Eicosapentaenoic acid and docosahexaenoic acid suppress the proliferation of vascular smooth muscle cells

ATHEROSCLEROSIS

Atherosclerosis 104 (1993) 95-103

Eicosapentaenoic acid and docosahexaenoic acid suppress the proliferation of vascular smooth muscle cells T. Shiina*, T. Terano, J. Saito, Y. Tamura, S. Yoshida The Second Department of Internal Medicine. Chiba University School of Medicine, 1-8-I Inohana Chuo-Ku, 260 Jopan

(Received I September 1992; revision received 19 July 1993; accepted 28 July 1993)

Abstract

Eicosapentaenoic acid, which is one of the n-3 polyunsaturated fatty acids (PUFA), is reported to exert its antithrombotic and anti-atherogenic effect partly through the modulation of vascular cell functions. Vascular smooth muscle cell (VSMC) proliferation plays an important role in the pathogenesis of atherosclerosis. We reported the differential effect of various PUFA on VSMC proliferation. First we established a method for preparing PUFA rich cells in culture to mimic the in vivo situation using PUFA triacylglycerol emulsion. Using these fatty acid rich cells, we found that only EPA and docosahexaenoic acid, although less potent than EPA, inhibited the proliferation of VSMC among the fatty acids tested. This effect of EPA was reversed by the addition of anti-oxidants. It is suggested that production of the oxidized species at a low concentration from EPA inhibited the proliferation of VSMC. This anti-proliferative effect of EPA and DHA on VSMC could partly explain the anti-atherosclerotic effect of marine lipids. Key words: Eicosapentaenoic Lipid peroxides

acid; Docosahexaenoic

acid; Vascular smooth muscle cells; Anti-proliferative

1. Introduction Eicosapentaenoic

acid (EPA),

which

is one of

the n-3 polyunsaturated fatty acids (PUFA) and which is plentiful in marine lipids, has been reported to exert its anti-atherosclerotic effect through the modulation of various cell functions related to atherosclerosis. Greenland Eskimos, whose intake includes a large amount of marine lipids, have a low incidence of cardiovascular diseases [ 1,2]. Oral administration of EPA pre-

* Corresponding author.

effect;

vented the reocclusion of coronary artery after angioplasty [3]. EPA improved the signs and symptoms of arteriosclerotic obliteration (ASO) [4]. These epidemiological and clinical data supported the anti-atherosclerotic and anti-thrombotic effect of EPA. The following mechanisms have been reported: first is the lipid lowering effect of EPA [5,6]; second is inhibition of platelet aggregability [7,8]; third is the enhanced production of vasodilator substance including PG12 [9- 111.Vascular smooth muscle cell (VSMC) proliferation is an important component in the pathophysiology of atherosclerosis, because it may be associated with lipid accumulation [ 121.

0021-9150/931$06.00 0 1993 Elszier Scientific Publishers Ireland Ltd. All rights reserved SSDI 0021-9150(93)05131-N

96

Several investigators have reported that polyunsaturated fatty acids (PUFA) generally inhibit the proliferation of cells in culture mainly through the lipid peroxidation process [ 13- 151. But in some reports, low concentrations of arachidonic acid actually promoted the proliferation of VSMC [16]. These differences might arise from the quantity and quality of lipid peroxides produced and the nature of the fatty acids, some of which are a good substrate for cyclooxygenase but others are not. Recently, a new hypothesis for the inhibitory effect of VSMC proliferation was presented by Fox et al. [ 171. They reported that the max-EPA might suppress the SMC proliferation by decreasing the production of the platelet-derived growth factor (PDGF) from endothelial cells which indicated that a free radical oxidative process was required for this inhibition. Others have proposed that the enhanced production of PGE2 or PGIz are involved in the inhibition of smooth muscle cell proliferation [ 18,191. In almost all the experiments, free PUFAs were used to make PUFA rich cells in culture [ 13- 151. Free PUFAs are easily oxidized in the medium and in some cases the hydroperoxide level may increase to cytotoxic levels in the medium and the cells. It is essential to ensure that the method for incorporating the PUFA into cells is as close to in vivo conditions as possible and avoids excessive peroxidation in order to confirm the physiological effect of PUFA on cell proliferation. For this purpose, triglyceride forms of EPA, docosahexaenoic acid (DHA), oleic acid and linoleic acid were used in order to modulate the fatty acid composition of VSMC. Using these cells, the effect of PUFA, especially of the n-3 fatty acids, on the proliferation of VSMC was studied. Also the mechanisms of the antiproliferative effect of n-3 PUFA were investigated with special reference to lipid peroxide formation from PUFA.

T. Shiina

et al. /Atherosclerosis

104 (1993)

95-103

synthesized by chemical condensation of glycerol and free EPA (prepared by hydrolysis of purified EPA ethyl ester by the Central Research Laboratories of Nipponsuisan Kaisha, Hachioji, Tokyo) was emulsified by Biotron mixer (Biotron, Switzerland) and then by high pressure homogenizer (Gaulin, USA) with phosphatidylcholine (PC), glycerol, tocopherol and water. It was filtered by cartridge and membrane filters to remove any particles over 1 pm in diameter and was sterilized under high pressure vapor and then packed. All procedures were performed under nitrogen gas. The mean particle size was 286 nm measured by a particle size analyzer (Autoalzer, Malvern, UK). Each emulsion contained 20.6% fatty acids triacylglycerol, 2.3% glycerol, 1.2% yolk egg PC and 0.004% tocopherol. Dulbecco’s modified Eagles medium (DMEM) was purchased from Nissui Seiyaku Co Ltd. (Tokyo) and fetal calf serum was purchased from Gibco Laboratories (New York, USA). Butylated hydroxytoluene (BHT) was purchased from Wako Pure Chemicals Co., Ltd (Tokyo). Alphatocopherol (vitamin E), indomethacin, thiobarbituric acid (TBA), FeCls. HZ0 and tetraethoxypropane were all purchased from Sigma Chemical Co (St. Louis, MO). [Methyl-3H]thymidine (2.59-3.33 TBq/mmol) was purchased from DupontNEN Research Products (Boston, MA). 2.2. Cell culture

Medial smooth muscle cells were isolated from the thoracic aorta of 12-week-old WKY and vascular smooth muscle cells (VSMC) in culture were prepared by the explant method according to Ross et al. [20] as previously reported [21]. VSMC were cultured in DMEM with 10% fetal calf serum, 400 &ml gentamicin and 0.67 mg/ml of sodium bicarbonate (Growth Medium) in humidified 95% sir/5% CO2 at 37°C. Cells of passage from the 5th to the 10th were used in this experiment.

2. Materials and methods 2.3. Fatty acid composition 2.1. Materials

Fatty acid emulsions of EPA (20:5, n-3), DHA(22:6, n-3), oleic acid (18:1, n-9) and linoleic acid (18:2, n-6) were prepared as follows; e.g. 1,2,3-trieicosapentaenoyl glycerol (98% pure) was

Cells at 70-80% confluence were made quiescent by incubation in DMEM without containing FCS for 24 h. Then these cells were cultured with a growth medium supplemented with 20-160 PM of the TG form of fatty acids (EPA, DHA, oleic

T. Shiina et al. /Atherosclerosis

91

104 (1993) 95-103

acid and linoleic acid) for 13 h. The cells were removed by rubber policeman and the fatty acid composition of the total phospholipid fraction of VSMC was measured by gas liquid chromatography as previously reported [22]. 2.4. Incorporation of L3H]thymidine into DNA DNA synthesis was measured by incorporation of [“Hlthymidine into VSMC. Cells (10 000) were seeded into 24-well plates and incubated for about 60 h with growth medium and then the medium was switched to serum free DMEM and cultured for a further 24 h. The cell stage was synchronized at Go by these procedures. This medium was changed to a growth medium supplemented with 20-160 PM of fatty acid triglycerides. After a 13 h incubation, this culture medium was aspirated and 1 ml of the growth medium containing 1 &i of [3H]thymidine was added to each well and incubated for a further 4 h. The [3H]thymidine containing medium was aspirated and the cells were washed with 1 ml of phosphate buffer saline (PBS). Then 1 ml of ice cold trichloroacetic acid (TCA) was added to each well and the plates were kept at 0°C for 20 min. The TCA was exchanged by 1 ml of the new TCA and aspirated again. The cells were washed by 1 ml of PBS twice and then 1 ml of 0.5 N KOH was added to each well. The plates were incubated at 37°C for 12 h and the radioactivity in the KOH solution was measured by a liquid scintillation counter. 2.5. The effect of indomethacin or antioxidants on the cell growth

Indomethacin, Vit E (10 and 100 PM) or BHT (1 and 10 PM) was added to the growth medium with fatty acids TG and incubated with VSMC at Go stage for 13 h. The effect of indomethacin or antioxidants on cell proliferation was evaluated by measuring the incorporation of [3H]thymidine for a further 4 h. 2.6. Determination of lipid peroxide levels

Lipid peroxides in the medium and in the cells supplemented with fatty acids for 13 h were assayed by the modified TBA procedure’ as previously reported [23]. A sample solution (0.1 ml) was mixed with ethanol (0.8 ml), 0.5 M

glycine-HCl buffer (pH 3.6, 1.5 ml) and 10 mM BHT (0.1 ml). Then 0.1 ml of 10 mM FeC13 and 1.5 ml of 0.5”/0 TBA were added to this solution and this mixture was heated for 15 min in boiling water, cooled on ice and then extracted with 4 ml of n-butanol. The mixture was centrifuged at 3000 rev./min for 10 min to separate the phases and the fluorescence intensity of the upper phase was determined at 535 nm (excitation) and 553 nm (emission) on a Hitachi F-2000 spectrophotometer. The data were presented as nmol MDA equivalent to a standard curve generated with tetraethoxypropane. Statistical comparisons were performed by the use of unpaired, two-tailed Student’s r-test. 3. Results

The fatty acid composition of VSMC before the treatment is shown in Table 1. C20:5, n-3 (EPA), C22:6, n-3 (DHA) and C20:4, n-6 (arachidonic acid) in the fractions of phospholipids of VSMC were 0.58, 2.03 and 5.75 mol%, respectively. The contents of EPA and C22:5, n-3 (docosapentaenoic acid, DPA) increased dose dependently by the addition of EPA from 20 to 160 PM as shown in Table 1A. The content of EPA increased to more than 350% of the pretreatment level after supplementation of 160 PM of EPA. Neither the content of DHA nor the arachidonic acid was influenced by the addition of EPA-TG. The content of DHA and EPA were increased dose dependently by the addition of DHA as shown in Table IS. However, no significant change was observed in the fatty acid composition by the addition of oleic acid or linoleic acid (Tables lC, 1D). Fig. 1 shows the effect of EPA, DHA, oleic acid and linoleic acid on [ 3H]thymidine incorporation into cells. Both EPA and DHA inhibited the [3H]thymidine incorporation dose dependently. This inhibitory activity was more prominent in EPA than DHA. Neither oleic nor linoleic acid suppressed the proliferation of VSMC. Therefore, we used EPA for further examination to clarify the mechanisms of n-3 fatty acids on the suppression of [ 3H]thymidine incorporation. Both Vit E and BHT dose dependently reversed the inhibitory activity of EPA on [“Hlthymidine

T. Shiina et al. /Atherosclerosis

98 Table 1 Change of fatty acids composition A: EPA

0

C 16:O C 18:O

C 22:s (n-3) C 22:6 (n-3)

14.8 12.6 31.2 1.3 5.9 0.6 1.4 2.10

B: DHA

0

C c C C C C C C

14.8 12.6 31.2 I.3 5.9 0.6 1.4 2.1

C C C C

l8:l 18:2 (n-6) 20:4 (n-6) 20:s (n-3)

l6:O 18:O l8:l IS:2 (n-6) 20:4 (n-6) 20:5 (n-3) 22:5 (n-3) 22:6 (n-3)

of phospholipids

f f f f f f f

1.3 0.1 2.4 0.2 0.9 0.1 0.2 ?? 0.1

15.2 12.7 28.7 1.4 6.8 1.1

f 1.3 f 0.1 f 2.4 ?? 0.2 * 0.9 f 0.1 ?? 0.2 f 0.1

14.8 12.8 28.7 I.4 6.9 0.8 1.7 2.7

C l6:O c l8:O

14.8 * 1.3 12.6 ?? 0.1

C C C C

C 22:5 (n-3) C 22:6 (n-3)

31.2 2.4 5.9 0.6 I.4 2.1

22.5 10.9 22.1 2.4 5.1 0.9 1.9 2.1

D: linoleic acid

0

C l6:O C 18:O

14.8 12.6 31.2 2.4 5.9 0.6

C 22:5 (n-3) C 22:6 (n-3)

160

f f f f f f

2.9 0.6 3.6 0.4 0.4 0.2

40

40

l8:l 18:2 (n-6) 20:4 (n-6) 20:5 (n-3)

80

1.7 f 0.1 2.3 * 0.7

0

c C C C

of various

40

C: oleic acid

18:l 1812 (n-6) 20:4 (n-6) 20:5 (n-3)

of VSMC by the addition

f f f f f *

1.4 0.6 0.9 0.1 0.2 0.1

I.3 0.1 2.4 0.6 0.9 0.1

1.4 * 0.2 2.1 f 0.1

22.4 12.1 24.0 2.4 5.9 1.3

f f f f f f f f

I.1 0.1 1.4 0.1 0.6 0.18 0.1 0.1

80 f f f f f f f f

1.4 0.3 2.8 0.1 0.5 0.3 0.1 0.1

f 4.7 f I.2 f 3.8 f 0.6 ?? 0.7 f 0.1 f 0.1 f 0.1

40 f f f f * *

14.7 12.8 30.2 1.3 6.0 1.4 1.6 2.1

15.4 12.7 29.6 I.4 6.4 0.7 I.5 3.1

?? 2.1 f 0.1 f 4.1 ?? 0.2 f 0.4 f 0.1 f 0.2 f 0.3**

3.9 0.3 4.5 0.6 0.6 0.4

1.5 f 0.2 2.1 f 0.2

14.6 12.4 28.3 I.5 6.6 0.8 1.6 3.6

80

160

27.5 f 7.7 10.5 f 1.4

18.1 12.0 21.2 2.4 5.9 I.1 2.0 2.2

24.0 2.3 3.8 0.8 I.2 1.6

20.1 II.7 22.6 2.5 5.9 1.4

fatty acids

f f f f f

I.1 0.2 1.2 0.2 0.8 ?? 0.1* ?? 0.1 f 0.1

160

f * f * f f

3.3 0.6 I.1 0.3 0.4 0.4

ZIZ1.7 f 0.5 f 4.1 f 0.1 f 0.7 * 0.4 * 0.1 f 0.5**

f 2.7 f 0.5 ?? 4.7 f 0.6 f 0.2 f 0.3 ?? 0.3 f 0.8

160

80 + * f f f f

14.7 12.9 29.1 1.4 6.0 2.2 1.7 2.1

104 (1993) 95-103

f 2.2 4.9 0.6 0.6 0.4

18.6 * 1.7 12.8 * 0.2 22.8 f 4.9 2.1 ?? 0.7 6.5 ?? 0.9 1.2 ?? 0.4

2.0 f 0.4 2.3 f 0.1

2.0 * 0.4 2.4 f 0.1

?? 0.3

f f f f

The values shown are the mean f SE. of three cultures from one experiment. *P < 0.05, **P < 0.01, significantly different from zero dose values.

incorporation. However, the potency was different between Vit E and BHT. Although the BHT almost completely reversed the suppression of [3H]thymidine incorporation by EPA, only 20% reversal was observed by Vit E (Fig. 2). As shown in Fig. 3, suppression of [3H]thymi-

dine incorporation by EPA-TG was slightly, but significantly, reversed by the addition of indomethacin. Also the proliferation of VSMC by growth medium was enhanced by the addition of indomethacin. As shown in Fig. 4, supplementation of EPA (80

T. Shiinaet al./ Atherosclerosis 104 (1993) 95-103

99

X i

s 5

4 .-g rr s z CI,

800 600 400 200 0

EPA

?? J

Oleic acid

DHA 0

m

20

0

40

m

60

B

inoleic acid 16O(bM)

Fig. 1. Effect of the addition of fatty acids TG emulsion on VSMC proliferation measured by [jH]thymidine incorporation. Results are the mean + SE. of 4 separate experiments, each in triplicate. **P < 0.01. ***P < 0.001, significantly different from zero dose values.

PM) increased the lipid peroxides in the culture medium. Co-incubation of 1 PM of BHT slightly lowered the lipid peroxides and 10 PM of BHT completely suppressed the lipid peroxides in the medium (Fig. 4, upper panel). Co-incubation of Vit E with EPA-TG slightly suppressed the formation of lipid peroxides in the medium in a dose dependent manner (Fig. 4, lower panel). The lipid

- **

** Ic .-ilr’

)L

0

EPA(-)

10 100 VitE(,uM) EPA 80 PM

0

1 10 BHT CUM) EPA 80 yM

Fig. 2. Effect of Vit E or BHT on the suppression of VSMC proliferation by the addition of EPA-TG emulsion. Values shown are mean f SE. of 6-8 separate experiments, each in triplicate. *P < 0.05, **P < 0.01, significantly different from the absence of antioxidant.

peroxides in the cells showed a similar increase on the addition of EPA to those in the medium (data not shown) 4. Discussion The beneficial effect of EPA in the treatment of thrombo-atherosclerotic diseases, such as arterial occlusive diseases has been reported [3] and much attention has been given to its possible influence on the progression of atherosclerosis. Weiner et al. and Cahilli et al. reported that fish oil feeding prevented intimal thickening in the coronary arteries of hyperlipidemic swine and reduced thickening of autologous vein grafts in normal dogs, respectively, without modulating plasma lipoprotein profiles [24,25]. The direct effect of EPA on vascular cells has shown it would play an important role. So far, enhanced production of PG12 by EPA ingestion [9,11,26] and reduced sensitivity of artery by vasocontracting agents after EPA treatment have been reported [27]. This data forced us to investigate the effect of EPA on VSMC. Proliferation of VSMC and the progression of hypertrophy of vascular walls play an important role in the pathogenesis of atherosclerosis

1121. In general, PUFA have been reported to inhibit the proliferation of VSMC [ 13- 151, but the differ-

T. Shiina et al. /Atherosclerosis

loo

2, 7

104 (1993) 95-103

? 0

EPA(&V 0 IndomethacinCpM) 0

40

40

0

0

10

0

k?B&B

80

80

0

0

0

10

0

10

n = 6, mean +S.E. * P < 0.05. * * I'< 0.01

Fig. 3. Effect of indomethacin on the suppression of VSMC proliferation by the addition of EPA-TG emulsion. Values shown are mean * S.E. of 6 separate experiments, each in triplicate. *P < 0.05, **P < 0.01, significantly different with the absence of indomethacin.

ence of potency between PUFAs and the mechanisms which are involved have not yet been fully understood. In this study, we concentrated on the effect of n-3 PUFA. EPA and DHA were used as representatives of n-3 PUFA, and oleic acid (n-9 monounsaturated fatty acid) and linoleic acid (n-6 PUFA) were used as control fatty acids. Oleic acid has been widely used as a control to n-3 PUFA in both in vitro and in vivo studies [28]. Linoleic acid was also used as a representatiqe of n-6 PUFA. Arachidonic acid by itself is converted to prostaglandins, which might directly influence cell proliferation. Linoleic acid might be a better fatty acid as a representative of the n-6 series because linoleic acid might change the characteristics of the membrane but is not, itself, converted to prostaglandins. Fatty acids exist in extracellular fluid as forms of TG and phospholipids but not in a free form. In order to mimic the conditions which pertain to physiological extracellular fluid, TG-emulsified forms of EPA, DHA, oleic acid and linoleic acid were prepared and used in this study. In most of the experiments so far reported, a free type of PUFA was used to prepare PUFA rich cells [13-161. Free PUFAs might be easily oxidized and produce lipid peroxides [ 131. In the

physiological conditions, extracellular fluid contains lipid peroxides, but the level is quite low. After administration of highly purified EPA to humans for 4 weeks, the level of serum MDA did not increase [22]. We used a modified method to measure lipid peroxides in which FeC13 was added to accelerate the conversion of lipid peroxides to TBA reactive substance and BHT was added to prevent autooxidation of PUFA during TBA reaction. Although it might not be correct to compare our lipid peroxide levels with those normally found in human serum (measured by ordinary TBA), nevertheless the MDA level in the medium after addition of 80 PM of EPA was 2-6 nmol/ml in our system, similar to the values found in human serum MDA [29]. EPA treatment increased the content of EPA and docosapentaenoic acid (DPA) without affecting the content of DHA and arachidonic acid. This change of fatty acid composition was similar to those observed in vivo after EPA administration to rats. Lokkette et al. [27] reported a similar change of EPA content (from 0.4 to 3.0 mol%) of rat aorta after fish oil administration. In our previous experiment highly purified EPA at doses of 60 mg/kg body weight per day was administered to rats [26] and 8 weeks after EPA treatment, the

T. Shiina

et al. / A!herosclerosis

104 (1993)

101

95-103

EPA 0 EHT 0

B

0

0

2

10

12

TiAle(hodur)

Fig. 4. Effect of BHT (I and 10 CM, upper panel) or Vit E (IO and 100 p M, lower panel) on the elevation of lipid peroxides by the addition of EPA-TG (80 PM) in the medium of cultured VSMC. BHT (10 PM) almost completely suppressed the elevation of MDA by EPA, but 1 PM of BHT and Vit E (IO and 100 PM) slightly lowered the MDA level. The values shown are mean of 3 cultures from I experiment.

increase in the content of EPA in the total phospholipids of aorta or mesenteric artery was in a similar range (data not shown). This data suggested that this subtle change of EPA content obtained in vitro in culture was similar to that obtained after EPA treatment in vivo. [ 3H]Thymidine incorporation was inhibited dose dependently by both EPA and DHA, although the latter was less potent than the former, but not by either oleic or linoleic acids. This effect of n-3 PUFA was specific because linoleic acid and

oleic acid in the TG form were inactive. Also this inhibition was not induced by a cytotoxic cell condition, because we confirmed cell viability as follows: first, cells were morphologically intact and not stained by trypan blue; second, leucine incorporation, as a marker of protein synthesis, was not affected (data not shown). The high polyunsaturation level of the fatty acids makes these lipids readily susceptible to free radical oxidative processes [13]. To see whether the lipid peroxides produced by EPA-TG are involved in the proliferation of VSMC, Vit E (alpha-tocopherol) or BHT were used as radical scavengers and investigated the effect of antioxidants on the suppression of VSMC proliferation by EPA. Vit E is one of the natural antioxidants and the serum level is reported to be IO-50 PM. We used Vit E at concentrations of lo-100 PM. Both antioxidants by themselves exerted minimal effects, but the inhibition by EPA was partially reversed by the addition of Vit E and almost totally reversed by BHT in a dose dependent manner. Lipid peroxides in the culture medium and cells after EPA addition were also determined to clarify the different effects of BHT and Vit E. Lipid peroxides in the medium increased and reached a maximum level 6 h after the EPA addition and then gradually decreased until 13 h of incubation. This increase in lipid peroxides was almost completely suppressed by the addition of 10 PM of BHT but just 1 PM of BHT delayed the time taken to reach maximum level and slightly suppressed the maximum level of MDA as shown in Fig. 4(A). Co-incubation of Vit E also reduced the maximum level of lipid peroxides in a dose dependent fashion but could not completely suppress them (Fig. 4(B)). BHT was more powerful than Vit E so as to step up the inactive proliferation of VSMC by EPA. This difference starts from the separate effect of the two anti-oxidants in reducing the lipid peroxide levels. We do not know the exact reason why 100 PM of Vit E partially suppressed the lipid peroxides. In the experiments in which free forms of fatty acids were used to prepare fatty acid rich cells, co-addition of Vit E almost completely suppressed the formation of lipid peroxides [ 131. Therefore TG forms of fatty acids might be differently affected by Vit E from free forms of fatty

102

acids. Fox reported that MaxEPA, which contained both EPA and DHA in the form of TG, inhibited PDGF production from endothelial cells, and this inhibition was partially reversed by the addition of Vit E and almost totally reversed by BHT [ 171. These different effects of BHT and Vit E on PDGF formation in endothelial cells were similar to our observations. Fox et al. also used TG forms of fatty acids. All our data suggested that oxidized species present at a low concentration may be responsible for the inhibitory activity of EPA. Inhibition of [3H]thymidine incorporation by EPA was slightly but significantly reversed by the addition of indomethacin. PGE2 and PG12 are reported to have an inhibitory effect on cell proliferation through the enhanced production of c-AMP [l&19]. PG13 was not produced in rat aorta or rat VSMC from EPA [23,30] but the formation of PG12 in VSMC was increased during the incubation with EPA TG emulsion as previously reported [31]. This data indicated that indomethacin might work through the suppression of enhanced prostaglandins formation by EPA but there is a possibility that indomethacin by itself has an anti-oxidant property. The change in the fatty acid composition of cell membrane has been reported to be related to cell function [32,33]. There is a possibility that EPA may alter the physicochemical characteristics of the cell membrane [34,35] and might exert its antiproliferative properties through the modulation of signal transduction of growth factors. In summary, we have shown that the method for preparing fatty acid rich VSMC in culture is similar to the in vivo situation using fatty acid TG emulsions and have found that only EPA and DHA, although less potent than EPA, inhibited the proliferation of VSMC mainly through the formation of an oxidized species at a low concentration. This effect of EPA and DHA could partly explain the anti-atherosclerotic effect of marine lipids. Further detailed study is now in progress in our laboratory to investigate the effect of EPA and EPA-derived minimal lipid peroxides on intracellular signal transduction with regard to growth factors.

T. Shiina et al. /Atherosclerosis

104 (1993) 95-103

5. Acknowledgements We would like to thank Mrs. Emiko Hatakeyama for her technical assistance and Dr. Satoru Kobayashi for his useful suggestions. We are indebted to Dr. A. Ozawa of the Central Research Laboratories of Nipponsuisan Kaisha for providing us with the purified TG form of fatty acids and for measuring the fatty acids composition of membrane phospholipids and to the Central Research Laboratories of Midorijuji Pharmaceutical Company for providing us with emulsified fatty acids. This study was supported by a Grant-in-Aid for Scientific Research by The Ministry of Education, Science and Culture of Japan (grant No. 02454505). 6. References Kromann, N. and Green, A., Epidemiological studies in the Upernavik district, Greenland: incidence of some chronic disease 1950-74, Acta Med Stand., 208 (1980) 401. Dyerberg, J., Bang, H.O., Stoffersen, E., Moncada, S. and Vane, J.R., Eicosapentaenoic acid and prevention of thrombosis and atherosclerosis, Lancet, 2 (1978) 117. Gregory, J.D., Jeffery, J.P., Egerton, K.B. et al., Reduction in the rate of early stenosis after coronary angioplasty by a diet supplemented with n-3 fatty acids, N. Engl. J Med., 22 (1988) 733. Sakurai, K., Tanabe, T., Mishima, Y., Sakaguchi, S., Katsumura, T., Kusaba, A. and Sakuma, A., Clinical evaluation of MND-21 on chronic arterial occlusion - double blind study in comparison with ticlopidine, J. Jap. Coil. Angiol., 28 (1988) 597. Thorngren, N. and Gustafson, A., Effect of II week increase in dietary eicosapentaenoic acid on bleeding time, lipids, and platelet aggregation, Lancet, 2 (198 I) 1190. Hirai, A., Terano, T., Tamura, Y. and Yoshida, S., Eicosapentaenoic acid and adult disease in Japan: epidemiological and clinical aspects, J. Intern. Med., 225 (Suppl.) (1989) 69. Siess, W., Roth, B., Sehere, B., Kurzman, B. and Weber, PC., Platelet membrane fatty acids, platelet aggregation and thromboxane formation during a mackerel diet, Lancet, I (1980) 441. Brox, J.H., Killie, J., Gunnes, S. and Nordoy, A., The effect of cod liver oil and corn oil on platelet and vessel wall in man, Thromb. Haemost., 46 (1981) 604. Fischer, S. and Weber, P.C., The prostacyclin/thromboxane balance is favorably shifted in Greenland Eskimo, Prostaglandins, 32 (1986) 235.

T. Shiina et al. /Atherosclerosis IO

II

12 13

104 (1993) 95-103

103

Shimokawa, H. and Vanhoutte, P.M., Dietary n3 fatty acids and endothelium-dependent relaxation in porcine coronary arteries, Am. J. Physiol., 256H (1989) 968. Abeywardens, M.Y.. Fisher, S., Schweer. H. and Charneck, J., In vivo formation of metabolites of prostaglandin I1 and I, in the marmoset monkey, following dietary supplementation with tuna fish oil, Biochim. Biophys. Acta, 1003 (1989) 161.

24

Ross, R., The pathogenesis of atherosclerosis; an up date, N. Engl. J. Med., 314 (1986) 488. Gavino, V.C., Miller, J.S., Ikharebha, S.O., Mile. G.E. and Cornwell, D.G., Effect of polyunsaturated fatty acids

26

25

and antioxidants on lipid peroxidation in tissue cultures, J. Lipid. Res., 22 (1981) 763. 14 Morisaki, N.. Sprecher, H., Mile, G.E. and Cornwell, D.G., Fatty acid specificity in the inhibition of cell proliferation and its relationship to lipid peroxidation and prostaglandin biosynthesis, Lipids, I7 (1982) 893. 15 Hutter, J.J., Gwebu, E.T., Pangnamala. R.V., Mile, G.E. and Cornwell, D.G., Fatty acids and their prostaglandin derivatives: Inhibitors of proliferation in aortic smooth muscle cells, Science, I97 (1977) 289. 16 Morisaki, N., Lindsey, J.A.. Mile, G.E. and Cornwell, D.G., Fatty acid metabolism and cell proliferation. 3. Effect of prostaglandin biosynthesis either from exogenous fatty acid or endogenous fatty acid release with hydralazine, Lipids, 18 (1983) 349. 17 Fox, P.L. and Dicorleto, P.E., Fish oils inhibit endothelial cell production of platelet derived growth factor like protein, Science, 241 (1988) 453. 18 Morisaki, N., Kanzaki, T., Motoyama, N., Saito, Y. and Yoshida, S., Ceil cycle-dependent inhibition of DNA synthesis by prostaglandin I, in cultured rabbit aortic smooth muscle cells, Atherosclerosis, 71 (1988) 165. Et inhibits 19 Nilsson, J. and Olsson, A.G., Prostaglandin DNA synthesis in arterial smooth muscle cell stimulated with platelet derived growth factor, Atherosclerosis, 53 (1984) 77. and the arter20 Ross, R. and Glomset, J.A., Atherosclerosis

27

ial smooth muscle cell, Science, 180 (1973) 1332. Hasunuma. K., Terano. T., Tamura, Y. and Yoshida, S., Formation of epoxyeicosatrienoic acids from arachidonic acid by cultured rat aortic smooth muscle cell microsomes, Prostagland. Leukotr. Essent. Fatty Acids., 42 (1991) 171. Terano, T.. Hirai, A., Hamazaki, T. et al., Effect of oral administration of highly purified eicosapentaenoic acid

34

on platelet function, blood viscosity and red cell deformity in healthy human subjects, Atherosclerosis, 46 (1983) 321. Hirai, A., Saito, J., Terano, T., Tamura, Y. and Yoshida, S., An improved method for the determination of lipid peroxides in biological samples containing a significant amount of polyunsaturated fatty acids, In: Yagi, K., Kondo, M., Niki, E. and Toshikawa, T. (Eds.), Oxygen Radicals, Excerpta Medica, Tokyo. 1992, p. 319.

35

21

22

23

28

29

30

31

32

33

Weiner, B.H.. Ockene, I.S., Levine, P.H. et al., Inhibition of atherosclerosis by cod liver oil in a hyperlipidemic swine model, N. Engl. J. Med., 315 (1986) 841. Landymore, R.W.. Kinley, C.E., Cooper, J.H.. MacAulay, M., Sheridan, B. and Cameron. B.N., Codliver oil in the prevention of intimal hyperplasia in autogenous vein grafts used for arterial bypass. J. Thorac. Cardiovasc. Surg., 89 (1985) 351. Hamazaki, T.. Hirai, A., Terano, T. et al., Effect of orally administered ethyl ester of eicosapentaenoic acid on PGI,-like substance production by rat aorta, Prostaglandins. 23 (1982) 557. Lockette, W.E., Webb. R.C., Gulp. B.R. and Pitt, B.. Vascular reactivity and high dietary eicosapentaenoic acid, Prostaglandins. 24 ( 1982) 63 I Mosconi. C., Colli, S., Medini, L. et al., Vitamin E influences the effect of fish oil on fatty acids and eicosanoid production in plasma and circulating cells in the rat. Biochem. Pharmacol.. 37 (1988) 3415. Marshall. P.J., Warso, M.A. and Lands, W.E.M., Selective microdetermination of lipid hydroperoxides, Anal. Biochem., 145 (1985) 192. Morita, 1.. Saito, Y.. Chang. W.C. and Murota, S.. Effects of purified eicosapentaenoic acid on arachidonic acid metabolism in cultured murine aortic smooth muscle cells, vessel walls and platelets. Lipids, I8 (1983) 142. Saito, J., Terano. T., Hirai, A.. Shiina. T.. Tamura, Y. and Yoshida. S., Enhancement of PGIz formation by eicosapentaenoic acid in rat vascular smooth muscle cells: Possible involvement of lipid peroxides. Jap. J. Pharmacol.. 58 (Suppl. 2) (1992) 287. Treen, M., Uauy. R.D., Jameson. D.M.. Thomas, V.L. and Hoffman, D.R., Effect of docosahexaenoic acid on membrane fluidity and function in intact cultured y-79 retinoblastoma cells, Arch. Biochem. Biophys.. 294 (1992) 564. Barry, H.G., Brown, T.J., Simon, 1. and Spector, A.A.. Effect of the membrane lipid environment on the properties of insulin receptors, Diabetes, 30 (1981) 773. Terano, T., Hirai, A.. Tamura, Y., Kumagai. A. and Yoshida, S., Effect of dietary supplementation of highly purified EPA and DHA on arachidonic acid metabolism in leukocytes and leukocyte function in healthy volunteers. In: Samuelsson, B.. Paoletti, R. and Ramuwell, P.W. (Eds.), Advances in Prostaglandin, Thromboxane, and Leukotriene Research, Raven Press, New York, Vol. 17, 1987, p. 880. Dratz, E.A. and Deese, A.J., The role of docosahexaenoic acid in biological membranes: examples from photoreceptors and model membrane bilayers. In: Simopoulos, A.P.. Kifer, R.P. and Martin, R.E. (Eds.), Polyunsaturated Fatty Acids in Sea Press, New York, 1986, p. 319.

Heart Food.

Effects of Academic