468

BBA 52916

Biochimica et Biophysics Acta, 963 (1988) 468-475 Elsevier

Effect of vitamin E enrichment on arachidonic acid release

and cellular phospholipids in cultured human endothelial cells

Khai Tran and Alvin C. Chan Department of Biochemistry, Faculty of Health Sciences University of Ottawa, Ottawa (Canada)

(Received 24 May 1988)

Key words: Arachidonic acid release; Vitamin E enrichment; Phospholipid; Cell culture; (Human endothelial cell)

The effect of (R,R,R)-~tocopherol on agonist-stimulated arachidonate release and cellular lipids was investigated in cultured human umbilical cord endothelial cells. Endothelial cells in culture incorporate added tocopherol in a dose-dependent manner at both physiological (23.2 PM) or pharmacological (92.8 PM) concentrations which were well tolerated by the cells, as judged by unaltered cell number and viability. Two experiments were conducted in which cells were either incubated with (R,R,R)-cr-tocopherol followed by labelling with [1-‘4C]arachidonic acid or they were labelled with arachidonate followed by incubation with tocopherol. Irrespective of the sequence of incubation with arachidonate and tocopherol, (R,R,R)-a- tocopherol-enriched cells released significantly more labelled arachidonate when stimulated with thrombin (2.5 U/ml) or ionophore A23187 (1 PM) for 10 min. The magnitude of [l- r4 Clarachidonate release was higher from ionophore A23187 stimulation than from thrombin stimulation, but the trend of increased arachidonate release in tocopherol-enriched cells was the same. Results from these studies demonstrate that (R,R,R)-ar-tocopherol can stimulate arachidonate release in human endothelial cells. This observation is in direct contrast to the role of tocopherol, which has been shown to inhibit platelet and cardiac phospholipase A, activity in rats, and to reduce thrombin-stimulated thromboxane release in rat platelets.

Introduction

When stimulated, human endothelial cells in culture actively synthesize potent lipid-derived mediators, such as prostaglandin I,, E, and platelet activating factor [l-4], which possess potent but varying biological activities. In intact cells, the pathways leading to the synthesis of these mediators are complex but they share com-

Abbreviations: Hepes, 4-(2-hydroxyethyl)-l-piperazineethane- sulfonic acid; HBS, Hepes-buffered saline; BSA, bovine serum

albumin; DMSO, dimethyl sulfoxide.

mon phospholipid precursors and require phos- pholipase A, activity. The release of arachidonic

acid is now generally accepted as the rate-limiting step in eicosanoid synthesis. In platelets, this re- lease reaction was initiated by the activation of

phospholipase C through the hydrolysis of phos- phatidylinositol [.5]. However, the bulk of arachidonate was found to be from the direct hydrolysis of phosphatidylcholine via a phos- pholipase A, activity [6-81.

In mammalian cells, a-tocopherol is an integral component of biomembranes and it represents the only hydrophobic antioxidant found in the bi- layer. When the dietary level of this vitamin is enriched, accumulation of tocopherol in mem- branes from various tissues is usually found to be logarithmically related to dietary tocopherol con- centrations. Dietary a-tocopherol has been re-

Correspondence: A.C. Chan, Department of Biochemistry,

University of Ottawa, 451 Smyth Road, Ottawa, Ontario,

Canada KlH 8M5.

0005-2760/88/$03.50 0 1988 Elsevier Science Publishers B.V. (Biomedical Division)

469

ported to influence cyclooxygenase product pro- file, and this effect is tissue-specific. For example, in thrombin-stimulated rat platelets, the release of thromboxane A 2 was reduced when dietary

tocopherol was proportionally increased [9]. In contrast, vascular prostaglandin I, release ap-

peared to require a certain amount of tocopherol in the diet [lo-121, and in cultured human en- dothelial cells, higher prostaglandin I, production was reported when cells were grown in the pres-

ence of a-tocopherol [13]. The present study was undertaken in order to

understand better the involvement of a-tocopherol in the metabolic turnover of arachidonic acid in

human endothelial cells in culture. In contrast to our previous reports, which showed that a-

tocopherol inhibits rat platelets and myocardial phospholipase A 2 activity [14,15], the present study demonstrated that vitamin E stimulates

arachidonate release in human endothelial cells in culture.

Experimental procedures

Materials. Medium 199 with Hanks’ salts and L-glutamine, heat-inactivated bovine serum, sodium penicillin G (10000 units/ml), strep- tomycin sulfate (10000 pgg/ml), fungizone (250 pg amphotericin B and 205 pg sodium deoxy- cholate/ml), trypsin-EDTA and culture dishes were from Gibco (Burlington, Ontario). Endothe- lial cell growth supplement was from Collabora-

tive Research (Bedford, MA). Hepes, collagenase type IV, gentamicin sulfate, heparin, thrombin (1000 NIH units per mg protein) and calcium ionophore A23187 were from Sigma (St. Louis, MO).

The medium 199 (pH 7.4) was supplemented with heparin (90 pg/ml), Hepes (25 mM), gentamicin sulfate (40 pg/ml), sodium penicillin G (100 U/ml), streptomycin sulfate (100 pg/ml), fungizone (2.5 pg amphotericin B/ml) and heat- inactivated fetal bovine serum (20%). The endo- thelial cell growth supplement (30 pg/ml) was added into the culture dishes after each feeding.

The (R, R, R)-a-tocopherol was a generous gift from Henkel Co. (Chicago, IL), and Eisai Co (Tokyo, Japan) and all-rat-a-tocopherol acetate was from Sigma (St. Louis, MO). HPLC grade

solvents were from BDH Chemicals. [l-‘4C]Arach-

idonate (specific activity 54.9 mCi/mmol) was purchased from New England Nuclear, Boston, MA. All glassware was silanized prior to use.

Culture of endothelial cells. Endothelial cells were isolated from human umbilical cord veins by the

method of Jaffe [16]. Cords were cannulated with tubing from butterfly needles and flushed with 50 ml of warm Hepes-buffered saline (HBS) (10 mM Hepes/0.14 M NaC1/4 mM KCl/ll mM glucose

(pH 7.4)) to remove residual blood. The vein was then filled with 10 ml of 0.2% collagenase in HBS

and incubated at 37°C in a bath of phosphate- buffered saline for 15 min. The collagenase/cell mixture was flushed with 40 ml HBS into a 50 ml plastic centrifuge tube containing 5 ml medium 199 and then centrifuged at 180 x g (or 1000 rpm) for 10 min. The cell pellet was resuspended in 8 ml of fresh medium 199 and cultured on a 0.2% gelatin-coated petri dish (58 cm2). After 24 h in

5% CO,/95% air incubator, the medium was re- placed with new medium 199 supplemented with

30 pg/ml endothelial cell growth factor [17]. Cells were fed every 3 days. Confluency was usually attained in 7-10 days. Cells were detached by trypsinization and were subcultured in a 1 : 3 ratio on 9 or 58 cm2 petri dishes. Cells in these experi- ments were in passages 2 and 3. Human endo- thelial cells were identified by the presence of factor VIII-related antigen by immunoflurescent microscopy [ 181.

Preparation of (R,R,R)-a-tocopherol-enriched

medium. An aliquot of 1 pg/pl of (R,R,R)-a-

tocopherol in ethanol was pipetted into a test-tube and the solvent was removed under N, gas. A

microliter volume of dimethyl sulfoxide (DMSO), which represented 0.4% of the total volume of medium, was added to solubilize the tocopherol. 20% of heat-inactivated fetal bovine serum was added, and the test-tube was vortexed vigorously. Medium 199 and antibiotics were added last, and the tocopherol-enriched medium was incubated at 37 o C for 15-20 min in the dark. Medium without added tocopherol was prepared with exactly the same protocol. Tocopherol carried by this method gave an actual tocopherol concentration of greater than 95% compared to the calculated one. The small volume of DMSO was found not to be harmful to the cells.

470

Preparation of [1 -‘4C]arachidonic acid-enriched

medium. [l-‘4C]Arachidonic acid in ethanol was dried under N, gas. A small volume of DMSO was added as vehicle (0.4%) and vortexed before 10% heat-inactivated fetal bovine serum/medium

199 and antibiotics were added.

[l -t4C]Arachidonic acid and tocopherol incuba-

tion. Confluent endothelial monolayers in 9 cm* dishes were used in these experiments. Two sets of experiments were performed as follows. In Expt. 1, the cells were incubated with 2 ml of medium containing 0, 23.2 and 92.0 PM of (R,R,R)-a-

tocopherol for 20 h at which time the tocopherol medium was removed and cells were rinsed with 0.25% bovine serum albumin (BSA) in warm HBS buffer. These cells were labelled with 0.25 PCi of [l-‘4C]arachidonic acid in culture medium for 5 h. In Expt. 2, cells were first labelled with 0.25 PCi

of [l-14C]arachidonic acid for 20 h, rinsed with HBS (0.25% BSA) and incubated with 2 ml of medium enriched with different concentrations of tocopherol for 5 h.

Thrombin and ionophore stimulation. Both thrombin (2.5 U/ml) and ionophore A23187 (1 PM) were dissolved in Ringer-Tyrode buffer con- taining 0.25% BSA and 1 mM Ca*+. All of the dishes were rinsed with HBS (0.25% BSA) to remove any nonspecific binding of arachidonic acid to the cells prior to agonist stimulation. At the end of the 10 min stimulation period, the medium was quickly removed and an aliquot was taken for determination of radioactivity by scintil- lation counting. Cells were rinsed with HBS (0.25% BSA) before being detached by trypsinization or by scraping in ice-cold methanol. Determination of cell number and viability by Trypan blue exclu- sion indicated that neither tocopherol nor the agonist stimulation had any effect on these cell

parameters. Lipid analysis and tocopherol determination. De-

tached cells were acidified with glacial acetic acid to pH 3.6 and cell lipids were extracted by the method of Bligh and Dyer [19]. Lipids were sep- arated by two thin-layer chromatographic proce- dures using the solvent system chloroform/ methanol/acetic acid/water (85 : 15 : 10 : 3.5, v/v) for phospholipids and hexane/ diethyl ether/ acetic acid (70 : 30 : 1, v/v) for neutral lipids. The spots comigrated with authentic standards, were

detected by autoradiography, and were directly scraped into scintillation vials containing 10 ml ACS II (Amersham). Radioactivity was de-

termined using a Beckman LS 1801 spectrometer equipped for automatic quenching correction.

(R, R, R)-cY-Tocopherol was determined from an aliquot of cell lipid extract with addition of all-

rat-tocopherol acetate as an internal standard. The tocopherols were separated and quantitated by reversed-phase HPLC using a C-18 column

with a solvent system which contained 99% methanol/l% water/O.l% trifluoracetic acid mod- ified from the method originally described by Bi- eri et al. [20].

Results

Distribution of incorporated labelled arachidonate in

endothelial cell lipids

The incorporation of labelled arachidonate into endothelial cell lipids after 5 and 20 h of incuba- tion was studied. Radioactivity was found chiefly in the phosphatidylcholine and neutral lipid frac- tions; however, the distribution of incorporated radioactivity in various phospholipid classes was different after the two incubation periods (Table I). After 20 h of arachidonate labelling, there was a considerable shift of radioactivity from phos- phatidylcholine and phosphatidylinositol fractions to those of phosphatidylethanolamine and neutral lipids. This transfer of labelled arachidonate is

consistent with the observation reported by Brown and co-workers [21].

Uptake of (R,R,R)-a-tocopherol by endothelial cells When incubated with physiological (23.2 PM)

and pharmacological (92.8 PM) concentrations of (R, R, R)-a-tocopherol, incorporation of this vitamin by the endothelial cells was dose- and time-dependent. At 92.8 PM, tocopherol uptake by the cells was consistently 3-fold higher than at 23.2 PM, irrespective of whether the incubation time was 5 or 20 h (Table II). Within the same concentration of tocopherol-enriched medium, cel- lular uptake was doubled in the 20 h incubation when compared to the 5 h incubation (1.80 vs. 0.98 nmol at 92.8 PM and 0.56 vs. 0.32 nmol at 23.2 pm). The levels of tocopherol present in the culture medium did not have any toxic effect on

471

TABLE I

DISTRIBUTION OF [1-‘4C]ARACHIDONIC ACID IN VARIOUS LIPID CLASSES OF HUMAN ENDOTHELIAL CELLS

AFTER 5 AND 20 h OF INCUBATION

Confluent endothelial cells were Iabelled with 0.25 PCi of [I-i4C]arachidonic acid (3.8 FM) for 5 or 20 h. The dishes were rinsed with

Ringer-Tyrode buffer (pH 7.4) containing 0.25% BSA; cells were scraped in ice-cold methanol and cellular lipids were extracted and

quantitated as described in Experimental procedures. Results are means+S.D. of three dishes.

Radioactivity (dpm x 10m4/dish)

5h (% total) 20 h (% total)

Total uptake by cells

Phosphatidylcholine

Phosphatidylethanolamine

Phosphatidylinositol Neutral lipid + free arachidonate

19.13*0.07 (100) 35.48 f 1.67 (100) 8.94 f 0.05 (46.7) 11.75 + 0.42 (33.1)

1.71*0.75 (8.9) 6.75kO.83 (17.3)

2.56 f 0.06 (13.4) 2.07 f 0.27 (5.8) 3.53kO.08 (18.5) 11.31 io.45 (31.9)

the cells, as judged by unaltered cell number and viability.

Effect of (R,R,R)-a-tocopherol on arachidonate re-

lease and cellular phospholipids: Expt. 1

When confluent monolayers were incubated with different levels of tocopherol for 20 h and then labelled with 0.25 PCi of arachidonate for 5 h, total uptake of radioactivity by cells was not significantly altered by the amount of tocopherol added to the medium ((19.11 f 1.4). 104, (19.46 k

TABLE II

EFFECT OF ADDED TOCOPHEROL ON CELLULAR

TOCOPHEROL CONTENT AND CELL NUMBER

Values are means f S.D. of three dishes; nd., not detectable. In

Expt. 1, confluent cells were incubated with different tocopherol

concentrations for 20 h followed by labelling with 0.25 PCi of

[l-‘4C]arachidonic acid for 5 h. In Expt. 2, cells were labelled

with 0.25 PCi of arachidonate for 20 h followed by incubation

with tocopherol for 5 h. Total lipids were extracted from cells

and all-rat-a-tocopherol acetate was added as internal stan-

dard. Tocopherol was quantitated by an HPLC method, as

described in Experimental procedures.

Tocopherol

in medium (PM)

0

23.2

92.8

Tocopherol in cells. nmol/dish

(cell number X 10 ‘/dish)

Expt. 1 Expt. 2

n.d. nd.

(3.70 f 0.31) (3.62k0.25)

0.56 f 0.08 0.32 f 0.04

(3.40 * 0.40) (3.74iO.11)

1.8OkO.10 0.98 + 0.09

(3.63 kO.24) (3.53 f 0.35)

1.1). lo4 and (18.23 k 0.5). lo4 dpm which corre- sponded to 0, 23.2 and 92.8 PM tocopherol, re- spectively). Agonist-stimulated arachidonate re- lease showed that ionophore A23187 (1 PM) in- duced a significantly higher arachidonate release

than thrombin (2.5 U/ml). Arachidonate release was significantly enhanced in the tocopherol-en- riched cells, and this increase was directly propor- tional to the concentrations of tocopherol present in the medium (Fig. 1A). This potentiating effect of tocopherol on arachidonate release was simi- larly observed with the two agonists used. In fact,

in the absence of agonist, there was a small but significant increase in the release of arachidonate from cells that were enriched with the high con- centration of tocopherol (92.8 PM).

Analysis of the radioactivity that remained in the cells after agonist stimulation revealed that the principal source of arachidonate was released from

phosphatidylcholine, which was followed by phos- phatidylethanolamine (Fig. 1B). The radioactivity retained by phosphatidylinositol was unchanged by the tocopherol status of the cells, but throm- bin-stimulation caused a significantly lower radio- activity in this phospholipid when compared to ionophore or control. Data from this experiment clearly indicate that tocopherol caused a signifi- cant increase of arachidonate release from phos- phatidylcholine and phosphatidylethanolamine in agonist-stimulated endothelial cells.

Effect of (R, R,R)-a-tocopherol on arachidonate re- lease and cellular phospholipids: Expt. 2

In order to determine whether the release of

412

arachidonate was a direct effect of tocopherol on cellular lipids or whether the presence of tocopherol in the cells discriminated the incor- poration of arachidonate into a particular lipid, Expt. 2 was conducted, in which the incubation

order of tocopherol and arachidonate was reversed from that of Expt. 1. The cells were prelabelled with [l-‘4C]arachidonic acid for 20 h and then incubated with varying tocopherol concentrations

for 5 h. During this 5 h of tocopherol incubation, about 16% of the incorporated arachidonate radio- activity was released into the culture medium (be- fore tocopherol incubation: (42.5 + 1.4) . lo4 vs. after 5 h of tocopherol incubation; (35.3 + 3.8).

J , I ,

0 23.2 92.9

( R,R,R 1 Q - TOCOPHEROL ( JIM 1

104, (35.3 + 3.9) . lo4 and (35.0 L- 3.6) . lo4

dpm/dish for 0, 23.2 and 92.8 PM of tocopherol in medium, respectively). Analysis of cellular lipids showed that triacylglycerol was the principal source of this arachidonate release (data not shown).

When these cells were stimulated with iono- phore and thrombin for 10 min, the tocopherol-in-

duced enhancement of arachidonate release was again observed. However, under these experimen-

tal conditions, the increased release of arachi- donate leveled off at 23.2 PM of tocopherol prein- cubation and, in contrast to Expt. 1, no further increase of release could be detected at a higher (92.8 PM) level of tocopherol preincubation (Fig. 2A). This could be in part due to the shorter duration of tocopherol enrichment. The cellular

tocopherol uptake in this experiment was 0.32 and 0.98 nmol/dish for 23.2 and 92.8 PM of tocopherol in the medium. In contrast, uptake of cellular tocopherol in Expt. 1 (20 h of tocopherol incuba- tion) was much higher, reaching 0.56 and 1.80 nmol/dish, as illustrated in Table II. Alterna- tively, the longer period of arachidonate labelling may result in chain elongation of arachidonate to adrenic acid, which has been demonstrated by Rosenthal and Hill [17] to be resistant to agonist- induced release.

Fig. 1. Expt. 1, dose-dependent stimulation by (R,R,R)-a- tocopherol on agonist-induced arachidonic acid release and

radioactivity retained in cellular phospholipids. Confluent cells

were incubated with 0, 23.2 and 92.8 PM of (R,R,R)-a-

tocopherol-enriched medium for 20 h. Following the removal

of medium, each dish was radiolabelled with 0.25 PCi of

[l-‘4C]arachidonic acid for 5 h. The dishes were rinsed with

0.25% BSA in Ringer-Tyrode buffer (37 o C) and were stimu-

lated with Ringer-Tyrode buffer alone (0) or with thrombin,

2.5 U/ml (m) or ionophore A23187 (A) as described in Experi-

mental procedures. The incubation medium was removed after

10 min and an aliquot was taken for quantitation of total radioactivity. Total lipids in cells were extracted and phos-

pholipids were resolved by thin-layer chromatography using

CHCl,/methanol/acetic acid/H,0 (85 : 15 : 10: 3.5. v/v) as

solvent system. Identification of phospholipid class was

achieved by I, vapor exposure and autoradiography with

authentic phospholipid standards. Different bands from TLC

were scraped and radioactivity was detected by scintillation counting. Values are means f SD. of three dishes, PC, phos-

phatidylcholine; PE, phosphatidylethanolamine and PI, phos- phatidylinositol.

1 I 1 92.8

0 0 23.2

( R,R,R ) a- TOCOPHEROL ( JJM )

Fig. 2. (A) Expt. 2, stimulation of arachidonic acid release by (R,R,R)-a-tocopherol . .

I 1 I ,

0 23.2 92.8

( R,R,R ) a-TOCOPHEROL ( pfd )

Confluent endothelial monolayer were labelled with 0.25 PCi of [I-“Clarachidonic acid for 20 h. After removal of medium and rising with 0.258 BSA in Ringer-Tyrode buffer

(37OC). the cells were incubated with 0, 23.2 and 92.8 PM of (R,R,R)-a-tocopherol-enriched medium for an additional 5 h. The

dishes were rinsed with the above buffer and cells were stimulated with Ringer-Tyrode buffer alone (0) or the thrombin, 2.5 U/ml

(H) or ionophore A23187 (A) as described under Experimental procedures. The incubation medium was removed after 10 min and

radioactivity was determined by scintillation counting. Values are means i: SD. of three dishes. (B) Expt. 2, radioactivity in cellular

phospholipids after agonist stimulation. Experimental conditions were identical to in (A). After cells were stimulated with

Ringer-Tyrode buffer alone (0) or with thrombin (m) or ionophore A23187 (A), they were harvested by scraping in ice-cold methanol

and cellular lipids were extracted and different phospholipid classes were resolved by thin-layer chromatography as described in Fig.

I. PC, phosphatidylcholine; PE. phosphatidylethanolamine and PI, phosphatidylinositol. Values are means& S.D. of three dishes.

The radioactivity retained in different phos- confirmed the previous finding that (R, R,R)-a- pholipid classes is presented in Fig. 2B. The loss tocopherol potentiates arachidonate release from of radioactivity by agonist stimulation was once phosphatidylcholine and phosphatidylethanol- again found in phosphatidylcholine and phos- amine, and the magnitude of radioactivity lost phatidylethanolamine, and the magnitude of ra- corresponded with the amount of arachidonate dioactivity lost corresponded with the amount of release. Results from this experiment confirmed

arachidonate release. Results from this experiment the previous finding that (R,R,R)-cr-tocopherol

474

TABLE III

DISTRIBUTION OF RADIOACTIVITY IN CELLULAR NEUTRAL LIPIDS AFTER AGONIST STIMULATION

Values are means f SD. of three dishes. An aliquot of cellular lipid extract from Expt. 2 (Fig. 2A and B) was used for quantitation of

neutral lipids by thin-layer chromatography, as described under Experimental procedures.

Tocopherol concn. Distribution of radioactivity (dpm x 10K3/dish)

in media (PM): 0 23.2 92.8

Free arachidonate

Control

Thrombin

A231 87

Diacylglycerol

Control

Thrombin

A23187

Triacylglycerol

Control

Thrombin

A23187

0.81+0.02

2.36 + 0.14

1.75 * 0.01

2.93 k 0.19 2.90 f 0.19

5.80+0.23 6.02 k 0.50

4.6OkO.28 4.82 k 0.20

49.21 f 2.82

43.60 + 2.62

38.63 k 0.90

0.79 f 0.01

2.35 kO.14

1.72kO.10

44.11* 4.49

41.90f1.80

38.33 f 3.53

0.77 * 0.02

2.50 + 0.23

1.79*0.01

2.81+0.16

5.69kO.17

4.65kO.18

41.93 * 0.71

39.10+0.80

34.60 * 1 .OO

potentiates arachidonate release from phosphati- dylcholine and phosphatidylethanolamine.

Distribution of radioactivity in cellular neutral lipids after agonist stimulation

Aliquots of cellular lipid extract from Expt. 2 were subjected to quantitation of radiolabelled neutral lipids (Table III). Radioactivity in tri- acylglycerol, diacylglycerol and free arachidonate was not affected by the tocopherol status of the cells. While the radioactivity associated with the triacylglycerol fraction remained unaffected by agonist stimulation, the amount of radioactivity found in diacylglycerol and free arachidonate was significantly higher from cells that were stimulated

by thrombin and ionophore.

Discussion

The release of arachidonic acid in intact cells involves several deacylation mechanisms which differ not only in fatty acid specificities, but also in the precursor phospholipid classes. It is now generally accepted that phosphatidylinositol turnover from phospholipase C activation initiates the release reaction. However, direct deacylation from phosphatidylcholine and phosphatidyl-

ethanolamine by a phospholipase A, activity con- tributes a significant mass of arachidonate re- leased [5-8,22-241. In human endothelial cells, however, phospholipase A, activity was also re- ported to hydrolyze phosphatidylinositol, phos- phatidylcholine and phosphatidylethanolamine

]7,81. Results from these experiments clearly show

that arachidonate release can be potentiated by enrichment of endothelial cells with ( R, R, R )-a-

tocopherol. Judging from the loss of cellular ra- dioactivity from phosphatidylcholine and phos- phatidylethanolamine observed in this study, the involvement of phospholipase A z activity is likely. The mechanism by which tocopherol elicited such a response in endothelial cells is unclear at pre-

sent, and merits further investigation. While the tocopherol enrichment effect observed herein is compatible with reported stimulation of tocopherol on prostaglandin I, release [10,13], it nevertheless is in strong contrast to previously reported ob- servations from our laboratory, which showed that dietary vitamin E-enriched rat platelets synthe- sized less thromboxane A, upon stimulation with thrombin [9]. Similarly, phospholipase A, activity in rat platelets and myocardium was inhibited by dietary vitamin E enrichment or by addition of

tocopherol in vitro [14,15]. It appears that the regulation of the arachidonate release reaction by tocopherol is tissue- and probably species-specific.

It is tempting to speculate that tocopherol may affect a mediator which controls acylhydrolase

activity. In this respect, the lipoxygenase product 15-hydroperoxy-5,8,11,13-eicosatetraenoic acid, which was reported to stimulate phospholipase A, [25], could possibly be the mediator, because tocopherol-enriched human platelets have been shown to elicit a transitory increase in 12-hydro- peroxy-5,8,10,14-eicosatetraenoic acid and 12-hy-

droxy-5,8,10,14-eicosatetraenoic acid when in- cubated with labelled arachidonate [26]. With its antioxidant property, tocopherol should affect the cyclooxygenase, which is stimulated by low levels of organic antioxidants but inhibited by high anti- oxidant concentrations [27,28]. Deactivation of

cyclooxygenase should allow more substrate to be available for the lipoxygenase pathway. Irrespec- tive of the mechanism, the observation that physi- ological and pharmacological concentrations of tocopherol can increase arachidonate release in

human endothelial cells may have direct relevance and implication to clinical situations such as thrombosis, inflammation and endothelial injury.

Acknowledgement

This work was supported by Grant MA-7626 from the Medical Research Council of Canada. We wish to thank David Pyke for his technical assistance and Julie Normand for preparation of the manuscript.

References

1 Weksler, B.B., Ley, C.W. and Jaffe, E.A. (1978) J. Clin.

Invest. 62, 923-930.

2 Brotherton, A.F.A. and Hoak, J.C. (1983) J. Clin. Invest.

72, 1255-1261.

3 Prescott, S.M., Zimmerman, G.A. and McIntyre, T.M.

(1984) Proc. Natl. Acad. Sci. USA 81, 3534-3538.

4 McIntyre, T.M., Zimmerman, G.A., Satoh, K. and Prescott,

SM. (1985) J. Clin. Invest. 76, 271-280.

8 Hong, S.L., McLanghlin, N.J., Tseng, C.Y. and Patton, G.

(1985) Thromb. Res. 38. l-10.

9

10

Hamelin, S.-St. and Chan, A.C. (1983) Lipids 18, 267-269.

Chan, A.C. and Leith, M.E. (1981) Am. J. Clin. Nun. 34,

2341-2347.

11

12

13

Okuma, M., Takayama, H. and Uchino, H. (1980) Pros-

taglandins 19, 527-536.

Karpen, C.W., Merola, A.J., Trewyn, R.W., Cornwell, D.G.

and Panganamala, R.V. (1981) Prostaglandins 22, 651-661.

Boogaerts, M.A., Van de Broeck, J., Deckmyn, H., Vermy-

len, J. and Verwilghen, R.L. (1984) Thromb. Haemost. 51,

89-92.

14

15

16

17

Douglas, C.E., Chan, A.C. and Choy, P.C. (1986) B&him.

Biophys. Acta 876, 639-645.

Cao, Y.Z., Karmin, O., Choy, P.C. and Chan, A.C. (1987)

B&hem. J. 247, 135-140.

Jaffe, J.A. (1984) Biology of Endothelial Cells, pp. l-13,

Martinus Nijhoff, Boston.

Rosenthal, M.D. and Hill, J.R. (1986) B&him. Biophys.

Acta 875. 382-391.

18

19

Jaffe, J.A., Nachman, R.L., Becker, C.G. and Minick, C.R.

(1986) J. Clin. Invest. 52, 2745-2756.

Bligh. E.G. and Dyer, W.J. (1959) Can. J. Biochem. Phys-

iol. 37, 911-917.

20

21

22

23

Bieri, J.G., Tolliver, T.J. and Catignani, G.L. (1979) Am. J.

Clin. Nutr. 32, 2143-2149.

Brown, M.L., Jakubowski, J.A., Leventis, L.L. and Deykin,

D. (1987) Biochim. Biophys. Acta 921, 159-166.

Thomas, J.M.F., Hullin, F., Chap, H. and Douste-Blazy. L.

(1984) Thromb. Res. 34, 117-123.

Willis, A.L. and Smith, J.B. (1981) Prog. Lipid Res. 20,

387-406.

24

25

McKean, M.L., Smith, J.B., Silver, M.J. and Ahem, D.

(1981) prog. Lipid Res. 20, 435-438.

Hong, S.L., Carty, T. and Deykin, D. (1980) J. Biol. Chem.

255, 9538-9540.

475

5 Bell, R.L., Kennerly, D.A.. Stanford, N. and Majerus, P.W.

(1979) Proc. Natl. Acad. Sci. USA 76, 3238-3241.

6 Irvine, R.F. (1982) Biochem. J. 204, 3-16.

7 Hong, S.L. and Deykin, D. (1982) J. Biol. Chem. 257,

7151-7154.

26 Ghan, A.C., Raynor, C., Patrick, J. and Boland, M. (1986)

Am. J. Clin. Nutr. 44, 278-282.

27 Egan, R.W., Gale, P.H., Baptista, E.M., Kennicott, K.L.,

Vanden Heuvel, W.J.A., Walker, R.W., Fagemess, P.E. and

Kuehl, F.A. (1981) J. Biol. Chem. 256, 7352-7361.

28 Egan, R.W., Gale, P.H., Beveridge, G.C., Marrett, L.J. and

Kuehl, F.A. (1980) Adv. Prostaglandin Thromboxane Res.

6. 153-155.