Thrombosis Research 100 (2000) 305–315

ORIGINAL ARTICLE

Antithrombotic Potential of Olive OilAdministration in Rabbits with Elevated CholesterolJose Pedro De La Cruz1, Marıa Auxiliadora Villalobos2, Jose Antonio Carmona1,Mercedes Martın-Romero2, Jose Marıa Smith-Agreda2 and Felipe Sanchez de la Cuesta1

Departments of 1Pharmacology and Therapeutics and2Anatomy, School of Medicine, University of Malaga, 29072 Malaga, Spain.

(Received 21 March 2000 by Editor L.C. Petersen; revised/accepted 9 June 2000)

phological lesions of the endothelium and vascularAbstractwall. We conclude that supplementation of theSFAED with 15% olive oil reduced vascular throm-Olive oil is the main source of dietary fatty acidsbogenicity and platelet activation in rabbits. Al-in the Mediterranean region. The objective of thisthough the percentage of olive oil in the diet wasstudy was to evaluate the effect of dietary supple-

mentation with virgin olive oil in an experimental higher than the amount in the human diet, thesemodel with rabbits fed an atherogenic diet (satu- results may be helpful in determining the effect ofrated fat 48% of total fat). Four different groups olive oil in the human thrombogenic system. 2000of 10 animals each were studied: (1) normolipemic Elsevier Science Ltd. All rights reserved.diet (NLD), (2) atherogenic diet or saturated fattyacid-enriched diet (SFAED), (3) NLD with 15% Key Words: Olive oil; Platelets; Vessel wall; Thrombox-olive oil (NLD1OLIV), and (4) SFAED with 15% ane; Prostacyclinvirgin olive oil (SFAED1OLIV). The animals werefed the experimental diets for 6 weeks, after which

Platelets and the vessel wall are both involvedwe determined serum lipid profile (total cholesterol, in the genesis and evolution of atherosclero-HDL-cholesterol, and triglycerides), platelet aggre- sis, as well as in acute lipid overcharges,gation, platelet thromboxane B2, aortic prostacyclin, participating in the appearance of the initial athero-and platelet and vascular lipid peroxidation. Scan- matous lesion and in the subsequent developmentning electron microscopic images of the vascular en- of thrombosis, the main complication [1,2]. A num-dothelium were studied, as were morphometric pa- ber of alterations in platelet function, includingrameters in the arterial wall and thrombogenicity

hypersensitivity to aggregating agonists and in-of the subendothelium (annular perfusion cham-creased thromboxane production, have been re-ber). Animals fed the SFAED showed platelet hy-ported in animal models and in humans, both inperactivity and increased subendothelial thrombo-early and late periods of the disease [3–7]. In thegenicity. Animals fed the SFAED1OLIV showed,arterial wall, decreased endothelial production ofcompared with the SFAED group, an improved lipidprostacyclin and increased thrombogenicity haveprofile with decreased platelet hyperactivity and sub-been described [3–5,8].endothelial thrombogenicity and less severe mor-

Dietary supplementation with unsaturated fattyacids, especially marine oils, has been shown to

Corresponding author: J.P. De La Cruz, M.D., Department of reduce the alterations found in experimental mod-Pharmacology and Therapeutics, School of Medicine, University

els of hyperlipidemia and atherosclerosis, decreas-of Malaga, Campus de Teatinos s/n, 29071 Malaga, Spain. Tel: 134(952) 131 567; Fax: 134 (952) 131 568; E-mail: <jpcruz@uma.es>. ing platelet hypersensitivity and attenuating the

0049-3848/00 $–see front matter 2000 Elsevier Science Ltd. All rights reserved.PII S0049-3848(00)00321-2

306 J.P. De La Cruz et al./Thrombosis Research 100 (2000) 305–315

elevated synthesis of thromboxane [9–13], as well percholesterolemia after a short follow-up period(Table 1).as decreasing, to a certain extent, thrombogenicity

Each group was kept on the diet for 6 weeks.of the arterial wall [9,14]. These changes in plateletFood intake was recorded periodically to avoidfunction have been considered beneficial in pre-differences between groups in the amount of feedventing cardiovascular disorders.consumed. The OLIV-SFAED diet was preparedIn Mediterranean regions olive oil is one of thefresh daily by mean triturating of the solid foodmain sources of dietary fatty acids. Olives and theirand mixing with olive oil (150 g OLIV in 850 goil contain oleic acid and a series of polyphenolstriturated solid food). Animals from the NLD1[15], which have been shown to have an inhibitoryOLIV and SFAED1OLIV groups ate 11.862.5%effect on platelet function and thromboxane syn-less food (NLD: 18265.2 g/rabbit/day; NLD1thesis [16]. The objective of the present study wasOLIV: 16164.4 g/rabbit/day; SFAED: 19965.1 g/therefore to evaluate the effects of dietary supple-rabbit/day; SFAED1OLIV: 17266.0 g/rabbit/day).mentation with olive oil on platelet–vessel-wall in-

teractions in a model of hyperlipidemia with ath-1.3. Sample Collectionerosclerosis in rabbits fed a diet rich in unsaturated

fatty acids from olive oil.The animals were anaesthetised with a subcutane-ous injection of thalamonal (0.025 mg/kg of phenta-nyl and 1.2 mg/kg of dibenhydroperidol), ketamine1. Materials and Methods(50 mg/kg body weight), and midazolam (2.5 mg/kg body weight). A medial laparotomy was done1.1. Animalsto obtain blood from the inferior vena cava. Triso-dium citrate 3.8% at a proportion of 1:10 was usedA total of 40 male white New Zealand rabbits wereas anticoagulant; moreover, 2 mL of nonanticoagu-used. The animals were 2 months old at the startlated blood was kept at 378C for 45 minutes andof the experiment and had a mean body weight ofcentrifuged at 45003g for 15 minutes at 48C to2498636 g. The experimental protocol was ap-obtain serum.proved by the Ethical Commitee for Animal Care

The aorta was dissected from its root at the leftof the University of Malaga; at all times the animalsventricle to its iliac bifurcation and divided into fivereceived care in compliance with the criteria in the1.5-cm segments for morphofunctional analyses.Guide for the Care and Use of Laboratory AnimalsIn a descending order we used these segments asprepared by the National Academy of Sciences.follows: (1) measured the stable metabolite ofprostacyclin (6-keto-PGF1a); (2) measured the fatty

1.2. Experimental Groups acid composition of the arterial wall; (3) observedthe endothelium with scanning electron micros-

The rabbits were randomly divided into four copy; (4) studied subendothelial thrombogenicitygroups of 10 animals each. Group 1 animals were with a perfusion system in an annular chamberfed a normolipemic standard diet (NLD) (Lapin (Baumgartner chamber, Labotron SA, Barcelona,Entretien Ref/112, Laboratorios Panlab SL, Barce- Spain); and (5) observed arterial wall thickness andlona) containing 34% of saturated fatty acid with cell nucleus counts, which were analyzed morpho-respect to total fat. Group 2 animals received an metrically.atherogenic diet (diet supplemented with 1% cho-lesterol and saturated fatty acids 48% of saturated 1.4. Analytical Techniquesfat with respect to total fat [SFAED], LaboratoriosPanlab SL). Group 3 animals were given the NLD 1.4.1. Basic Serum Profilewith 15% virgin olive oil (NLD1OLIV) (Aceites To determine the basic serum lipid profile, we usedMinerva S.A., Malaga, Spain). Group 4 animals standard spectrophotometric techniques to mea-were fed with the atherogenic diet with 15% olive sure the concentration of total cholesterol, highoil (SFAED1OLIV). We used high cholesterol density lipoprotein (HDL) cholesterol, and triglyc-

erides.content in the atherogenic diet to obtain high hy-

307J.P. De La Cruz et al./Thrombosis Research 100 (2000) 305–315

1.4.2. Vascular 6-keto-PGF1a at 10,0003g for 5 minutes. The supernatant wasseparated and incubated at 1008C for 15 minutes,To measure 6-keto-PGF1a, samples of aortic tissue

were placed in 1 mL buffer (composition in mmol/ and absorbance was measured spectrophotometri-cally at 532 nm. Lipid peroxides (main componentL: 100 NaCl, 4 KCl, 25 NaHCO3, 2.1 Na2SO4, 20

sodium citrate, 50 Tris) and incubated with 100 mM malondialdehyde) were calculated by comparingthe results with a standard curve prepared witharachidonic acid for 10 minutes at 378C. After this

period two samples of 500 mL each were frozen malondialdehyde-bis-diethyl-acetal.at2808C until 6-keto-PGF1a was determined withradioimmunoassay (Amersham International plc, 1.4.6. Scanning Electron MicroscopyLittle Chalfont Buckinghamshire, England). For scanning electron microscopic examination of

the vascular endothelium, samples were fixed in1% glutaraldehyde and dehydrated in a series of1.4.3. Platelet Aggregometry

We determined platelet aggregation in whole alcohol-acetones (50, 70, and 80% ethyl alcoholfor 5 minutes each, 96 and 100% ethyl alcoholblood with the electrical impedance method [17] in

a Chrono-Log model 540S aggregometer (Chrono- for 30 minutes each, 100% acetone for 1 hour).Critical-point drying on dry ice was followed byLog Corp., Haverton, PA, USA). Aggregation was

induced with different concentrations of adenosine metallization with gold-palladium. The prepara-tions were observed and photographed in a Jeoldiphosphate (ADP) or collagen (Menarini Diag-

nostica, Barcelona, Spain). In each experiment the JSM-840 scanning electron microscope (Jeol In-struments, Tokyo, Japan).change in electrical impedance was recorded 10

minutes after the aggregate was added. Concentra-tion–effect curves were drawn from the percentage 1.4.7. Vascular Wall Morphometryaggregation obtained at each concentration of in- For morphometric studies of the vascular wall,ducer, and the concentration of inducer that pro- samples were fixed in 10% formaldehyde, embed-duced 50% of the maximal aggregation (EC50) ded in paraffin, and cut into 4-mm-thick sections.was calculated. One series of sections was stained with orcein to

examine wall thickness, and another series wasstained with hematoxylin-eosin for cell nucleus1.4.4. Serum Thromboxane B2

Thromboxane B2 (a stable metabolite of TxA2) counts. For morphometric analyses of the vesselwall we used the Visilog v. 3.6 program (Microptic,was measured by radioimmunoassay (Amersham

International plc, Little Chalfont, Buckingham- Barcelona, Spain) running on a Sun-Sparc Station(Sun Microsystems, Palo Alto, CA, USA). Imagesshire, England), with (3H)thromboxane B2. The

sample of whole blood without anticoagulant was were digitalized at 5123512 pixels with a 256-levelgray scale. Once the arterial wall, lumen, and ather-placed in a bath at 378C for 45 minutes and centri-

fuged at 45003g at 48C for 15 minutes. The serum omatous plaque had been identified as differentlevels of gray, the area of each portion and vesselwas removed and kept frozen at2808C until analy-

sis. To assess the possible influence of platelet num- diameter were measured in five sections from eachartery. Cell nuclei were counted in four differentber on platelet TxB2 production, we used the for-

mula described by Carter & Hanley [18]: TxB2 zones of five sections per artery (20 zones per rab-bit) at a magnification of 5003with a 100-mm2 grid.(nmol/109)5TxB2 (nmol/L) (12(hematocrit/100)/

platelet number (cells3109/L))3109.1.4.8. Platelet–Subendothelium InteractionFor perfusion in the Baumgartner annular chamber1.4.5. Lipid Peroxides

Platelet-rich plasma and aortic sections were used (20), the endothelium was removed with a-chymo-trypsin (0.4 mg/ml for 12 hours at 378C under con-to measure enzymatic lipid peroxidation [19] by

incubating samples at 378C with 200 mmol/L arachi- stant shaking), and the sample was placed in thechamber and perfused with samples of humandonic acid and adding 100 mmol/L indomethacin

5 minutes after the arachidonic acid. Then 500 mL blood obtained from healthy volunteers (men witha mean age of 38.562.2 years, none of whom hadof 0.5% thiobarbituric acid in 20% trichloroacetic

acid was added, and the sample was centrifuged taken any drugs during at least 15 days before the

308 J.P. De La Cruz et al./Thrombosis Research 100 (2000) 305–315

samples were collected). The chamber was coupled imental period to 25536139 g at the end of theexperimental period (21.360.5%) in the normoli-to a perfusion pump set at a flow rate correspond-

ing to a shear rate of 800 sg21 for 10 minutes. At pemic group with olive oil supplementation. In ani-mals fed the hyperlipidemic diet plus olive oil, bodythe end of the period of blood perfusion the arterial

segment was fixed in 10% formaldehyde, embed- weight changed from 2388642 g before the experi-mental period to 2763671 g at the end of the exper-ded in paraffin (modified from [20]), and sectioned

lengthwise. Each segment was cut into 120 trans- imental period (115.662.6%).In the serum lipid profile (Table 2), the SFAEDverse sections 2 mm in thickness, which were

stained with hematoxylin-eosin and used to quan- increased total cholesterol 9.2-fold and increasedtriglycerides 4.5-fold. The diet decreased HDL-tify parameters of platelet–subendothelial interac-

tion (20): Contact (C) was assessed as platelets cholesterol by 29.5% and led to a 13.3-fold increaseattached but not spread onto the subendothelium in the total/HDL-cholesterol index. This high tri-(nonactivated platelets). Spread (A), platelets glyceride values may be due to the high fat contentspread and firmly bound to the subendothelium, of the diet; moreover, the livers of these animalsmeasuring less than 5 mm high (representing the showed steatotic areas (data not shown). In choles-first phase of platelet aggregation). Thrombus (T) terol-elevated rabbits, dietary supplementationwere considered groups of platelets with clear signs with olive oil significantly reduced total cholesterolof activation, measuring 5 mm or higher. These (20%) and triglycerides (29.5%) in the serum lipidstructures were examined at 10 different points in profile. Serum HDL-cholesterol increased byeach section for a total of 1200 points analyzed in 67.4%, and the total/HDL-cholesterol index de-each perfused artery. creased by 50% (Table 2).

In animals fed the SFAED, the endothelial lininghad zones of endothelial denudation alternating1.4.9. Fatty Acidswith abnormal but not denuded zones. In areasFatty acids were quantified by gas chromatogra-where the subendothelium was exposed, erythro-phy. Samples were previously dried, methylated,cytes, leukocytes, and activated platelets were com-and extracted with hexane, then injected into amon. The nuclei were not distributed linearly, as inHewlett-Packard Model 5890 (Avondale, Pennsyl-normolipemic animals, and atheromatous plaquesvania, USA) gas chromatograph.were evident in zones of bifurcation (e.g., supraaor-tic trunks). Although the endothelial normal pat-1.4.10. Statistical Analysistern was not observed in all of the normolipemicAll results were expressed as the mean6SEM.and hyperlipidemic animals fed the olive-oil–When the variances were equal and the data weresupplemented diet (Figure 1, lower left), most ofnormally distributed, the groups were comparedthe endothelial lining was intact, and no denudedwith ANOVA analysis and subsequent transforma-zones or activated cellular elements were observedtion with the Bonferroni test; if not, the Kruskall-(Figure 1, lower right).Wallis test was used. All analyses were done with

Platelet aggregation in whole blood was in-the Statistical Package for Social Sciences (SPSS,creased in hyperlipidemic rabbits regardless ofversion 6.0, SPSS Inc., Chicago, IL, USA).whether ADP or collagen was used as the inducer(Figure 2). The EC50 values were significantly lowerin hyperlipidemic animals (Table 3). Platelet aggre-2. Resultsgation in animals fed the SFAED and olive oil wassignificantly reduced (Figure 2). The EC50 valuesBody weight increased from 2636636 g before thein these animals were 3.3-fold greater in ADP-experimental period to 30286155 g at the end ofinduced aggregation and 2.5-fold greater in colla-the experimental period (115.560.64%) in thegen-induced aggregation in rabbits fed the SFAEDnormolipemic group. In animals fed the SFAED,and olive oil than in rabbits that were not giventhe body weight changed from 2495631 g beforeolive oil (Table 3).the experimental period to 3281690 g at the end

Platelet production of thromboxane B2 (Table 3)of the experimental period (131.862.1%). Bodyweight decreased from 2570635 g before the exper- was nearly 30-fold greater in hypercholesterolemic

309J.P. De La Cruz et al./Thrombosis Research 100 (2000) 305–315

Table 1. Fatty acids (g/500 g of food) in different diets and in olive oila

NLD SFAED NLD1OLIV SFAED1OLIV OLIV

C14:0 0.19260.005 0.4360.037 0.2160.02 0.4560.02 0.0160.001C16:0 4.0360.18 13.3661.56 5.1460.36 14.7761.15 10.3360.93C16:1 0.08160.005 0.00460.001 0.4960.01 0.0660.002 0.6260.07C17:0 0.05160.006 0.3660.02 0.05960.002 0.1360.015 0.0760.008C:17:1 0.0360.0007 0.06960.004 0.08860.005 0.0860.006 0.1160.009C18:0 2.4560.05 5.2960.21 2.8660.09 5.6260.41 3.6560.19C18:1 2.7560.09 6.5660.50 12.9461.93 15.0561.01 74.8866.15C18:2 9.9461.12 9.5560.55 10.8560.99 10.4261.09 8.7060.72C18:3 0.1660.08 5.5060.18 0.2460.04 5.5960.42 0.6560.06C20:0 0.1760.009 0.3960.009 0.2160.02 0.4160.03 0.5460.06C20:1 0.30460.02 0.2660.04 0.6460.02 0.6160.04 0.3860.03Total Fat 20.263.2 42.165.1 35.363.8 53.366.1 97.8669.15MUFA 3.260.4 6.960.7 15.361.9 17.161.8 74.3868.2*SFA 6.960.8 20.162.5 9.061.1 20.262.5 14.3061.5*PUFA 10.161.1 15.161.9 11.261.3 16.562.1 9.1860.9*

a Each value is the mean6SEM of six samples (one each week); NLD5normolipemic diet, SFAED5saturated fatty acid-enriched diet,OLIV5olive oil; SFAs5saturated fatty acids; PUFAs5polyunsaturated fatty acids; MUFAs5monounsaturated fatty acids.

* percent of total fat

animals than in controls. Mean 6-keto-PGF1a pro- SFAED after treatment with olive oil. Aortic lipidperoxidation was reduced by 14% in normolipemicduction was 2.1-fold lower in animals fed the

SFAED than in normolipemic controls (Table 3). rabbits and by 75% in hypercholesterolemic rab-bits after treatment with olive oil (Table 3).In animals with the olive-oil–supplemented diet,

platelet production of thromboxane B2 was signifi- Morphometric analyses of the wall of the aorta inhypercholesterolemic animals showed thickeningcantly reduced (67.8% in hypercholesterolemic

rabbits), although it was not as low as in normo- and increased nuclear counts in comparison withnormolipemic rabbits. In the former group 9.236lipemic animals. In the groups given the normal

diet, 6-keto-PGF1a production increased 78% after 0.56% of the vascular wall was occupied by foamcells (Figures 3 and 4). Dietary supplementationtreatment with olive oil. In animals fed the SFAED,

6-keto-PGF1a production increased 325% with with olive oil significantly modified the compo-nents of the arterial wall in hypercholesterolemicolive oil treatment (Table 3).

Platelet malondialdehyde equivalents were 9.3- rabbits, decreasing wall area, foam cell nuclearcount, and the percentage of the wall occupied byfold greater in hypercholesterolemic animals than

in controls, and aortic malondialdehyde equiva- foam cells in comparison with animals that did notreceive dietary olive oil (Figures 3 and 4).lents were 2.04-fold greater (Table 3). Platelet lipid

peroxide production was inhibited by 25% in nor- Functional tests of arterial wall-platelet interac-tions showed that in animals fed the SFAEDmolipemic rabbits and by 57% in rabbits fed the

Table 2. Basic serum lipid profile in rabbits

NLD NLD1OLIV SFAED SFAED1OLIV

Total cholesterol (mmol/L) 2.1160.32 2.3660.41 19.5563.05* 15.6462.36**Triglycerides (mmol/L) 3.0260.45 3.3960.45 13.5562.03* 9.5561.01**HDL-cholesterol (mmol/L) 0.6160.09 0.8760.09 0.4360.05 0.7260.09**Total cholesterol/

HDL-cholesterol 3.2660.26 2.8660.40 43.4668.80* 21.7264.12**

NLD5normolipemic diet, SFAED5saturated fatty acid-enriched diet, OLIV5olive oil.* p50.05 (vs NLD); ** p50.05 (vs SFAED)

310 J.P. De La Cruz et al./Thrombosis Research 100 (2000) 305–315

Fig. 1. Examples of scanning electron micrographs of aortic endothelium from a normolipemic rabbit (upper left), ahyperlipidemic rabbit (upper right) and hyperlipidemic rabbits with 15% (wt/wt) dietary supplementation with olive oil(lower left and lower right).

thrombogenicity was increased, as reflected by the 3. Discussionsignificant increase in surface screened in the arte-

Our results show that dietary supplementation withrial segment mounted in the Baumgartner chamber15% olive oil, in our experimental model of feeding(32.9% increase in spread, 11.1% increase inrabbits with an atherogenic diet, reduced thrombo-thrombus) (Table 4).genic factors and increased antithrombotic factors.The percentage of covered subendothelium wasThese effects were seen both in platelets and in thenot significantly greater in normolipemic animalsvessel walls. In platelets, dietary olive oil reducedtreated with olive oil. In hypercholesterolemic ani-hypersensitivity to aggregation, thromboxane syn-mals olive oil significantly decreased the percent-thesis, and lipid peroxide production. In the arterial

age of subendothelium screened by 37%, mainly wall the supplemented diet increased prostacyclinas a result of the decrease in thrombus (59%) synthesis and reduced thrombogenicity of the sub-(Table 4). endothelium. We also observed a reduction in hy-

Table 5 shows the results of chromatographic perlipidemic status, with increased levels of HDL-analyses of the arterial wall to determine fatty acid cholesterol. All effects were seen especially clearlycomposition after feeding for 6 weeks with each in animals fed the atherogenic diet and appeared

less clearly in normolipemic rabbits.of the diets.

311J.P. De La Cruz et al./Thrombosis Research 100 (2000) 305–315

genic diet led to large increases in cholesterol lev-els, which are difficult to reduce to normal withdietary measures alone. However, the comparisonof the lipid metabolism between humans and rab-bits is not so exact.

The platelet hyperactivity found in rabbits thatconsumed the atherogenic diet was reduced in ani-mals that consumed olive oil. Some phenolic com-ponents of olive oil are able to inhibit platelet ag-gregation in platelet-SFAED plasma in vitro [16],and one study found that sensitivity to collagen-stimulated aggregation was reduced in patientswith type II hypercholesterolemia treated with di-etary olive oil [23]. Neither normal diet norSFAED contain polyphenolic substances ora-tocopherol; olive oil contains 1024 mg/kg of poly-phenolic components and 150 mg/kg of a-tocoph-erol (data obtained from manufacturers). Our re-sults show that in rabbit whole blood, platelets

Fig. 2. Concentration-dependent curves of platelet aggre- are less sensitive to aggregating agents and showgation induced by ADP or collagen after 6 weeks on a aggregating behavior similar to that in control ani-normolipemic (open symbols) or hyperlipidemic (solid

mals. This inhibitory action may involve inhibitionsymbols) diet, without (circles) or with (squares) 15% sup-of thromboxane synthesis, which is fundamentalplementation with olive oil.for platelet aggregation, or reduced lipid peroxideproduction. Some antioxidants such as flavonoidsare able to inhibit platelet functioning, althoughThe conclusions of earlier studies that investi-this effect does not always correlate with an inhibi-gated reductions in hyperlipidemia in humans havetion of thromboxane synthesis [24]. These antioxi-not been unanimous: One study [21] failed to finddant substances are among many possible compo-significant changes in the lipid profile, whereas an-nents that could play a role in the beneficial effectother [22] reported a decrease in total cholesterolof olive oil.and an increase in the HDL-cholesterol fraction.

An important consideration in analyzing our re-In the present study of hyperlipidemic rabbits, wesults is that dietary supplementation with olive oilfound significant changes in cholesterol, triglycer-affected both the platelet and the vascular compo-ides, and HDL-cholesterol, although the reduc-nent of atherosclerosis, reducing subendothelialtions did not return values to those found in controlthrombogenicity and stimulating the endothelialanimals that were fed a normal diet. This finding

may have resulted from the fact that the athero- synthesis of prostacyclin. This latter effect does not

Table 3. Aggregometric and prostanoid parameters in the four groups of animals

NLDa NLD1OLIV SFAED SFAED1OLIV

EC50 ADP (mmol/L 2.5060.19 2.0460.54 0.8460.14* 2.8360.35**EC50 Collagen (mg/mL) 0.5560.07 0.5160.05 0.2260.04* 0.5560.09**Thromboxane B2 (ng/mL) 115615.91 127618.42 3095689.57* 9976183**6-keto-PGF1a (ng/mg aorta) 3.4960.64 6.2560.80* 1.6860.06* 7.1560.51**Platelet MDA (nmol/109 cells) 0.0860.01 0.0660.009 0.7560.06* 0.3260.04**Aortic MDA (nmol/mg protein) 2.4360.02 2.0960.16 4.9660.25* 1.2460.09**

a NLD5normolipemic diet, SFAED5saturated fatty acid-enriched diet, OLIV5olive oil; MDA5malondialdehyde equivalents;EC505concentration of ADP or collagen that produced 50% of maximal aggregation.

* p50.05 (vs. NLD); ** p50.05 (vs. SFAED).

312 J.P. De La Cruz et al./Thrombosis Research 100 (2000) 305–315

Fig. 3. Examples of optic micrographs of aortic wall from anormolipemic rabbit (upper) and a hyperlipidemic (lower).All the slices were obtained in the same aortic area.

Fig. 4. Morphometric parameters of aortic sections in nor-molipemic (NLD) or atherogenic (SFAED) rabbits (10per group) without or with (1OLIV) 15% (wt/wt) dietarysupplementation with olive oil (OLIV). *p,0.05 with re-appear to have been the result of a direct stimula-spect to NLD; **p,0.05 with respect to the SFAED. Thesetion of prostacyclin synthase; in animals that con-values were obtained in alleatory aortic slices (always in

sumed the normolipemic diet we found no effect the same aortic section); in each one, four measures wereon the synthesis of this prostanoid. Some compo- made, with an interval of 908.nents of olive oil have antioxidant effects [25,26],as corroborated by our finding that lipid peroxida-tion in the vascular wall was inhibited (Table 3). dothelial cells allowed them to synthesize moreThus, the decrease in peroxides implies that prosta- prostaglandin.cyclin synthesis was not inhibited, because lipid The reduced thrombogenicity of the subendo-peroxides reduce the synthesis of this prostaglan- thelium (Table 4) reflected the effect of olive oildin [27]. This indirect effect has been found with not only on the rabbit platelet function, but also onsome drugs that show antioxidant properties, such the arterial wall, because perfused blood samplesas dipyridamole [28] and S-adenosyl-L-methionine were taken from healthy volunteers who did not[29]. However, we cannot rule out that the im- consume dietary olive oil. Our results are quantita-

tively different from those published by Baumgart-provement in the morphological lesions in the en-

313J.P. De La Cruz et al./Thrombosis Research 100 (2000) 305–315

Table 4. Percentage of vascular subendothelium covered by platelets after 10 minutes of human bloodperfusion in the Baumgartner annular chamber (shear rate: 800 sg21)

Percentages with respect to the surface screened

NLDa NLD1OLIV SFAED SFAED1OLIV

Contacts 53.1564.68 47.0964.00 14.6662.57* 30.0762.05**Spread 22.7162.03 29.9163.24 55.6465.48* 49.8266.77Thrombus 23.4463.12 21.0263.91 34.5864.56* 14.1260.85**Covered surface 25.1663.12 23.8763.30 38.7065.13* 24.3663.84**Height of thrombus (mm) 5.6160.19 5.4560.18 7.1360.54* 5.3160.20**

a NLD5normolipemic diet, SFAED5saturated fatty acid-enriched diet, OLIV5olive oil.* p50.05(vs. NLD); ** p50.05(vs. SFAED).

ner et al. [20], but the distribution of contacts, urated fatty acids in the arterial wall to decrease,although this reduction is not statistically signifi-spread, and thrombus are proportionally similar to

theirs. The differences may be due to the histologi- cant. Nonetheless, it is important to note that thereis a proportional decrease in saturated fatty acids. Itcal method (electron microscopic process in the

Baumgartner description and optical microscopic has been shown [30] that dietary supplementationwith linoleic acid displaces other fatty acids fromprocess in our study). However, the interpretation

of the results is the same in all groups, including the vessel wall. In our experimental model, dietaryolive oil increased the amount of oleic acid in thecontrol groups.

Morphological findings included a thinner vessel arterial wall and also appeared to displace othersaturated fatty acids while leaving polyunsaturatedwall with fewer foam cells (Figures 3 and 4), i.e.,

changes that perhaps have been involved in the fatty acids unaffected. This change in the composi-tion of the vessel wall is notable for three reasons:lower thrombogenicity in this group. Moreover,

the fatty acid content in the vessel wall changed (1) Saturated fatty acids are atherogenic and favorplatelet aggregation, decreasing prostacyclin pro-in animals that were fed with the olive-oil–

supplemented diet (Table 5), as shown by the in- duction and increasing thromboxane production.They can thus be considered prothrombotic sub-crease in C18:1 fatty acids. This finding, also found

in normolipemic rabbits, was logical in view of the stances. (2) Polyunsaturated fatty acids reduceplatelet activity and the thrombogenic capacity offact that this fatty acid is a major component of

olive oil. In general, when olive oil is consumed the arterial wall [9,11,14]; it is therefore noteworthythat dietary olive oil did not significantly decreasethere is a tendency for the proportion of polyunsat-

Table 5. Fatty acid composition (percentages) of aortic vascular tissue from the differentgroups of rabbits

Fatty acid NLDa NLD1OLIV SFAED SFAED1OLIV

C 14:0 4.160.4 1.560.3* 3.160.1 1.860.3**C 16:0 41.362.2 25.561.0* 36.561.5 24.761.3**C 18:0 8.660.6 5.860.4* 9.660.8 6.660.6**C 18:1 28.360.5 51.662.2* 33.062.9 53.061.5**C 18:2 18.061.9 15.362.6 15.961.6 13.661.7C 18:3 ND ND 0.0360.01 NDC 20:0 2.160.2 0.860.3* 1.360.4 1.260.1MUFAs 26.960.8 51.662.2* 33.662.1 53.061.5**SFAs 53.862.4 33.861.4* 50.562.1 34.561.7**PUFAs 18.262.0 15.362.6 16.262.6 13.661.7

a NLD5normolipemic diet, SFAED5saturated fatty acid-enriched diet, OLIV5olive oil; ND5not detected; SFAs5

saturated fatty acids; PUFAs5polyunsaturated fatty acids; MUFAs5monounsaturated fatty acids.* p,0.05 vs. NLD, ** p,0.05 vs. SFAED.

314 J.P. De La Cruz et al./Thrombosis Research 100 (2000) 305–315

the proportion of these fatty acids in the vessel 5. Fitzgerals GA, Catella F, Oates JA. Eicosanoidbiosynthesis in human cardiovascular disease.wall. (3) The accumulation of monounsaturated

fatty acids provides the arterial wall with an antioxi- Human Pathol 1987;18:248–52.6. Kawaguch H, Ishibashi T, Imai Y. Thrombox-dating agent [25,26]. Moreover, the polyphenols in

olive oil also contribute to the effects described ane B2 biosynthesis and phospholipids hydroly-sis in platelets from hypercholesterolemic rab-above [16].

In summary, dietary supplementation with olive bits. Lipids 1981;16:37–42.7. Packham MA, Mustard JF. The role of plate-oil in rabbits previously fed an atherogenic diet

reduced prothrombotic activity in both of the com- lets in the development and complications ofatherosclerosis. Semin Hematol 1986;23:8–26.ponents involved in thrombus formation. In plate-

lets we found decreases in aggregation, thrombox- 8. Zmuda A, Dembinska-Kiec A, Chitkowski A,ane synthesis, and lipid peroxide formation; in the Gryglewski RJ. Experimental atherosclerosisarterial wall, we found reduced thrombogenicity of in rabbits: Platelet aggregation, thromboxanethe subendothelium and increased antithrombotic A2 generation and antiaggregatory potency ofactivity in the endothelium (i.e., increased prosta- prostacyclin. Prostaglandins 1977;14:1035–41.cyclin synthesis and reduced lipid peroxide forma- 9. Bertomeu MC, Crozier GL, Haas TA, Fleithtion). These effects may be the result of a protective M, Buchanan MR. Selective effects of dietaryeffect against the morphological lesions that the fats on vascular 13-HODE synthesis and plate-atherogenic diet caused in the vessel wall, a modifi- let/vessel wall interactions. Thromb Res 1990;cation in the lipid composition of the arterial wall, 59:819–30.or a combination of both. 10. Davis HR, Bridenstine RT, Vesselennovitch

D, Wissler RW. Fish oil inhibits developmentWe thank M.V. Anaya, M.D. Villatoro, A. Pino, and J. Diaz for of atherosclerosis in rhesus monkeys. Arterio-their excellent technical assistance and K. Shashok for translating

sclerosis 1986;7:441–9.parts of the original manuscript into English.11. De La Cruz JP, Martın-Romero M, Carmona

JA, Villalobos MA, Sanchez de la Cuesta F.Effect of evening primrose oil on platelet ag-Referencesgregation in rabbits fed an atherogenic diet.Thromb Res 1997;87:141–9.1. Kristensen SD, Schmidt EB, Dyerberg J. Di-

12. Knapp HR, Reily IA, Alessandrini P, Fitzger-etary supplementation with n-3 polyunsatu-ald GA. In vivo indexes of platelet and vascularrated fatty acids and human platelet function:function during fish-oil administration in pa-A review with particular emphasis on implica-tients with atherosclerosis. N Engl J Med 1986;tions for cardiovascular disease. J Intern Med314:937–42.1989;225(1):S141–50.

13. Von Schacky C, Weber PC. Metabolism and2. Siess W, Roth P, Scherer B, Kurzmann I, Boh-effects on platelet function of the purified eico-lig B, Weber PC. Platelet-membrane fattysapentaenoic and docosahexaenoic acids in hu-acids, platelet aggregation, and thromboxanemans. J Clin Invest 1989;76:2446–50.formation during a mackerel diet. Lancet 1980;

14. Villalobos MA, De La Cruz JP, Martin-Romero1:441–4.M, Carmona JA, Smith-Agreda JM, Sanchez De3. Arnold JH, Pritchard KA, Greco NJ, Pangana-La Cuesta F. Effect of dietary supplementationmala RV. Production of platelet thromboxanewith evening primrose oil on vascular thrombo-A2 and arterial prostacyclin I2 from hypercho-genesis in hyperlipemic rabbits. Thromb Hae-lesterolemic rats. Prostaglandins Leukotr Es-mostas 1998;80:696–701.sent Fatty Acids 1992;46:215–7.

15. Montedoro GF, Servili M, Baldioli M, Miniati4. Gryglewski RJ, Dembiska-Kiec A, Zmuda A,E. Simple and hydrolyzable phenolic com-Gryglewska T. Prostacyclin and thromboxanepounds in virgin olive oil. Their extraction, sep-A2 biosynthesis or capacities of heart, arteries,aration, and quantitative and semiquantitativeand platelets at various stages of experimen-evaluation by HPLC. J Agricult Food Chemtal atherosclerosis in rabbits. Atherosclerosis

1978;31:385–94. 1992;40:1571–6.

315J.P. De La Cruz et al./Thrombosis Research 100 (2000) 305–315

16. Petroni A, Blasevich M, Salami M, Papini N, 24. Laughton MJ, Evans PJ, Moroney MA, HoultMontedoro GF, Galli C. Inhibition of platelet JRS, Halliwell B. Inhibition of mammalianaggregation and eicosanoid production by phe- 5-lipoxygenase and cyclo-oxygenase by flavo-nolic components of olive oil. Thromb Res noids and phenolic dietary additives: Relation-1995;78:151–60. ship to antioxidant activity and to iron ion-

17. Cardinal DC, Flower RJ. The electronic ag- reducing ability. Biochem Pharmacol 1991;42:gregometer: A novel device for assessing plate- 1673–81.let behaviour in blood. J Pharm Meth 1980; 25. Reaven P, Parthasarathy S, Grasse B, Miller6:135–58. E, Steinberg D, Witztum JL. Effects of oleate-

18. Carter AJ, Hanley SP. The effect of platelet rich and linoleate-rich diets on the susceptibil-number and haematocrit on whole blood throm- ity of low density lipoprotein to oxidativeboxane synthesis. Thromb Haemostas 1985;54: modification in mildly hypercholesterolemic225–7. subjects. J Clin Invest 1993;91:668–76.

19. Smith J, Ingerman-Wojenski C, Silver M. Ma- 26. Scaccini C, Nardini M, D’Aquino M, Gentilllondialdehyde formation as an indicator of V, DiFelice M, Tomassi G. Effect of dietaryprostaglandin production by human platelets. oils on lipid peroxidation and on antioxidantJ Lab Clin Meth 1976;881:167–72. parameters of rat plasma and lipoprotein frac-

20. Baumgartner HR. Platelet interaction with tions. J Lipid Res 1992;33:627–33.vascular structures. Thromb Diath Haemorrh 27. Warso M, Lands W. Lipid peroxidation in rela-Suppl 1972;51:161–71. tion to prostacyclin and thromboxane physiol-21. Schick PK, Wojenski CM, Walker J. The ef-

ogy and pathophysiology. Br Med Bull 1983;fects of olive oil, hydrogenated palm oil, and39:277–80.n-3 fatty acid-enriched diets on megakaryo-

28. De La Cruz JP, Ortega G, Sanchez De Lacytes and platelets. Arteriosclerosis and Throm-Cuesta F. Differential effects of the pyrimido-bosis 1993;13:84–9.pyrimidine derivatives, dipyridamole and mop-22. Grundy SM, Florentin N, Nix D, Whelan MF.idamol, on platelet and vascular cyclooxygen-Comparison of monounsaturated faty acidsase activity. Biochem Pharmacol 1994;47:and carbohydrates for reducing raised levels209–15.of plasma cholesterol in man. Am J Clin Nutr

29. De La Cruz JP, Gonzalez-Correa JA, Martin-1988;47:965–9.Aurioles E, Ortiz P, Sanchez De La Cuesta23. Sirtori C, Gati E, Tremoli E, Galli C, Gian-F.Effects of S-adenosyl-L-methionine on plate-franceschi G, Franceschini G, Colli S, Madernalet thromboxane and vascular prostacyclin.P, Marangoni F, Perego P, Stragliotto E. OliveBiochem Pharmacol 1997;53:1761–3.oil, corn oil, and n-3 fatty acids differently af-

30. Horrobin DF. Omega-6 and omega-3 essentialfect lipids, lipoproteins, platelets, and superox-fatty acids in atherosclerosis. Sem Thrombide formation in type II hypercholesterolemia.

Am J Clin Nutr 1992;56:113–22. Haemost 1993;19:129–37.