated into both monolayers (see Experimental Procedures). Dependence of inhibition of lipid peroxidation in liposomes upon a-tocopherol. The accumulation of lipid peroxidation products (malondialdehyde and/or thiobarbituric acid reactants) in liposomes from rat cerebral cortex lipids in the presence or in the absence of a-tocopherol is shown in Fig. 2. In the absence of cu-tocopherol, a pronounced formation of lipid peroxidation products was observed in the first 15 min of incubation, after which the process was inhibited (30-60 min). The rates of MDA accumulation were 1.6,0.4,0.06 nmol/mg lipid/ min, during time intervals O-5 min, 5-15 min, 15-30
149
IN LIPOSOMES
min, respectively. Earlier data have shown that this is mainly due to a decrease in the ascorbate concentration and to accumulation of ascorbate oxidation product(s) capable of chelating Fe(I1) (19). Therefore, time intervals of lipid peroxidation induction not exceeding 30 min were used to evaluate the antioxidant effect of a-tocopherol in liposomes. Concentration-dependent lipid peroxidation inhibition was found within the range of 0.16-6.66 nmol (Ytocopherol/mg lipid. When concentrations of tocopherol exceeded 3.33 nmol/mg lipid, no accumulation of lipid peroxidation products was observed. Since the a-tocopherol concentration in donor liposomes decreases in the course of its transfer to acceptor liposomes, the susceptibility of donor liposomes to lipid peroxidation inducers should increase. Hence it is necessary to choose tocopherol concentrations which, when decreased twofold in donor liposomes (due to tocopherol migration into acceptor liposomes until equilibrium), still prevent lipid peroxidation in donor liposomes. From Fig. 2 it is clear that a-tocopherol concentrations equal to and
, 0
30
60
Time bin)
FIG. 2. Accumulation of products of lipid peroxidation in unilamellar liposomes from rat cerebral cortex lipids. Incubation medium: 0.1 M K,Na-phosphate buffer, pH 7.4,80 pM Fe(U), 0.5 mM ascorbate, 0.6 mg lipid/ml. The reaction mixture was shaken in air at 37°C. There were five replicates of each measurement. Absence, 1, and presence, 2-9, of a-tocopherol; 2-9, cu-tocopherol concentrations: 0.16,0.25,0.33, 0.40,0.80,1.66,3.33, and 6.66 nmol/mg lipid, respectively.
KAGAN
,-
x
I!
0
15 Time
5
5a 30
ET AL.
plete inhibition of lipid peroxidation was not observed (curves 2-4). Curves 2-4 are observed under the theoretical curve (dashed line), reflecting lipid peroxidation accumulation in acceptor liposomes in the case when no a-tocopherol transfer from donor to acceptor liposomes occurred. An estimation of the amount of a-tocopherol in acceptor liposomes can be made by comparing the efficiency of lipid peroxidation inhibition in acceptor liposomes with the calibration curves given on Fig. 2. The comparison shows that about 0.8-1.0 nmol of a-tocopherol/mg lipid were present in acceptor liposomes after 60 min preincubation, which are 12-E% of the amount in donor liposomes, respectively. Similarly, when the unsaturated fatty acid containing egg-yolk lecithin liposomes were used as tocopherol donors, lipid peroxidation inhibition in acceptor liposomes by a-tocopherol increased with preincubation time of liposomes and with lipid peroxidation inducers (Table II). After a 60-min preincubation, lipid peroxidation corresponded to that caused by 1.0 nmol of a-tocopherol/mg lipid in acceptor liposomes. This corresponds to an exchange of 15% of the a-tocopherol present in donor liposomes. This finding agrees with data of Nakagawa et al. (15) on the rate of intermembrane transfer of a-[3H]tocopherol between liposomes from egg-yolk lecithin. In
(mill)
FIG. 3. Accumulation of lipid peroxidation products in acceptor unilamellar liposomes. The reaction was started by adding inducers of lipid peroxidation. Experimental conditions were the same as in Fig. 2. There were five replicates of each measurement. Experiments were done in the absence, 1, of tocopherol or in the presence of donor liposomes with a-tocopherol at 6.66 nmol/mg lipids, 2-4, or a-tocopherol concentration in donor and acceptor liposomes of 3.33 nmol/mg lipid, 5. Dashed line, the theoretical curve of accumulation of TBA-reactive products when no a-tocopherol transfer occurs between donor and acceptor liposomes. Preincubation of donor and acceptor liposomes before addition of inducers of lipid peroxidation was as follows: 2, 1 min; 3,30 min; 4,60 min.
above 6.66 nmol/mg lipid fulfill this requirement. Thus, in further experiments a standard concentration of 6.66 nmol of a-tocopherol/mg lipids was used. Inhibition of lipid peroxidation and intermembrane atocopherol transfer. To examine the antioxidant effect of a-tocopherol, determined by its intermembrane transfer, we used mixtures of donor (a-tocopherol containing) and acceptor (no a-tocopherol containing) liposomes. Figure 3 shows the accumulation of lipid peroxidation products in liposomes preincubated for l-60 min before the addition of inducers of lipid peroxidation. Curves 1 and 5 represent the lipid peroxidation kinetics in the mixtures of liposomes in the absence and presence of a-tocopherol (both in donor and acceptor liposomes), respectively. When a-tocopherol was incorporated only into the donor liposomes (6.66 nmoles/mg lipid) com-
TABLE Inhibition
II
of Lipid Peroxidation during Intermembrane Transfer of a-Tocopherol in Liposomes of Different Composition” Percentage inhibition of lipid peroxidation in acceptor liposomes *
Donor liposome composition Lipids from rat cerebral cortex Egg-yolk
DMPCd
OPPC’
lecithin
Preincubation time (min) 0 30 60 0 30 60 0 30 60 0 30 60
5 min
Reaction time’ 15 min
19f2 36+2 57 f 2 37 f 3 42 +- 4 55 -1-4 0 18 f 2 24 rf- 2 -
22 k 3 45 f 3 62 k 4 45 * 4 56 f 4 67 2 5 7+2 23 f 2 35 f 3 9*2 36 f 4 49 f 4
30 min 22 48 63 47 62 74 20 28 37
f 3 f 2 k 4 t 3 + 5 f 6 f 2 24 f 2
n There were four replicates of each measurement. The values are given as mean values + SD. * Acceptor liposomes were prepared from rat cerebral cortex lipids. c The reaction was started by adding inducers of lipid peroxidation (see Experimental Procedures). d Dimiristoylphosphatidylcholine. e Oleoyl-palmitoylphosphatidylcholine.
wTOCOPHEROL
151
IN LIPOSOMES
not leak from the acceptor liposomes into the incubation medium (for details see Experimental Procedures and Fig. 1). Figure 4 illustrates the fluorescence spectra of a-tocopherol after a 2-h incubation of the mixtures of liposomes. The fluorescence intensity of a-tocopherol did not change with time, but dropped to zero values after sonication, which made interaction of a-tocopherol with ferricyanide possible. Thus during this 2-h period no fusion of donor and acceptor liposomes would occur in the absence of accumulated lipid peroxidation products, which are known to stimulate liposome fusion (21). DISCUSSION
Intermembrane
310
370
43onm
incorporated into uniFIG. 4. Fluorescence spectra of a-tocopherol lamellar liposomes from rat cerebral cortex lipids. 1, Donor liposomes with a-tocopherol incorporated into the inner monolayer only were mixed with liposomes of the same composition containing no a-tocopherol and incubated for 2 h; 2, donor liposomes with a-tocopherol incorporated into the inner monolayer only were mixed with acceptor liposomes with potassium ferricyanide (0.33 pmol/mg lipid) inside and incubated for 2 h; 3, the same as in 2 but after sonication (22 kHz, 10 s).
experiments where monounsaturated phosphatidylcholine (/3-oleoyl-r-palmitoyl-phosphatidylcholine) donor liposomes were used, lipid peroxidation inhibition in cerebral cortex lipid acceptor liposomes was less efficient (Table II). When liposomes from saturated phospholipid (DMPC) were used as donor liposomes, the antioxidant effect in acceptor liposomes was less pronounced. After 60 min preincubation it did not reach the values corresponding to the presence of 0.3-0.5 nmol of cr-tocopherol/mg lipids in acceptor liposomes (Table II), which corresponds to 4.5-7.5% of a-tocopherol present in donor liposomes. Qualitatively similar results were obtained with lower concentrations of a-tocopherol in donor liposomes (0.9 nmol/mg lipids), which did not provide 100% inhibition of lipid peroxidation. However, the quantitative interpretation of these results was difficult because the induction of lipid peroxidation in donor liposomes was occurring simultaneously with the transfer and antioxidant behavior of a-tocopherol in acceptor liposomes. In a series of experiments, Test for liposomal fusion. the possibility that fusion of donor with acceptor liposomes had occurred was examined. Using gel-filtration and ferricyanide as the oxidant, we obtained donor liposomes containing a-tocopherol only in the inner membrane monolayers, and acceptor liposomes with ferricyanide trapped inside. It was found that ferricyanide did
Transfer and Exchange of a-Tocopherol
We have developed a simple and convenient method to measure tocopherol membrane transfer. The inhibitory effect of a-tocopherol on lipid peroxidation seen in mixtures of donor and acceptor liposomes is an exclusive property of its transfer from donor and exchange into acceptor liposomes. The present results on inhibition of lipid peroxidation by a-tocopherol incorporated into liposomes show that the efficiency of its antioxidant action depends on its localization in the lipid bilayer of either saturated or unsaturated phospholipids. There are two possible explanations why donor liposomes are not susceptible to lipid peroxidation: first, a-tocopherol is in a monomeric form and in high concentration in both monolayers of these liposomes it completely prevents any lipid peroxidation during the time intervals used or, second, liposomes formed from the saturated phospholipid, dimiristoylphosphatidylcholine, do not give TBAreactive products (2,20). Regardless, this indicates that accumulation of lipid peroxidation products occurs only in acceptor liposomes. Intermembrane transfer of tocopherol might result either from: (A) fusion of donor and acceptor liposomes and subsequent lateral diffusion of a-tocopherol or (B) by intermembrane exchange of tocopherol when two types of liposomes come in contact each other (Fig. 5).
l
‘1
A -Tocopherol lipids
FIG. 5. Scheme illustrating intermembrane transfer or fusion as the mechanism of a-tocopherol exchange between donor and acceptor liposomes. A, direct membrane contact; B, fusion.
152
KAGAN TABLE
III
Fatty Acid Composition of Rat Cerebral Cortex Lipids and Egg-Yolk Lecithin Cerebralcortex Fatty acids
lipids
Egg-yolk lecithin
% of total amount 14:o 16:0 16:l 18:O 181 18:2 2O:l 20:4w6 20:5w3 22:6w3 Monounsaturated Polyunsaturated Saturated
Trace 22.2 20.3 24.4 1.0 2.1 12.6 4.7 12.1
2.0 29.3 0.9 17.4 26.4 14.2 5.6
26.5 30.4 42.5
27.3 23.8 48.7
4.0
Note. Traces, less than 0.5%.
Obviously, in the case of intermembrane exchange of tocopherol from donor liposomes, the efficiency of lipid peroxidation inhibition in acceptor liposomes is determined by the rate of intermembrane transfer of a-tocopherol from the outer monolayer of donor liposomes and its exchange into the outer layer of acceptor liposomes. This is because a-tocopherol located in the inner monolayer of donor liposomes cannot participate in intermembrane transfer due to its very low rate of transbilayer mobility (8,9,14). The phospholipids of the inner monolayer of acceptor liposomes are probably susceptible to lipid peroxidation inducers (at least after accumulation of certain amounts of lipid peroxidation products which act as perturbants and would be expected to increase permeability of liposomal bilayers (20)). This could explain the incomplete inhibition of lipid peroxidation in acceptor liposomes. Liposomal Lipid Selectivity Membrane Exchange
in Tocopherol
The physiological role of tocopherol membrane transfer may relate to the lipid composition. The efficiency of lipid peroxidation inhibition decreased in the order: eggyolk lecithin > cerebral cortex lipids > ,&oleoyl-y-palmitoyl-phosphatidylcholine > dimiristoylphosphatidylcholine. Hence the efficiency of the a-tocopherol exchange appears to be determined by unsaturation of liposome lipids. This may be related to less ordering of bilayers formed from unsaturated phospholipids. The degree of unsaturation of fatty acid residues in egg-yolk lecithin was approximately the same as that of cerebral cortex lipids (Table III). The differences in lipid peroxi-
ET AL.
dation inhibition and a-tocopherol transfer between acceptor and donor liposomes, when the latter were prepared from cerebral cortex lipids or egg-yolk lecithin, might be due to the presence of other lipid classes (phosphatidylethanolamine, cholesterol, gangliosides, sphingomyeline) in cerebral cortex lipids. Further experiments are needed to elucidate the role of minor lipid components. Moreover the role of tocopherol-binding proteins in intermembrane transfer and exchange of vitamin E in vivo is still unknown. ACKNOWLEDGMENTS Research supported and NIH (CA47597).
by the Bulgarian
USSR Academy
of Sciences
REFERENCES 1. Tappel, A. L. (1962) Wt. Horm. 20,493-508. 2. Burton, G. W., and Ingold, K. U. (1986) Act. Chem. Res. 19,194201. 3. Kagan, V. E., Bakalova, R. A., Serbinova, E. A., and Stojtchev, Ts. S. (1990) in Methods in Enzymology (Packer, L., and Glezer, A. N., Eds.), Vol. 186, pp. 355-367, Academic Press, New York. 4. Tyurin, V. A., Kagan, V. E., Avrova, N. F., and Prozorovskya, M. P. (1988) Bull. Exp. Biol. Med. USSR 6,667-669. 5. Machlin, L. J. (Ed.) (1984) in Handbook of Vitamins: Nutritional, Biochemical and Clinical Aspects, p. 99, Dekker, New York. 6. Serbinova, E. A., Kadiiska, M. A., Bakalova, R. A., Koynova, G. M., Stoyanovsky, D. A., Karakachev P., Wolinksy, I., and Kagan, V. E. (1989) Z’onicol. Lett. 47,119-123. 7. Packer, L. (1984) Med. Biol. 62,105-109. 8. Tyurin, V. A., Kagan, V. E., Serbinova, E. A., Gorbunov, N. V., Erin, A. N., Prilipko, L. L., and Stoytchev, Ts. S. (1986) Bull. Enp. Biol. Med. USSR 12,689-691. 9. Kagan, V. E., Serbinova, E. A., Bakalova, R. A., Novikov, K. N., Skrypin, V. I., Evstrigneeva, R. P., and Stoytchev, Ts. S. (1988) in Free radicals, Oxidant Stress and Drug Action. (Rice-Evans C., Ed.), pp. 425-442, Richelieu Press, London. J. C., Villalain, J., Aranda, F. J., Ortiz, A., 10. Gomez-Fernandez, Mikob, V., Goutinho, A., Berberan-Santon, M. N., and Prieto, M. J. E. (1989) in Vitamin E: Biochemistry and Health Implications (Diplock, A., Machlin, L., Packer, L., and Pryor, W., Eds.), pp. 109-120, N.Y. Acad. Sci., New York. 11. Behrens, W. A., and Medere, R. (1983) Fed. Proc. 42,813-817. 12. Massey, J. B. (1984) Biochim. Biophys. Acta 793,387-392. 13. Guarinieri, C., Flamigni, F., and Caldarera, C. M. (1980) Biochem. J. 190,469-471. 14. Niki, E., Yamamoto, Y., and Kamiya, Y. (1985) Life Chem. Rep. 3,35-40. 15. Nakagawa, Y., Nojima, S., and Inoue, K. (1980) J. Biochm. 87, 497-502. 16. Fukuzawa, K., Chida, H., Tokumura, A., and Tsukatani, H. (1981) Arch. Biochem. Biophys. 206,173-180. 17. Folch, J., Lees, M., and Stanly, G. (1957) J. Biol. Chem. 226,2-8. 18. Porter, N. A., Nolon, Y., and Ramas, I. (1976) Biochim. Biophys. Acta. 441,506-X2. D. A., Afanas’ev, I. B., Kagan, V. E. (1990) Bull. 19. Stoyanovsky, Exp. Biol. Med. USSR. 2,235-238. in Biomembranes, p. 182, 20. Kagan, V. E. (1988) Lipid Peroxidation CRC Press, Boca Raton, FL. 21. Barsukov, L. I., Victorov, A. V., Vasilenko, I. A., Evstigneeva, R. P., and Bergelson, L. D. (1980) Biochim. Biophys. Actu 598, 136-147.
OF BIOCHEMISTRY
AND
BIOPHYSICS
Vol. 280, No. 1, July, pp. 147-152,199O
Intermembrane Transfer and Antioxidant of a-Tocopherol in Liposomes
Action
V. E. Kagan,” R. A. Bakalova,” Zh. Zh. Zhelev,* D. S. Rangelova,* V. A. Tyurin,? N. K. Denisova,? and L. Packe#
E. A. Serbinova,*
*Institute of Physiology, Bulgarian Academy of Sciences, Sofia 1113, Bulgaria; tlnstitute of Evolutionary and Physiology, USSR Academy of Sciences, Leningrad 194223, USSR; and SDepartment of Molecular Cell Biology, 2.51 LSA, University of California, Berkeley, California 94720
Biochemistry
and
Received December 4,1989, and in revised form March 2,1SS0
Intermembrane transfer and exchange of tocopherol are not well understood. To study this we tested the ability of a-tocopherol containing unilamellar donor liposomes to inhibit the accumulation of lipid peroxidation products in acceptor liposomes. With molar ratios of cz-tocopherol:phospholipids from 1:lOO to 1:lOOO in donor liposomes prepared by sonication of lipid dispersions, a-tocopherol was incorporated into both monolayers and was homogenously distributed in monomeric form without forming clusters in the liposomes. Concentrations of a-tocopherol which completely prevented the peroxidation of lipids were chosen for donor liposomes. Hence inhibition of lipid peroxidation in mixtures of donor and acceptor liposomes was determined by the antioxidant effect of a-tocopherol in acceptor liposomes which resulted from intermembrane transfer and exchange of a-tocopherol. Evidence was obtained that this was not due to fusion of donor with acceptor liposomes. The efficiency of the “intermembrane” antioxidant action of tocopherol was more pronounced when donor liposomes contained unsaturated phospholipids, indicating that the presence of unsaturated fatty acids in the outer monolayer phospholipids 8 1990 facilitates intermembrane tocopherol exchange. Academic
Press,
Inc.
The regulation of free radical lipid peroxidation in eucaryotic membranes is predominantly dependent upon vitamin E, which is mainly in the form of a-tocopherol. w-Tocopherol interacts with lipid alkoxy and peroxy radicals, acting as a chain-breaking antioxidant (1, 2). In the lipid bilayer of membranes a-tocopherol is not uniformly distributed, but forms clusters and is prefer1 To whom correspondence
should be addressed.
0003.9861/90 $3.00 Copyright 0 1990 by Academic Press, All rights of reproduction in any form
entially concentrated in domains rich in polyenoic phospholipids (3,4). Oxidative stress is usually accompanied by a loss of vitamin E (5-7). Hence this may cause a localized antioxidant deficiency in the membrane. Thus domains rich in enzymatic generators of active oxygen species and lipid radicals might become tocopherol-deficient microenvironments. This “local E-hypovitaminosis” in the membrane might be overcome by the migration of tocopherols from regions with a high content of vitamin E to those poor in tocopherois. The efficiency of trans-bilayer tocopherol migration (“flip-flop”) is very low (8,9). Replenishment of E-deficient membrane domains might be brought about by either lateral diffusion or intermembrane transfer of tocopherols. Since the rate of the lateral diffusion of tocopherols is sufficiently high (10) this process may not be limiting. However opinions differ concerning the intermembrane transfer and exchange of tocopherols. It has been suggested that intermembrane exchange is only efficient in the presence of tocopherol-binding proteins (11-13). Existing information about lipid peroxidation inhibition in the case of intermembrane transfer of tocopher01s requires further elucidation. Niki et al. (14) and Nakagawa et al. (15) have found that donor liposomes containing unsaturated phospholipids and a-tocopherol do not influence the rate of lipidperoxidation in acceptor liposomes. However, Fukuzawa et al. (16) have demon-
strated a protective effect under similar conditions when the donor liposomes are obtained from saturated phospholipids. The present study was undertaken to clarify this discrepancy and to obtain further evidence for intermembrane exchange of tocopherols. To study this we used tocopherol-loaded (donor) and tocopherol-free (acceptor) liposomes under conditions where no membrane fu147
Inc. reserved.
148
KAGAN
ET AL. ascorbate
sion occurs. The antioxidant effects of cY-tocopherol were examined by incorporating it into unilamellar donor liposomes prepared from either saturated or unsaturated phospholipids, and monitoring its inhibition of lipid peroxidation in acceptor liposomes prepared from rat cerebral cortex lipids. EXPERIMENTAL
PROCEDURES
Liposomes. Total lipids from the cerebral cortex of male Wistar rats (150-180 g) were isolated by the method of Folch (17). Lipid fatty acid composition was determined by gas-liquid chromatography (Pye104) using a flame-ionization detector. Unilamellar liposomes were obtained bysonication (10 min, 22 kHz at 25°C for dimiristoylphosphatidylcholine and at 4°C for unsaturated lipids) of lipid dispersions (0.6 buffer (pH 7.4 at 25°C) until mg of lipids/ml) in 0.1 M K,Na-phosphate the suspension became clear. Sonication was performed under nitrogen stream. The amount of lipid peroxidation products in unsaturated lipids under conditions used did not exceed 0.9 nmol MDA*/mg lipids. Negatively stained electron micrographs showed that the average diameter of unilammelar liposomes was 30-40 nm. Incorporation of (Ytocopherol into liposomes was accomplished either by addition of the ethanol solution of ol-tocopherol to the liposome suspension or by dissolving lipids and o-tocopherol in chloroform, evaporating to dryness, and subsequent dispersion and sonication in K,Na-phosphate buffer, pH 7.4, as described above. Tocopherol distribution in liposomes. The incorporation and monomeric distribution of a-tocopherol in liposomes were estimated using a fluorescence method (3). This method is based on an increase of fluorescence intensity of a-tocopherol (excitation = 292 nm, emission = 325 nm) which occurs when there is a decrease in local a-tocopherol concentration. This increase of fluorescence intensity is due to elimination of tocopherol concentration-dependent fluorescence selfquenching. The maximum incorporation and the fully monomeric distribution of a-tocopherol in liposome bilayers was determined by the fluorescence intensity obtained in the presence of the detergent, sodium desoxycholate (at a concentration of 25 mM, which exceeded the critical micellar concentration). In the presence of detergent, a-tocopherol is distributed uniformly in mixed detergent-lipid micelles. Lipid peroxidation in liposomes was Lipid peroxidation products. induced by the Fe*+ + ascorbate oxidation system. Incubation medium contained: 0.1 M K,Na-phosphate buffer, pH 7.4 at 37°C 80 pM Fe’+, 0.5 mM ascorbate, 0.6 mg lipid/ml. The reaction was started by simultaneous addition of Fe’+ and ascorbate. The reaction mixture was shaken in air at 37°C. Lipid peroxidation products were determined by reaction with 2-thiobarbituric acid (TBA) and measured spectrophotometrically at 535 nm (E = 1.56 X lo5 M-’ cm-‘) (18). Lipid peroxidation induction was performed in mixtures of two types of liposomes: donor liposomes containing a-tocopherol and acceptor liposomes containing no oc-tocopherol. Donor and acceptor liposomes were preincubated for the time periods described in the legends to the figures from 1 to 60 min at 37°C. Test for fusion of donor and acceptor liposomes. To assess fusion of donor and acceptor liposomes, we obtained donor liposomes containing a-tocopherol only in the inner membrane monolayer and acceptor liposomes with ferricyanide trapped inside. For these experiments liposomes containing ferricyanide were prepared by sonication (10 min at 22 kHz) of lipid dispersions (2.5 mg/ ml) in the presence of ferricyanide (0.5 M). The suspension was then applied onto a column filled with Sephadex G-25, which was presaturated by a mixture of lipids and ferricyanide, and was chromatographed in 0.1 M K,Na-phosphate buffer at pH 7.4. The spectrophoto2 Abbreviations used: TBA, 2-thiobarbituric acid; DMPC, toylphosphatidylcholine; MDA, malondialdehyde.
dimiris-
0
1
2
Time ecurs)
FIG. 1. Changes in the absorbance of potassium ferricyanide in unilamellar liposomes prepared from cerebral cortex lipids after addition of ascorbate and subsequent sonication (22 kHz, 10 s). Incubation medium: liposomes, 0.65 mg lipids/ml; potassium ferricyanide inside liposomes or absorbed on the surface of liposomes, 0.33 pmol/mg lipids; ascorbate, 1.0 mM; 0.1 M K,Na-phosphate buffer, pH 7.4, at 37°C. There were four replicates of each measurement. metric determination of ferricyanide at 418 nm (3) revealed two fractions: liposome-bound ferricyanide and free ferricyanide. The amount of liposome-bound ferricyanide was measured spectrophotometrically in the above mentioned buffer after extraction of lipids by a chloroform-methanol mixture (2:l v/v). This amount of ferricyanide in liposomes was equal to 0.33 pmol/mg lipids. The absorption of ferricyanide on the outer surface of liposomes and the leakage of ferricyanide from the inside of liposomes were examined by the changes in the absorbance of ferricyanide after the addition of 1.0 mM ascorbate as the reductant to the incubation medium. Figure 1 shows that the absorbance of the liposome-bound ferricyanide decreased lo-12% immediately after ascorbate was added, and thereafter remained unchanged during the 2 h of incubation at 37°C. Upon sonication all ferricyanide proved to be accessible to ascorbate. Liposomes in which a-tocopherol in its reduced form was incorporated only into the inner monolayer were obtained by treatment of cu-tocopherol-containing liposomes with ferricyanide (5 X 10m4M) to oxidize the ol-tocopherol in the outer monolayer. After that the suspension was applied onto a Sephadex G-25 column to separate liposomes from ferri- and ferrocyanide. The liposomes obtained contained nonoxidized o-tocopherol in the inner monolayer and oxidized a-tocopherol (which does not possess characteristic fluorescence) in the outer monolayer. Reagents. d&Tocopherol, dimiristoylphosphatidylcholine (DMPC), P-oleoyl-y-palmitoyl-phosphatidylcholine (Serva), egg-yolk lecithin (Sigma), KzHPOl, NaH2P04, trichloroacetic acid (Merck), ascorbate (Reanal), sodium desoxycholate, potassium ferricyanide (Fluka), 2thiobarbituric acid (Sigma), chloroform, methanol (Reachim), Sephadex G-25 (Pharmacia).
RESULTS
Distribution of a-tocopherol in liposomes. It is known that tocopherols undergo diffusion in the lipid bilayer in monomeric form and also associate in domains or form clusters (3). Nonuniform distribution may influence the efficiency of the intermembrane transfer of a-tocoph-
a-TOCOPHEROL TABLE
I
Effect of Sodium Desoxycholate on a-Tocopherol Fluorescence in Liposomes Fluorescence
intensity
a-Tocopherol in the outer monolayer* Liposomal
lipids
Rat cerebral cortex lipids Egg-yolk lecithin DMPCd
Control
2.8 f 0.3 2.1 t- 0.3 1.8 t 0.2
+Detergent
4.0 + 0.4 4.0 f 0.3 3.9 f 0.4
(arbitrary
units) a
a-Tocopherol in the outer and inner monolayer’ Control
4.5 + 0.3 4.2 f 0.3 4.4 + 0.4
+Detergent
4.3 * 0.5 3.9 f 0.4 4.1 i 0.3
’ There were four replicates of each measurement. The values are given as mean values f SD. ’ Ethanol solution of a-tocopherol was added to liposomes to a final concentration 6.66 nmol/mg lipids. ’ Lipids sonicated (22 kHz, 10 min) in the presence of 6.66 nmolo1tocopherol/mg lipid. d Dimiristoylphosphatidylcholine.
erol. Thus we examined the uniformity of a-tocopherol distribution in liposome monolayers by comparing the fluorescence intensity of a-tocopherol in liposomes in the presence and absence of the detergent sodium desoxycholate. Table I shows results of fluorescence measurements of a-tocopherol incorporated into one or both monolayers of liposomes. When a-tocopherol was incorporated into both monolayers of liposomes from saturated (DMPC) or unsaturated lipids (egg-yolk lecithin or rat brain lipids), the addition of the detergent did not cause any significant changes in fluorescence intensity. When a-tocopherol was incorporated only into the outer monolayer of liposomes (by the addition of cy-tocopherol in ethanol to the liposome suspension), in the presence of detergent the fluorescence intensity was much higher than that in the absence of the detergent. This indicates that, at the concentrations used, a-tocopherol does not form clusters when incorporated into both monolayers, rather it, is present in monomeric form. In the experiments described below, we used the procedure where a-tocopherol is incorporated into both monolayers (see Experimental Procedures). Dependence of inhibition of lipid peroxidation in liposomes upon a-tocopherol. The accumulation of lipid peroxidation products (malondialdehyde and/or thiobarbituric acid reactants) in liposomes from rat cerebral cortex lipids in the presence or in the absence of a-tocopherol is shown in Fig. 2. In the absence of cu-tocopherol, a pronounced formation of lipid peroxidation products was observed in the first 15 min of incubation, after which the process was inhibited (30-60 min). The rates of MDA accumulation were 1.6,0.4,0.06 nmol/mg lipid/ min, during time intervals O-5 min, 5-15 min, 15-30
149
IN LIPOSOMES
min, respectively. Earlier data have shown that this is mainly due to a decrease in the ascorbate concentration and to accumulation of ascorbate oxidation product(s) capable of chelating Fe(I1) (19). Therefore, time intervals of lipid peroxidation induction not exceeding 30 min were used to evaluate the antioxidant effect of a-tocopherol in liposomes. Concentration-dependent lipid peroxidation inhibition was found within the range of 0.16-6.66 nmol (Ytocopherol/mg lipid. When concentrations of tocopherol exceeded 3.33 nmol/mg lipid, no accumulation of lipid peroxidation products was observed. Since the a-tocopherol concentration in donor liposomes decreases in the course of its transfer to acceptor liposomes, the susceptibility of donor liposomes to lipid peroxidation inducers should increase. Hence it is necessary to choose tocopherol concentrations which, when decreased twofold in donor liposomes (due to tocopherol migration into acceptor liposomes until equilibrium), still prevent lipid peroxidation in donor liposomes. From Fig. 2 it is clear that a-tocopherol concentrations equal to and
, 0
30
60
Time bin)
FIG. 2. Accumulation of products of lipid peroxidation in unilamellar liposomes from rat cerebral cortex lipids. Incubation medium: 0.1 M K,Na-phosphate buffer, pH 7.4,80 pM Fe(U), 0.5 mM ascorbate, 0.6 mg lipid/ml. The reaction mixture was shaken in air at 37°C. There were five replicates of each measurement. Absence, 1, and presence, 2-9, of a-tocopherol; 2-9, cu-tocopherol concentrations: 0.16,0.25,0.33, 0.40,0.80,1.66,3.33, and 6.66 nmol/mg lipid, respectively.
KAGAN
,-
x
I!
0
15 Time
5
5a 30
ET AL.
plete inhibition of lipid peroxidation was not observed (curves 2-4). Curves 2-4 are observed under the theoretical curve (dashed line), reflecting lipid peroxidation accumulation in acceptor liposomes in the case when no a-tocopherol transfer from donor to acceptor liposomes occurred. An estimation of the amount of a-tocopherol in acceptor liposomes can be made by comparing the efficiency of lipid peroxidation inhibition in acceptor liposomes with the calibration curves given on Fig. 2. The comparison shows that about 0.8-1.0 nmol of a-tocopherol/mg lipid were present in acceptor liposomes after 60 min preincubation, which are 12-E% of the amount in donor liposomes, respectively. Similarly, when the unsaturated fatty acid containing egg-yolk lecithin liposomes were used as tocopherol donors, lipid peroxidation inhibition in acceptor liposomes by a-tocopherol increased with preincubation time of liposomes and with lipid peroxidation inducers (Table II). After a 60-min preincubation, lipid peroxidation corresponded to that caused by 1.0 nmol of a-tocopherol/mg lipid in acceptor liposomes. This corresponds to an exchange of 15% of the a-tocopherol present in donor liposomes. This finding agrees with data of Nakagawa et al. (15) on the rate of intermembrane transfer of a-[3H]tocopherol between liposomes from egg-yolk lecithin. In
(mill)
FIG. 3. Accumulation of lipid peroxidation products in acceptor unilamellar liposomes. The reaction was started by adding inducers of lipid peroxidation. Experimental conditions were the same as in Fig. 2. There were five replicates of each measurement. Experiments were done in the absence, 1, of tocopherol or in the presence of donor liposomes with a-tocopherol at 6.66 nmol/mg lipids, 2-4, or a-tocopherol concentration in donor and acceptor liposomes of 3.33 nmol/mg lipid, 5. Dashed line, the theoretical curve of accumulation of TBA-reactive products when no a-tocopherol transfer occurs between donor and acceptor liposomes. Preincubation of donor and acceptor liposomes before addition of inducers of lipid peroxidation was as follows: 2, 1 min; 3,30 min; 4,60 min.
above 6.66 nmol/mg lipid fulfill this requirement. Thus, in further experiments a standard concentration of 6.66 nmol of a-tocopherol/mg lipids was used. Inhibition of lipid peroxidation and intermembrane atocopherol transfer. To examine the antioxidant effect of a-tocopherol, determined by its intermembrane transfer, we used mixtures of donor (a-tocopherol containing) and acceptor (no a-tocopherol containing) liposomes. Figure 3 shows the accumulation of lipid peroxidation products in liposomes preincubated for l-60 min before the addition of inducers of lipid peroxidation. Curves 1 and 5 represent the lipid peroxidation kinetics in the mixtures of liposomes in the absence and presence of a-tocopherol (both in donor and acceptor liposomes), respectively. When a-tocopherol was incorporated only into the donor liposomes (6.66 nmoles/mg lipid) com-
TABLE Inhibition
II
of Lipid Peroxidation during Intermembrane Transfer of a-Tocopherol in Liposomes of Different Composition” Percentage inhibition of lipid peroxidation in acceptor liposomes *
Donor liposome composition Lipids from rat cerebral cortex Egg-yolk
DMPCd
OPPC’
lecithin
Preincubation time (min) 0 30 60 0 30 60 0 30 60 0 30 60
5 min
Reaction time’ 15 min
19f2 36+2 57 f 2 37 f 3 42 +- 4 55 -1-4 0 18 f 2 24 rf- 2 -
22 k 3 45 f 3 62 k 4 45 * 4 56 f 4 67 2 5 7+2 23 f 2 35 f 3 9*2 36 f 4 49 f 4
30 min 22 48 63 47 62 74 20 28 37
f 3 f 2 k 4 t 3 + 5 f 6 f 2 24 f 2
n There were four replicates of each measurement. The values are given as mean values + SD. * Acceptor liposomes were prepared from rat cerebral cortex lipids. c The reaction was started by adding inducers of lipid peroxidation (see Experimental Procedures). d Dimiristoylphosphatidylcholine. e Oleoyl-palmitoylphosphatidylcholine.
wTOCOPHEROL
151
IN LIPOSOMES
not leak from the acceptor liposomes into the incubation medium (for details see Experimental Procedures and Fig. 1). Figure 4 illustrates the fluorescence spectra of a-tocopherol after a 2-h incubation of the mixtures of liposomes. The fluorescence intensity of a-tocopherol did not change with time, but dropped to zero values after sonication, which made interaction of a-tocopherol with ferricyanide possible. Thus during this 2-h period no fusion of donor and acceptor liposomes would occur in the absence of accumulated lipid peroxidation products, which are known to stimulate liposome fusion (21). DISCUSSION
Intermembrane
310
370
43onm
incorporated into uniFIG. 4. Fluorescence spectra of a-tocopherol lamellar liposomes from rat cerebral cortex lipids. 1, Donor liposomes with a-tocopherol incorporated into the inner monolayer only were mixed with liposomes of the same composition containing no a-tocopherol and incubated for 2 h; 2, donor liposomes with a-tocopherol incorporated into the inner monolayer only were mixed with acceptor liposomes with potassium ferricyanide (0.33 pmol/mg lipid) inside and incubated for 2 h; 3, the same as in 2 but after sonication (22 kHz, 10 s).
experiments where monounsaturated phosphatidylcholine (/3-oleoyl-r-palmitoyl-phosphatidylcholine) donor liposomes were used, lipid peroxidation inhibition in cerebral cortex lipid acceptor liposomes was less efficient (Table II). When liposomes from saturated phospholipid (DMPC) were used as donor liposomes, the antioxidant effect in acceptor liposomes was less pronounced. After 60 min preincubation it did not reach the values corresponding to the presence of 0.3-0.5 nmol of cr-tocopherol/mg lipids in acceptor liposomes (Table II), which corresponds to 4.5-7.5% of a-tocopherol present in donor liposomes. Qualitatively similar results were obtained with lower concentrations of a-tocopherol in donor liposomes (0.9 nmol/mg lipids), which did not provide 100% inhibition of lipid peroxidation. However, the quantitative interpretation of these results was difficult because the induction of lipid peroxidation in donor liposomes was occurring simultaneously with the transfer and antioxidant behavior of a-tocopherol in acceptor liposomes. In a series of experiments, Test for liposomal fusion. the possibility that fusion of donor with acceptor liposomes had occurred was examined. Using gel-filtration and ferricyanide as the oxidant, we obtained donor liposomes containing a-tocopherol only in the inner membrane monolayers, and acceptor liposomes with ferricyanide trapped inside. It was found that ferricyanide did
Transfer and Exchange of a-Tocopherol
We have developed a simple and convenient method to measure tocopherol membrane transfer. The inhibitory effect of a-tocopherol on lipid peroxidation seen in mixtures of donor and acceptor liposomes is an exclusive property of its transfer from donor and exchange into acceptor liposomes. The present results on inhibition of lipid peroxidation by a-tocopherol incorporated into liposomes show that the efficiency of its antioxidant action depends on its localization in the lipid bilayer of either saturated or unsaturated phospholipids. There are two possible explanations why donor liposomes are not susceptible to lipid peroxidation: first, a-tocopherol is in a monomeric form and in high concentration in both monolayers of these liposomes it completely prevents any lipid peroxidation during the time intervals used or, second, liposomes formed from the saturated phospholipid, dimiristoylphosphatidylcholine, do not give TBAreactive products (2,20). Regardless, this indicates that accumulation of lipid peroxidation products occurs only in acceptor liposomes. Intermembrane transfer of tocopherol might result either from: (A) fusion of donor and acceptor liposomes and subsequent lateral diffusion of a-tocopherol or (B) by intermembrane exchange of tocopherol when two types of liposomes come in contact each other (Fig. 5).
l
‘1
A -Tocopherol lipids
FIG. 5. Scheme illustrating intermembrane transfer or fusion as the mechanism of a-tocopherol exchange between donor and acceptor liposomes. A, direct membrane contact; B, fusion.
152
KAGAN TABLE
III
Fatty Acid Composition of Rat Cerebral Cortex Lipids and Egg-Yolk Lecithin Cerebralcortex Fatty acids
lipids
Egg-yolk lecithin
% of total amount 14:o 16:0 16:l 18:O 181 18:2 2O:l 20:4w6 20:5w3 22:6w3 Monounsaturated Polyunsaturated Saturated
Trace 22.2 20.3 24.4 1.0 2.1 12.6 4.7 12.1
2.0 29.3 0.9 17.4 26.4 14.2 5.6
26.5 30.4 42.5
27.3 23.8 48.7
4.0
Note. Traces, less than 0.5%.
Obviously, in the case of intermembrane exchange of tocopherol from donor liposomes, the efficiency of lipid peroxidation inhibition in acceptor liposomes is determined by the rate of intermembrane transfer of a-tocopherol from the outer monolayer of donor liposomes and its exchange into the outer layer of acceptor liposomes. This is because a-tocopherol located in the inner monolayer of donor liposomes cannot participate in intermembrane transfer due to its very low rate of transbilayer mobility (8,9,14). The phospholipids of the inner monolayer of acceptor liposomes are probably susceptible to lipid peroxidation inducers (at least after accumulation of certain amounts of lipid peroxidation products which act as perturbants and would be expected to increase permeability of liposomal bilayers (20)). This could explain the incomplete inhibition of lipid peroxidation in acceptor liposomes. Liposomal Lipid Selectivity Membrane Exchange
in Tocopherol
The physiological role of tocopherol membrane transfer may relate to the lipid composition. The efficiency of lipid peroxidation inhibition decreased in the order: eggyolk lecithin > cerebral cortex lipids > ,&oleoyl-y-palmitoyl-phosphatidylcholine > dimiristoylphosphatidylcholine. Hence the efficiency of the a-tocopherol exchange appears to be determined by unsaturation of liposome lipids. This may be related to less ordering of bilayers formed from unsaturated phospholipids. The degree of unsaturation of fatty acid residues in egg-yolk lecithin was approximately the same as that of cerebral cortex lipids (Table III). The differences in lipid peroxi-
ET AL.
dation inhibition and a-tocopherol transfer between acceptor and donor liposomes, when the latter were prepared from cerebral cortex lipids or egg-yolk lecithin, might be due to the presence of other lipid classes (phosphatidylethanolamine, cholesterol, gangliosides, sphingomyeline) in cerebral cortex lipids. Further experiments are needed to elucidate the role of minor lipid components. Moreover the role of tocopherol-binding proteins in intermembrane transfer and exchange of vitamin E in vivo is still unknown. ACKNOWLEDGMENTS Research supported and NIH (CA47597).
by the Bulgarian
USSR Academy
of Sciences
REFERENCES 1. Tappel, A. L. (1962) Wt. Horm. 20,493-508. 2. Burton, G. W., and Ingold, K. U. (1986) Act. Chem. Res. 19,194201. 3. Kagan, V. E., Bakalova, R. A., Serbinova, E. A., and Stojtchev, Ts. S. (1990) in Methods in Enzymology (Packer, L., and Glezer, A. N., Eds.), Vol. 186, pp. 355-367, Academic Press, New York. 4. Tyurin, V. A., Kagan, V. E., Avrova, N. F., and Prozorovskya, M. P. (1988) Bull. Exp. Biol. Med. USSR 6,667-669. 5. Machlin, L. J. (Ed.) (1984) in Handbook of Vitamins: Nutritional, Biochemical and Clinical Aspects, p. 99, Dekker, New York. 6. Serbinova, E. A., Kadiiska, M. A., Bakalova, R. A., Koynova, G. M., Stoyanovsky, D. A., Karakachev P., Wolinksy, I., and Kagan, V. E. (1989) Z’onicol. Lett. 47,119-123. 7. Packer, L. (1984) Med. Biol. 62,105-109. 8. Tyurin, V. A., Kagan, V. E., Serbinova, E. A., Gorbunov, N. V., Erin, A. N., Prilipko, L. L., and Stoytchev, Ts. S. (1986) Bull. Enp. Biol. Med. USSR 12,689-691. 9. Kagan, V. E., Serbinova, E. A., Bakalova, R. A., Novikov, K. N., Skrypin, V. I., Evstrigneeva, R. P., and Stoytchev, Ts. S. (1988) in Free radicals, Oxidant Stress and Drug Action. (Rice-Evans C., Ed.), pp. 425-442, Richelieu Press, London. J. C., Villalain, J., Aranda, F. J., Ortiz, A., 10. Gomez-Fernandez, Mikob, V., Goutinho, A., Berberan-Santon, M. N., and Prieto, M. J. E. (1989) in Vitamin E: Biochemistry and Health Implications (Diplock, A., Machlin, L., Packer, L., and Pryor, W., Eds.), pp. 109-120, N.Y. Acad. Sci., New York. 11. Behrens, W. A., and Medere, R. (1983) Fed. Proc. 42,813-817. 12. Massey, J. B. (1984) Biochim. Biophys. Acta 793,387-392. 13. Guarinieri, C., Flamigni, F., and Caldarera, C. M. (1980) Biochem. J. 190,469-471. 14. Niki, E., Yamamoto, Y., and Kamiya, Y. (1985) Life Chem. Rep. 3,35-40. 15. Nakagawa, Y., Nojima, S., and Inoue, K. (1980) J. Biochm. 87, 497-502. 16. Fukuzawa, K., Chida, H., Tokumura, A., and Tsukatani, H. (1981) Arch. Biochem. Biophys. 206,173-180. 17. Folch, J., Lees, M., and Stanly, G. (1957) J. Biol. Chem. 226,2-8. 18. Porter, N. A., Nolon, Y., and Ramas, I. (1976) Biochim. Biophys. Acta. 441,506-X2. D. A., Afanas’ev, I. B., Kagan, V. E. (1990) Bull. 19. Stoyanovsky, Exp. Biol. Med. USSR. 2,235-238. in Biomembranes, p. 182, 20. Kagan, V. E. (1988) Lipid Peroxidation CRC Press, Boca Raton, FL. 21. Barsukov, L. I., Victorov, A. V., Vasilenko, I. A., Evstigneeva, R. P., and Bergelson, L. D. (1980) Biochim. Biophys. Actu 598, 136-147.