Biochimwa et Biophysica A('ta. 1081 (1991) 33-38

' 1991 ElsevierScience PublishersB.V. (Bi,3medicalDivision)0005-2760/91/$03.50 A DONIS 000527609100053D

Eicosapentaenoic acid ethyl ester as an antithrombotic agent: comparison to an extract of fish oil C a r o l M . W o j e n s k i , M e l v i n J. Silver a n d J e a n W a l k e r Cardeza Foundation for Hematologw Research. Department of Medwtne and Department of Pharma¢oh~gv. Jefferwm Medwal College. Thomas Jefferson Unirer~ttv, Philadelphia, P,4 t U.S.A. I

(Received 30 April 1990) Key words: Omega-3fatty acid: Platelet: Fish oil: Thrombosis:Eicosapentaenoicacid Evidence suggesting that dietary o:-3 polyunsaturated fatty, acids decrease the risk of thrombosis comes mainly from studies involving supplementation with large amounts (15-20 g / d a y ) of fish oil extract. We investigated the inhibition of platelet function by a moderate amount (4 g / d ) of ethyl eico~pentaenoate (E-EPA) compared to a concentrated fish oil extract (6 g / d ) when given as a supplement to an ordinary diet. We also determined the effects of t h e ~ supplements on platelet EPA incorporation, thromboxane synthesis, calcium mobilization and fibrinogen binding. After 4 weeks, both ~0-3 supplements increased the amount of EPA in platelet phospholipids. The fish oil extract, which contained docosahexaenoic acid (DHA), had increased the amount of DHA also. The total increase in t0-3 fatty acids was similar for both supplements. E-EPA decreased serum cholesterol by 13% and triacylglycerols, 35%; iucreased the bleeding time by 5"/% and the threshold dose of collagen needed to induce platelet aggregation by 46%. Thromboxane synthesis in response to collagen was decreased 65% by E-EPA. Thus, the dietary, supplement of pore E-EPA was more effective in limiting platelet reactivity than a concentrated fish oil extract providing an equivalent amount of 0:-3 fatty acids. As an antithrombotic agent, E-EPA should allow for reasonable daily dosis in long-term treatment of cardiovascular di~ase.


Blood platelets, vascular endothelial cells and smooth muscle cells all play important roles in hemostasis. They also participate in the slow and progressive process of atherogenesis which culminates in cardiovascular disease and thrombosis [1,2]. One of the mechanisms involved in the response of these cells to various stimuli is the synthesis of biologically potent prostaglandins and thromboxane from arachidonic acid (AA), a polyunsaturated fatty acid released from phospholipids of platelets and vascular cells. Epidemiologic studies comparing populations with high and low incidences of vascular disease and myocardial infarction suggested that diets providing large amounts of AA and its fatty acid precursors of the ,0-6 family might promote atherogenesis while diets providing high amounts of polyunsaturated fatty acids of the ~-3 family might be protective [3]. Many clinical and experimental studies now support the hypothesis that eicosapentaenoic acid

Correspondence: C.M. Wojenski. Cardeza Foundation for Hematologic Research, 1015 Walnut Street, Philadelphia, PA 19107-5099. U.S.A.

(EPA) and docosahexaenoic acid (DHA), found in high quantity only in marine foods, may inhibit the synthesis of prothrombotic eicosanoids and promote the synthesis of antithrombotic or neutral eicosanoids by displacing AA in blood and vessel wall cellular phospholipids [4]. As reviewed recently [5], most trials to date have involved use of diets containing large amounts of fish, or diets supplemented by cod liver oil or other fish oils {up to 50 ml daily), or large numbers of gelatin capsules containing a fish oil extract (10-20/d). While these studies have provided evidence that dietary (o-3 fatty acids decrease platelet responses and prolong the bleeding time, the regimens are too unpalatable to allow for good patient compliance for the long-term, possibly life-long, use of these supplements. In addition, while fish and fish oils can increase dietary ¢0-3 fatty acids, they also may contain fat soluble vitamins, saturated fatty acids, cholesterol and, possibly, heavy metal and organic chemical pollutants making inadvisable their long-term consumption in large quantity. It would be preferable to administer dietary supplements of pure (0-3 fatty acids in small quantities. Only a few studies have employed such fatty acids and none of these has included a direct comparison to the effect of fish oil [6-91.

The present study was designed to compare the effect on platelet function of an ethyl ester of EPA (E-EPA) to that of a concentrated fish oil extract or ethyl oleate added as a supplement to an ordinary diet. In addition. we investigated whether the increase in polyunsaturated o~-3 fatty acids in platelet phospholipids might inhibit platelet responses through events which precede synthesis of eicosanoids, such as mobilization of calcium ion and binding of fibrinogen [10,11]. Our results show that E-EPA is as effective as aspirin in prolonging the bleeding time and much more effective than concentrated fish oil in inhibiting platelet reactivity while ethyl oleatc has no effect. The marked increase in the bleeding time after dietary supplementation with E-EPA is most probably due to a combination of effects on platelet and vessel wall responses to vessel injury, but does not depend on decreased platelet calcium mobilization or fibrinogen binding induced by collagen. These results were reported in part at the 73rd Annual Meeting of the Federation of American Societies for Experimental Biology (FASEB J. (1989) 3, A334). Materials and Methods Protocol for dietary supplement,.ttion with fat O, acid~" The participants in this study were nine healthy male students and staff members 22 to 40 years of age. Each volunteer was instructed to maintain his usual dietary pattern throughout the study and not to use over-thecounter preparations containing aspirin or ibuprofen. A questionnaire was filled out by each volunteer at the beginning and again at the end of the study, to define the general eating habits, alcohol consumption and smoking habits of the participants. All participants denied taking any drugs known to affect platelet function during the 2 weeks preceding blood collection. As shown in Table !, the subjects were instructed t o supplement their ordinary diet for three periods of 4

TABLE I Protocolfor dtetaO"supph,mentation with o~-3fat O"ucid; Period I" 2 3' 4

Duration 4 week 5 ~eck 4 week

Supplement none Ethyl oleale none Res-Q1000

5 6

4 lll~lllth none 4 ~cek Ethyl EPA


S ~cek


Amount FAhcomposition 3 g/day 3.0 g oleic _

6 g/day

1.8 g eicosapentaenoic 1.2 g docosahexaenoic 0.6 g other ~-3 4 g/day 3.6 g eicosapentaenoic 0.4 g non w-3 -

" The results from these periods were not significanllydifferent from each other. The average appears as Control in T~tbles111,IV and V. h FA = fatty acid.

weeks each with gelatin capsules containing 1 g of ethyl oleate (3 g/d). or a concentrated extract of fish oil (6 g / d Res-Q 1000. including 1.8 g EPA and 1.2 g DHA), or ethyl EPA (4 g/day, >/90~ pure). Seven of the volunteers completed the entire study. Two participants. for personal reasons, tested only one of the c0-3 supplements each. None of the supplements contained cholesterol or vitamins A or D. The oleate and EPA capsules contained 2 I U / g and the Res-Q capsules 1 I U / g of vitamin E. an antioxidant. The periods of supplementation were separated by a minimum of 5 weeks without supplementation. This protocol was approved by the Jefferson Institutional Review Board in accord with an assurance filed with, and approved by. the U.S. Department of Health and Human Services. Source o f supplements Capsules containing ethyl oleate were provided by the R.P. Scherer Corp. (Troy. MI). Res-Q 1000 capsules were purchased from Res-Q International (Marlton, N J). The ethyl eicosapentaenoate was purchased from the Nippon Oil and Fats (Tokyo, Japan) through the Nissho lwai American (New York, NY) and encapsulated by R.P. Scherer, North America (St. Petersburg, FL). Blood sampling and routine tests Blood was drawn after an overnight fast. Blood cell counts and platelet size distribution were done on an Ortho Autocounter (model ELT-8/ds, Becton Dickenson, Braintree, MA). Serum triacylglycerols, total cholesterol and HDL cholesterol were measured in the hospital laboratory by automated methods. The bleeding time test was done by the Simplate I1 method employing a vertical incision with venostasis [12] following the periods during which the subjects ingested each of the fatty acid supplements. Platele t. aggregation tests For the whole blood studies, platelet aggregation in response to ADP (Sigma, St. Louis, MO) and collagen (Horme Reagent. Munich. F.R.G.) was recorded by the electrical impedance method [131 using a Whole Blood Lumi-Aggregometer (Chrono-Log, Havertown, PA). For studies in platelet-rich plasma (PRP), ;.he platelet count in the PRP was adjusted to 2.2 • 10S/ml with autologous platelet-free plasma to be equivalent to that in whole blood. The amount of ATP released in response to collagen was measured by adding a luciferin-luciferase preparation (Chronolume, Chrono-Log) to the sample after completion of aggregation [14]. After completion of the aggregation and ATP release tests the PRP sample was mixed with 1/10 vol. 0.1 M EDTA and centrifuged at 1 2 0 0 0 × g for 3 min. The supernatant platelet-free plasma was stored at - 2 0 ° C

until assayed for thromboxane B> the stable metabolite of thromboxane A,, by specific radioimmunoassay [15].

Preparation of platelet suspension Citrated PRP was recentrifug~',t for 5 rain at 200 × g to pellet contaminating erythrocytes. The supernatant PRP was cooled and acidified with citric acid (1 M, 9 /H/ml) before centrifugation at 1500 × g for 6 min at 5 ° C. The platelet pellet was washed in TSG-EDTA (pH 6.5) (0.9% NaCI, 0.015 M Tris-HCI, 0.1% dextrose. 5 mM EDTA) and centrifuged at 1000 × g for 6 rain. The platelets were resuspended (10'~,/ml) in the same buffer at pH 7.4 without EDTA and containing 0.2% fatty acid-free BSA (ICN Immuno Biologicals, Lisle, IL).

pellet was washed in Hepes-buffered saline (pH 7.6), containing 5 mM EDTA and 1 /tM prostaglandin E~ (PGE~, Biomol, Philadelphia, PA) and the platelets were IoadeZ with aequorin (Dr. John Blinks, Mayo Clinic, Rochester, MN) [21]. After loading, the platelets were washed in 3 ml Hepes-buffered saline without EDTA or PGE~ and centrifuged at 1000 × g for 10 min. Finally the platelets were resuspended at 2-10S/ml in Hepes-buffered saline without EDTA or PGE~ and calcium chloride was added to a concentration of 1 mM. Calcium mobilization was measured in a Platelet Ionized Calcium Aggregometer (PICA, Chronc,-Log) [22] in the presence of human fibrinogen (50 .ttg/ml) in response to U46619, collagen or ADP.

Total fatty atqds of platelet phospkolipid~

Ana(vsis of results

Duplicate aliquots (l-ml) of platelet suspension were extracted with chloroform/methanol (1:2) and the extracts were then evaporated under nitrogen at 37°C and stored in chloroform/methanol [16]. Butylated hydroxytoluene (50 /tg/ml) was added to the solvents as an antioxidant [17]. The individual phospholipids were separated on activated silica GHL (Analtech, Newark, DE) plates in chloroform/methanol/acetic acid/water (50 : 25 : 8 : 4, v / v ) and the phospholipid fatty acids were meth2,iated with freshly made 1 M HCI in methanol at 70°C for 2 h [181, after adding 5 ~g of heptadecanoic acid (NUChek Prep. Elysian, MN) as an internal standard. Gasliquid chromatography was carried out in a Varian Model 3700 Gas Chromatograph on a Supelco column 2-1721 or in a Hewlett-Packard 5830A on a Supelco column 2-1694, both packed with 10% SP2330 on 100/120 chromosorb WAW.

Statistical significance was determined by one way analysis of variance for repeated measures with Yate's Procedure to estimate missing variables.

Fibrinogen binding to platelets Platelet suspension was diluted to a platelet count of 2-108/ml with TSG (pH 7.4) containing 4 m g / m l BSA (Fraction V, 96-99% pure, Sigma. St. Louis, MO) and calcium chloride was added to yield a final concentration of 1 mM. Aliquots of platelet suspension were incubated at room temperature for 3 min with Aft, P, collagen, a thromhoxane analog, U46619 (Biomol, Philadelphia, PA), or saline without stirring, after which I"Sl-fibrinogen was added and incubation continued for 45 min. At the end of the incubation, platelet-bound labeled fibrinogen was separated by centrifugation over silicone oil [19,20]. Radioactivity in the platelet pelle! was measured in a Beckman 5500 Gamma Counter. Fibrinogen binding to platelets in the presence of saline solution alone was subtracted to calculate binding due to platelet activation by the aggregating agents.

Mobilization of cytoplasmic" calcium Citrated PRP was acidified with citric acid as before and centrifuged at 800 × g for 15 min. The platelet


All of the participants remained healthy throughout the study. The only reported side effect was mild gastric distress for a few days after beginning supplementation with the capsules (four of nine subjects). This was reported as often after the oleate capsule as after either of the 0~-3 supplements. No volunteer experienced increased bleeding or bruising with any supplement.

Changes in plutelet phospholipid fatty acids The amounts of t~-3 (EPA (20 : 5) and DHA (22 : 6)) and t0-6 (AA (20:4) and linoleic acid (18:2)) fatty acids esteri fled in platelet phosphatidylcholine (PC) and ethanolamine (PE) are shown in Table 11. Dietary supplementation with Res-Q increased the amount of EPA and DHA, while the E-EPA supplement only increased the EPA content of the phospholipids compared to the


Fffect t'¢dwtarv supplementationon phm@t phospholipidfitt(v actd~ ttg/lO'~ platelets EPA DHA {20:5) (22:0)

LA (18:2)

PC Control" Res-Q E*EPA

0.2+O.1 0.7+0.3 1.2±0.3"* 1.3+-0.2" 2.4±0.6** 11.9+-0,6

7.0+ 1.0 18.8+_3.8 7.0+1.2 17.4_+1.7 6.4±1.5 16.9_+3.2

PE Control Rcs-Q E-EPA

1).3+-0.2 I).9+-0.2 " 1.5±0.7"*

1.8+-0.5 21.O± 4.8 1.5_+0.4 17.4_+3.3 1.4+-0.5 15.6+-4.5


1.3 +-0.7 I.S +-0.5 1.1_+O.5

AA (20:4)

Control valueswereobtained beforedietary supplementation,Mean + S.D.. n = q. Difference fromControh * P < 0.05: * * P < 0.O1.

36 control period without dietary supplements. The A A content did not decrease significantly with either s u p p lement. The fatty acids h a d returned to control levels 8 weeks after s t o p p i n g s u p p l e m e n t a t i o n with E-EPA: EPA, 0.3 + 0.1 a n d D H A , 0.7 + 0.3 p , g / 1 0 ' platelets in PC: EPA, 0.2 + 0.2 a n d D H A , 1.3 + 0.8 / t g / 1 0 ~ platelets in PE. Linoleic acid did not c h a n g e significantly at a n y time. The fatty acid c o m p o s i t i o n of p h o s p h a t i d y l inositol or p h o s p h a t i d y l s e r i n e w a s not d e t e r m i n e d since EPA a n d D H A are not i n c o r p o r a t e d significantly into these platelet p h o s p h o l i p i d s in vivo [7]. The increase in EPA + D H A in platelet PC was 1.6 + 0.6 a n d 2.4 + 0.8 b t g / 1 0 ~ platelets a f t e r R e s - Q or E-EPA, respectively. T h e ratio of these fatty acids to A A increased f r o m 0.05 + 0.01 to 0.14 +_ 0.03 a f t e r ResQ a n d to 0.19 _+ 0.07 after E - E P A . In platelet PE, the increase in EPA + D H A was 1.1 + 0 . 7 a n d 1.4 + 0 . 9 p , g / 1 0 '~ platelets a f t e r R e s - Q a n d E - E P A , respectively. The ratio of these fatty acids to A A increased f r o m 0.07 + 0.02 to 0.16 + 0.02 after R e s - Q a n d to 0.17 + 0.03 after E-EPA.

Effect on other henlatologic parameters T h e only c h a n g e in b l o o d cell n u m b e r s d u r i n g the s t u d y was a 20% decrease in the whole b l o o d platelet c o u n t after b o t h R e s - Q a n d E - E P A (Table Iil). Platelet c o u n t s were still decreased 8 weeks after c e a s i n g E - E P A (2.4 + 0 . 5 . 105/ill). Several studies have n o t e d similar decreases in platelet c o u n t but in two l o n g - t e r m studies the c h a n g e was t e m p o r a r y (for review see Ref. 23). T h e r e was n o c h a n g e in platelet size d i s t r i b u t i o n at a n y time. Serum cholesterol decreased 13% a f t e r E - E P A , P < 0.05 a n d r e t u r n e d to control levels 8 weeks later (180 + 22 m g / d l ) . H D L cholesterol did not change: C o n t r o l ,

TABLE 111 I'ffect of dietao' vupph,mentation on hematologic parameters The whole blood platelet count, serum total cholesterol and triilcylglycerols were measured by automated methods. The bleeding time was done by the Simplate method. Mean+_S.D.. n = 9 . Difference from Control. * P < 0.05: from Oleate, * * P < 0.05, Control




Platelet count (.105/,ul) 2.7+_ 0.5 2.7_+ 0.6 2.2+_ 0.5* 2.2+- 0.3* Cholesterol (mg/dl) 193 +_33 187 +-35 196 +-28 168 +35* Triacylglycerols (mg/dl) 105 +_49 90 +_26 91 +_35 68 +_22" Bleeding time (rain) 4.0_+ 0.9 4.6_+ 1.7 6.2_+ 1.3"*

50 +_ 6; E - E P A , 49 + 7 m g / d l . S e r u m triacylglycerols d e c r e a s e d 33% a f t e r E - E P A , P < 0.05. T h e bleeding time, w h i c h w a s p e r f o r m e d b y technologists u n a w a r e of the s t u d y protocol, increased in five subjects a n d d e c r e a s e d in three subjects a f t e r R e s - Q c o m p a r e d to oleate, while E - E P A increased the b l e e d i n g time in six subjects a n d t w o s h o w e d n o c h a n g e . A 57% increase in the m e a n a f t e r E - E P A w a s highly significant. P < 0.01 ('l'able III).

Effects on platelet function e x vivo F o r e a c h subject the m i n i m u m c o n c e n t r a t i o n o f A D P a n d collagen sufficient to i n d u c e 50% of the m a x i m u m a g g r e g a t i o n r e s p o n s e ( t h r e s h o l d ) w a s d e t e r m i n e d in b o t h h e p a r i n i z e d w h o l e b l o o d a n d c i t r a t e d P R P . T h e res p o n s e to A D P d i d n o t c h a n g e significantly at a n y time p o i n t tested ( d a t a n o , shown). O n the o t h e r h a n d , E - E P A c a u s e d a n increase in the t h r e s h o l d c o n c e n t r a -


l-fleet of dietarl" supplementation on platelet flmction Collagen aggregation threshold (/tg/ml)" Control Whole blnnd PRP

0.52_+ 0.23 1.07+_ 0.50

Oleate 0.50_+ 0.23 0.98+_ 0.54

Res-Q 0.42+_ 0.17 1.12+_ 0.50

E- EPA 0.76+ 0.33* 1.37+_ 0.50*

Thrnmhoxane synthesis (pmol/lO s platelets)h ('oll~lgen 1 /~g/ml 2 t~g/ml

57 101

+- 1~', +32

55 89

_+19 +_38

44 83

_+13 +_29

20 37

zi: 12 * * +24 **

ATP rele se (nmo/10 ~ plutelet~)h Collagen 1 /~g/ml 2 t~g/ml

0.82 +_ 0.46 1.14+_ 0.49

0.84 +_ 0.38 1.14+_ 0.49

0.69 +_ 0.42 1.|8+_ 0.46

0.34 +_ 0.31 * * 0.90+_ 0.51

" The concentration of eonagen sufficient to induce a 10 ~ increase in electrical impedance (whole blood) or u 4()e; increase in light transmission (PRP) was determined, i, Synthesis of thromboxane (radioimmunoassav and ATP release (chemiluminescence) were measured in PRP at the completion of the aggregation repsonse. Mean _+S.D., n = 9. Difference from Control: * P < 0.05: * * P < 0.01.

37 tion for aggregation to collagen in both PRP and whole hlood. P < 0.05. while oleate and Res-Q had no effect (Table IV). The amount of thromhoxane syntheszed and the release of ATP were also measured in PRP in respome to collagen at concentrations close to the threshold for aggregation (1 and 2 pg/ml). E-EPA. hut not Res-Q. significantly decreased synthesis of thromhoxane in response to both concentrations of cc’lagen and inhibited ATP release stimc!ated by the lower concentration of collagen. all P < 0.01 (Table IV). The ability of platelets to increase cytoplasmtc calcium ion and to bind radiolabeled ftbrinogen m response to stimulation were studied as a measure of the effect of w-3 fatty acids on the signal-response coupling which precede platelet aggregation. ATP release and thromhoxane synthesis. For calcium mobilization. the platelet response to the thromboxane analog U46619 ,s independent of thromboxane synthesis and IS only partially enhanced by platelet aggregation. while the response to collagen is completely dependent on the extent of aggregation and. therefore. is greatly enhanced by thromboxane synthesis (data not shown). In netther case. however. was a decrease in calcium mobilization detected after dietary supplementation with w-3 fatty acids (Table V). Similarly, no significant change was detectable in fibrinogen binding in response to collagen or ADP after any of the supplements (Table V). Compurisott with effect of osptwr ON pkttelet /um mm After the completion of the entire protocol, three of the subjects took a single dose of apirm (7 mg/kg) and platelet function was studied 2 h later As shown in Table VI. aspirin had a much greater effect on platelet function as measured ex vivo than did the E-EPA

TABLE V EJJec I OJ &run ,~rppb~~rr~rorro~r I,,, c rrk !u,u nrohd,:urro~r ,e,cl /rhrr,wy hrndrng F o r c&,urn mc~h,hzeuon. the platuletr uere rcmn\ed fnmr pld\m.,. loaded n,th arquorm and re\u\pended m Hepevhu,f>I pl

Eicosapentaenoic acid ethyl ester as an antithrombotic agent: comparison to an extract of fish oil.

Evidence suggesting that dietary omega-3 polyunsaturated fatty acids decrease the risk of thrombosis comes mainly from studies involving supplementati...
434KB Sizes 0 Downloads 0 Views