A

Oh(M[aaaRddCc©

K

1

[gaar

0d

Atherosclerosis 194 (2007) e141–e146

The chronic intake of a Mediterranean diet enriched in virgin olive oil,decreases nuclear transcription factor �B activation in peripheral

blood mononuclear cells from healthy men

Pablo Perez-Martinez a,∗, Jose Lopez-Miranda a, Luis Blanco-Colio b, Cecilia Bellido a,Yolanda Jimenez a, Juan A. Moreno a, Javier Delgado-Lista a,

Jesus Egido b, Francisco Perez-Jimenez a

a Lipids and Atherosclerosis Research Unit, Hospital Universitario Reina Sofıa, Avenida Menendez Pidal s/n, 14004-Cordoba, Spainb Vascular Research Laboratory, Fundacion Jimenez Dıaz, Universidad Autonoma de Madrid, Madrid, Spain

Received 14 October 2006; received in revised form 24 November 2006; accepted 27 November 2006Available online 3 January 2007

bstract

bjetive: Nuclear transcription factor �B (NF-�B) plays a key role in the inflammatory response and can be modulate by dietary fat. Weave examined the effect of three diets, with different fat composition, on the activation of NF-�B on peripheral blood mononuclear cellsPBMCs).

ethods: Sixteen healthy men followed three 4-week diets, in a randomised crossover design: a Western diet, rich in saturated fat (SFA)22% SFA, 12% monounsaturated fat (MUFA) and 0, 4 �-linolenic acid]; a Mediterranean diet [<10% SFA, 24% MUFA and 0.4% �-linoleniccid], and a low fat diet enriched in �-linolenic acid [<10% SFA, 12% MUFA and 2% �-linolenic acid]. NF-�B (electrophoretic mobility shiftssay) in mononuclear cells and plasma concentrations (ELISA) of soluble vascular cellular adhesion molecule 1 (VCAM-1) were examinedfter either diets.esults: Western diet increased 2.7-fold NF-�B compared with the Mediterranean diet (p = 0.038) and 1.79-fold with the �-linolenic acidiet (p = 0.07). No differences were found between the last two. Furthermore, an increase on plasma VCAM-1 was observed with the Western

iet (p < 0.05).onclusions: The Mediterranean diet diminished NF-�B activation in mononuclear cells, compared with Western diet, supporting itsardioprotective properties. The effect of the n − 3 enriched diet was intermediate.

2006 Elsevier Ireland Ltd. All rights reserved.

tor NF-

bfbim

eywords: Diet; Blood cells; Mediterranean diet; Nuclear transcription fac

. Introduction

Diet is a cornerstone of cardiovascular disease prevention1]. It is well known that the Mediterranean diet, rich in vir-in olive oil as a major fat source, fruits, vegetables, legumes

nd other plant foods and low in saturated and trans fattycids and cholesterol, is associated with a low cardiovascularisk. The cardiovascular protection from this diet is proba-

∗ Corresponding author. Tel.: +34 957 218250; fax: +34 957 218250.E-mail address: pablopermar@yahoo.es (P. Perez-Martinez).

ccmaat

021-9150/$ – see front matter © 2006 Elsevier Ireland Ltd. All rights reserved.oi:10.1016/j.atherosclerosis.2006.11.033

�B; Olive oil; VCAM-1

ly not brought up only by their effects of plasma lipid riskactors but by effects on several other pathways, includinglood pressure, insulin sensitivity, arterial wall function andnflammation [2]. In addition, olive oil, besides high levels of

onounsaturated fatty acid (MUFA), contains several minoromponents with biological properties [3]. Atherosclerosis isonsider an inflammatory disease, characterized by the accu-

ulation of macrophage-derived foam cells in the vessel wall

nd accompanied by the production of cytokines, chemokinesnd growth factors [4]. Substantial evidence suggests thathe Mediterranean diet can modulate biomarkers involved

e eroscle

iamtasonaciicaiaCpbowNettdof

2

2

swaspotbag

2

iceSM4(

nSPtoTii

utsaeaFrf−melcin

2

1ttTsPmw[iw

2(

stwit

2

142 P. Perez-Martinez et al. / Ath

n biological pathways implicated in the development oftherosclerosis. As a major transcription factor in inflam-atory responses, nuclear factor �B (NF-�B) is involved in

he regulation of inflammatory and immune genes, apoptosisnd cell proliferation [5]. In this regard, NF-�B is a redox-ensitive transcription factor implicated in the transmissionf different signals from the cytoplasm to the nucleus ofumerous cell types. This transcription factor is found inhibits a trimer consisting of p50, p65 and I�B subunits in theytosol. Several stimuli are able to induce the release of thenhibitory I�B subunit and subsequently NF-�B activation,ncluding tumor necrosis factor � (TNF-�), hyperglycemia,ytokines and oxidized-LDL. Many of these NF-�B inducersnd NF-�B regulated genes have been implicated directly orndirectly in atherosclerosis [6]. Other factors inhibit NF-�Bctivation, including glucocorticoids, drugs such as HMG-oA reductase inhibitors and antioxidants [7]. In addition, arevious study showed that the consumption of a fat-enrichedreakfast increased NF-�B activation in humans [8]. More-ver, we have observed that the acute intake of butter oralnut, but not olive oil elicited the postprandial activation ofF-�B, on mononuclear cells from healthy men [9]. How-

ver, no data exist about the chronic effect of the diet onhe NF-�B activation and its potential effects on the promo-ion of the inflammation process. Therefore, our aim was toetermine the effect of three dietary models on the activationf NF-�B on peripheral blood mononuclear cells (PBMCs)rom healthy male volunteers.

. Materials and methods

.1. Study subjects

Sixteen male medical students were included in thistudy. All of them gave their informed consent and under-ent a comprehensive medical history, physical examination

nd clinical chemistry analysis before enrolment. Subjectshowed no signs of any chronic disease or obesity, and noneracticed unusually high levels of physical activity. Nonef the subjects was taking medication or vitamins knowno affect plasma lipids. The study protocol was approvedy our Hospital Human Investigation Review Committee,ccording to the institutional and the Good Clinical Practiceuidelines.

.2. Study design

Each volunteer in the trial was submitted to three dietntervention periods of 4 weeks of duration, in a randomizedrossover design. The composition of three diets was: West-rn diet rich in SFAs: 15% protein, 47% CHO, 38% fat [22%

FAs, 12% MUFAs, 4% PUFAs (0.4% �-linolenic acid)];editerranean diet enriched in virgin olive oil: 15% protein,

7% CHO, 38% fat [<10% SFAs, 24% MUFAs, 4% PUFAs0.4% �-linolenic acid)] and a high-CHO enriched in vegetal

a

H

rosis 194 (2007) e141–e146

− 3 fatty acids: 15% protein, 55% CHO, <30% fat [<10%FAs, 12% MUFAs, 8% PUFAs (2% �-linolenic acid)]. TheUFA enrichment of the high-CHO diet was achieved via

he use of natural food components rich in �-linolenic acidf vegetable origin (based on walnuts (Juglans regia L.)).he cholesterol content of the diets was <300 mg/day, and

t was kept at a constant level throughout the three dietaryntervention periods.

The composition of the experimental diets was calculatedsing the United States Department of Agriculture [10] foodables and Spanish food composition tables for local food-tuffs [11]. All meals were prepared in the hospital kitchennd were supervised by a dietitian. Lunch and dinner wereaten in the hospital dining room, whereas breakfast and anfternoon snack were eaten in the medical school cafeteria.ourteen menus were prepared with regular solid foods andotated during the experimental period. Duplicate samplesrom each menu were collected, homogenized and stored at70 ◦C. The study took place during January and March toinimize seasonal effects and academic stress. Subjects were

ncouraged to maintain their regular physical activity andifestyle and were asked to record in a diary any event thatould affect the outcome of the study, such as stress, changen smoking habits and alcohol consumption or intake of foodsot included in the experiment design.

.3. Lipid analysis and biochemical determinations

Venous blood samples were collected in tubes containingmg/mL EDTA in fasting at 4 weeks after the ingestion of

he three diets. Plasma was obtained by low-speed centrifuga-ion (1500 × g) for 15 min at 4 ◦C within 1 h of venipuncture.o reduce interassay variation, all plasma samples weretored at −80 ◦C and were analyzed at the end of the study.lasma cholesterol and triacylglycerol concentrations wereeasured by enzymatic techniques [12,13]. HDL-cholesterolas measured after precipitation with phosphotungstic acid

14]. Apolipoproteins (Apo) A-I and B were determined bymmunoturbidimetry [15]. LDL-cholesterol concentrationsere calculated by using the Friedewald formula [16].

.4. Isolation of peripheral blood mononuclear cellsPBMCs)

The blood samples were diluted 1:1 in PBS, and cells wereeparated in 5 mL Ficoll gradient (lymphocyte isolation solu-ion, Rafer) by centrifugation at 2000 × g for 30 min. PBMCsere collected, washed twice with cold PBS and resuspended

n buffer A (see Protein Extraction). Approximately 95% ofhe cells were mononuclear cells (flow cytometry, not shown).

.5. Protein extraction and electrophoretic mobility shift

ssays

Proteins of PBMCs were extracted as described byernandez-Presa et al. [17].

P. Perez-Martinez et al. / Atherosclerosis 194 (2007) e141–e146 e143

Table 1Plasma lipids and apoliproteins (Apo) at the end of each dietary period

Western diet Mediterranean diet CHO and n − 3 diet

Total cholesterol (mmol/L) 3.83 ± 0.11 3.60 ± 0.12* 3.55 ± 0.14*

HDL-cholesterol (mmol/L) 1.16 ± 0.20 1.17 ± 0.18 1.11 ± 0.20LDL-cholesterol (mmol/L) 2.24 ± 0.11 2.04 ± 0.10* 2.03 ± 0.12*

Triacylglycerols (mmol/L) 0.91 ± 0.10 0.86 ± 0.08 0.87 ± 0.08Apo B (g/L) 0.70 ± 0.03 0.64 ± 0.02* 0.63 ± 0.02*

A

aBGbi(5fia

2

1m(iit

2

coaasguet

3

3

niiLM

dnftH

3

dcaactNt(

3

td

ewdpVrwTamtcp

bc

po A-I (g/L) 1.25 ± 0.03

* Significantly different from typical Western diet.

Gel shift assays were performed with a commercial kitccording to the instructions of the manufacturer (Promega).riefly, NF-�B consensus oligonucleotide (5′-AGTT-AGGGGACTTTCCCAGGC-3′) was end-labeled with 32Py incubation with 10 U T4 polynucleotide kinase (Promega)n a reaction containing 10 �Ci [�-32P]ATP (3000 Ci/mmol)Amersham), 70 mmol/L Tris–HCl, 10 mmol/L MgCl2 andmmol/L DTT. The gel was dried and exposed to an X-raylm. The autoradiograph was subjected to densitometry usingImage Quant densitometric scanner (Molecular Dynamics).

.6. Adhesion molecules immunoassay

Plasma concentrations of vascular cell adhesion molecule-(VCAM-1), intercellular adhesion molecule-1 (ICAM-1),onocyte chemoattractant protein-1 (MCP-1), interleukin 6

IL-6) and tumor necrosis factor � (TNF-�) were determinedn duplicate with commercially available enzyme-linkedmmunosorbent assay kits (R & D Systems, Inc.) accordingo the manufactures guidelines.

.7. Statistical analysis

Statistical analyses were carried out using SPSS statisti-al software, Version 11.0 (SPSS Inc., Chicago). Analysisf variance (ANOVA) for repeated measures was used tonalyze the differences in plasma lipids levels and NF-�Bctivation. When statistically significant effects were demon-trated, Tukey’s post hoc comparison test was used to identifyroup differences. Analysis of covariance (ANCOVA) wassed to analyze the VCAM-1 data. Differences were consid-red significant when p < 0.05. All data are presented in theext and tables as means ± S.E.

. Experimental results

.1. Diet intake and lipids profile

Before the dietary intervention study, all volunteers haveormal lipids parameters (data not shown). The intake dur-

ng 4 weeks of Western diet induced a significant incrementn plasma concentrations of total cholesterol (p < 0.05) andDL-cholesterol (p < 0.05) compared with the ingestion of aediterranean and CHO enriched in vegetal n − 3 fatty acids

c(Np

1.20 ± 0.03 1.22 ± 0.03

iet. Apo B was lower after both Mediterranean and CHO− 3 diets, compared with a Western diet rich in saturated

atty acids. No significant differences were found betweenhe three nutritional models on triacylglycerols, Apo A-I andDL-cholesterol (Table 1).

.2. Diet intake and NF-κB activation

To examine whether chronic ingestion of three differentiets could regulate NF-�B activation, blood samples wereollected at the end of each dietary period. Electrophoreticctivity shift assay of nuclear proteins from PBMCs showedretarded band that was increased 2.7-fold with Western dietompared with Mediterranean diet (p = 0.038) and 1.79-foldhan the intake of a high CHO enriched in n − 3 (p = 0.07).o differences were found in NF-�B activation between

he Mediterranean and high CHO enriched in n − 3 dietsp = 0.230) (Fig. 1).

.3. Adhesion molecules plasma concentrations

One of the adhesion molecules induced by NF-�B activa-ion is VCAM-1. For this reason, we explored the effect of aiet intake on circulating VCAM-1 concentrations.

Fig. 2 shows VCAM-1 plasma concentrations. West-rn diet increase VCAM-1 plasma concentrations comparedith a Mediterranean and a high CHO enriched in n − 3iets (663 ± 97 versus 495 ± 35 and 486 ± 49, respectively,< 0.05). However, no significant differences were found inCAM-1 concentrations between the ingestion of a Mediter-

anean diet and a high CHO diet enriched in n − 3. In addition,e measured plasma levels of ICAM-1, MCP-1, IL-6 andNF-� at the end of the dietary intervention period with theim of determining the effect of different types of fats on theseolecules. Comparisons of the effects of the three diets on

he concentrations of these parameters revealed no signifi-ant differences (ICAM-1 p = 0.596; MCP-1 p = 0.513; IL-6= 0.463; TNF-� p = 0.181).

On the other hand, we have studied the relationshipetween NF-�B and VCAM-1, and lipid and lipoprotein con-entrations at the end of the dietary intervention period. No

orrelations were found for NF-�B and lipids parameterstotal cholesterol, LDL-C, TC/HDL-C and LDL-C/HDL-C).or was a significant correlation between VCAM-1 and thesearameters observed.

e144 P. Perez-Martinez et al. / Atheroscle

Fig. 1. NF-�B activation in peripheral mononuclear cells obtained after theintake of the three dietary models. (A) Representative EMSA of nuclearproteins from peripheral mononuclear cells. (B) Densitometer quantificationof NF-�B activity in peripheral mononuclear cells. Two-factor analysis ofvariance (ANOVA) for repeated measures. Results are expressed in arbitraryunits (generated by densitometer). Results are mean for all of the volunteers(n = 16) ±S.E.M. *p < 0.05 vs. Western diet.

Fig. 2. Plasma concentrations of VCAM-1 after the intake of the threedietary models. Results are means of all volunteers (n = 16) ±S.E.M.*p < 0.05 vs. Western diet.

4

MpWftaticRidoT�ophostenarheeefctfiotwoIa�hwecetostdwgw

rosis 194 (2007) e141–e146

. Discussion

Our results showed that the chronic consumption of aediterranean diet decreases the activation of NF-�B and

lasma VCAM-1 concentration compared with a typicalestern diet. The effect of a high CHO diet enriched in n − 3

atty acid was intermediate. In this study, we have adminis-rated three different diets during 4 weeks in healthy subjectsnd we have analyzed the effect of regimens on NF-�B activa-ion in PBMCs and plasma VCAM-1 concentrations. NF-�Bs a transcription factor implicated in various transcriptionallyontrolled processes such as inflammation and cell growth.ecent studies strongly suggest that NF-�B plays a key role

n the pathogenesis of atherosclerosis. Previous studies haveemonstrated that fat intake is associated with the inductionf inflammatory genes, and therefore with NF-�B activation.he consumption of a fat-enriched breakfast increased NF-B activation in humans [8]. In addition, we have recentlybserved that the acute intake of butter or walnut elicited theostprandial NF-�B activation on mononuclear cells fromealthy men [9]. In contrast, the acute consumption of oliveil decreased the activation of NF-�B system. This previoustudy was design to determine whether the acute intake ofhree foods with different fat contents had a selective influ-nce on the postprandial activation of NF-�B. Nevertheless,o data exist about the chronic effect of dietary models on thectivation of NF-�B in a healthy population, which is moreeliable for potential recommended intervention of dietaryabits. The current study is in our knowledge the first toxamine the effect of three dietary models, Mediterraneannriched in virgin olive oil, typical Western and high CHOnriched in n − 3 diets on the activation of NF-�B in PBMCsrom healthy volunteers. Interestingly, we have observed thatonsumption of the Mediterranean diet decrease the activa-ion of NF-�B compared with a typical Western diet. Thesendings suggest that olive oil, which is the principal sourcef fat of the Mediterranean diet, could prevent the activa-ion of NF-�B system, and this effect could be associatedith either the intake of MUFAs or the protective influencef the antioxidants components contained in virgin olive oil.n these sense, it is known that antioxidants reduce NF-�Bctivity. These protective effects of the olive oil on the NF-B activity, partly attributed to its antioxidant compoundsave also been suggested for red wine, indicating that foodsith higher antioxidant content could have a major inhibitory

ffect on NF-�B. Besides olive oil, the Mediterranean dietontents other source of potentially cardioprotective nutri-nts from fruits and vegetables which could also enhancehis beneficial effect [18]. In contrast, the opposite effect wasbserved after the chronic intake of a Western diet rich inaturated fatty acids, corroborating our previous data afterhe acute intake of a butter meal. The effect of a high CHO

iet enriched in n − 3 fatty acids on the NF-�B activationas intermediate. In this sense, previous studies have sug-ested that n − 3 �-linolenic acid found mainly in plants andalnuts may reduce cardiovascular risk through a variety of

eroscle

bmeNatsiotfitd

eiiVtNiMccliHooibsmioh

tMNoaagNvcto

5

e

ve

A

f00Jtitwe

R

[

[

[

[

[

P. Perez-Martinez et al. / Ath

iological mechanisms, including inhibiting vascular inflam-ation [19]. In our study, the intake of a high CHO diet

nriched in n − 3 fatty acids showed a tendency to decreaseF-�B activation, indicating that �-linolenic acid exerts an

nti-inflammatory effect through the NF-�B system. In addi-ion, we also observed changes in the lipid profile of healthyubjects after the consumption of the different diets. Thentake of a saturated diet increases the plasma concentrationsf total cholesterol and LDL-C compared with the inges-ion of a Mediterranean and CHO enriched in vegetal n − 3atty acids. These data corroborate that dietary saturated fatntake is significantly associated with serum cholesterol andhe risk of coronary heart disease compared with the other twoiets.

NF-�B has been shown to regulate the expression of sev-ral adhesion molecules in response to inflammatory stimuli,ncluding P-selectin, E-selectin, ICAM-1 and VCAM-1, allmplicated in atherosclerosis development. We have studiedCAM-1 plasma concentrations after the consumption of

he three diets, because this protein is also regulated by theF-�B system. Consequently, the intake of a Western diet

ncreases VCAM-1 plasma concentrations compared with aediterranean and a high CHO enriched in n − 3 diets, indi-

ating that the last two diets may provide additional benefitsompared with the typical Western diet. As well as the plasmaevels of VCAM-1, we determined other components of thenflammatory response: ICAM-1, MCP-1, IL-6 and TNF-�.owever, we did not observe significant differences in anyf these parameters at the end of each of the dietary peri-ds. The fact that, in our study, we only found differencesn plasma VCAM levels may be explained, at least in part,y the characteristics of our sample population, which con-isted of completely healthy young people, which may haveinimized the effects observed. For this reason, it would be

nteresting to replicate our study using persons with variousther sets of characteristics: metabolic syndrome, hyperc-olesterolemia, diabetes, etc.

These data, together with previous studies, suggest thathe anti-inflammatory effects associated with the intake of

editerranean diet could be related with the diminution onF-�B activation. However, despite the significant advancesf the recent years, the final proof about the specific mech-nisms and contributing role of the different dietary modelsnd nutrients to its beneficial effects requires further investi-ations. On the other hand, we found no relationship betweenF-�B and lipid parameters at the end of the dietary inter-ention period. Nor was this effect observed in the plasmaoncentrations of VCAM-1. This fact may serve to sustainhe hypothesis of an anti-inflammatory activity independentf the effects on lipid parameters.

. Conclusion

In conclusion, this study suggests that the cardioprotectiveffect of a chronic intake of a Mediterranean diet, based in

[

rosis 194 (2007) e141–e146 e145

irgin olive oil, could be due, at least in part, by its protectiveffect on NF-�B activation.

cknowledgments

This work was supported in part by research grantsrom the Spanish Ministry of Education and Science (SAF1/0366, AGL2006-01979/ALI to J L-M, SAF/01/2466-C054 to F P-J), the Spanish Ministry of Health (FISS 01/0449 toL-M) and Consejerıa de Salud, Junta de Andalucia (65/02

o J L-M), the CAM (08.4/0021.1/2003 to J E) and the Span-sh Cardiovascular Network (to J E). We extend our thankso Canoliva (Antonio Cano e Hijos SA, Luque, Cordoba),ho generously donated the virgin olive oil for the dietary

xperiments, and also to the volunteers.

eferences

[1] Kris-Etherton P, Eckel RH, Howard BV, et al. Lyon Diet Heart Study.Benefits of a Mediterranean-style, National Cholesterol EducationProgram/American Heart Association Step I Dietary Pattern on Car-diovascular Disease. Circulation 2001;103:1823–5.

[2] Perez-Jimenez F. International conference on the healthy effect of virginolive oil. Eur J Clin Invest 2005;35:421–4.

[3] Carluccio MA, Siculella L, Ancora MA, et al. Olive oil and red wineantioxidant polyphenols inhibit endothelial activation: antiatherogenicproperties of Mediterranean diet phytochemicals. Arterioscler ThrombVasc Biol 2003;23:622–9.

[4] Kriszbacher I, Koppan M, Bodis J. Inflammation, atherosclerosis, andcoronary artery disease. N Engl J Med 2005;353:429–30.

[5] de Winther MP, Kanters E, Kraal G, et al. Nuclear factor kappaB signal-ing in atherogenesis. Arterioscler Thromb Vasc Biol 2005;25:904–14.

[6] Kutuk K, Basaga H. Inflammation meets oxidation: NF-�B as a medi-ator of initial lesion development in atherosclerosis. Trends Mol Med2003;9:549–57.

[7] Ortego M, Gomez-Hernandez A, Vidal C, et al. HMG-CoA reductaseinhibitors reduce I kappa B kinase activity induced by oxidative stress inmonocytes and vascular smooth muscle cells. J Cardiovasc Pharmacol2005;45:468–75.

[8] Blanco-Colio LM, Valderrama M, Alvarez-Sala LA, et al. Red wineintake prevents nuclear factor-kappaB activation in peripheral bloodmononuclear cells of healthy volunteers during postprandial lipemia.Circulation 2000;102:1020–6.

[9] Bellido C, Lopez-Miranda J, Blanco-Colio LM, et al. Butter andwalnuts, but not olive oil, elicit postprandial activation of nuclear tran-scription factor kappa B in peripheral blood mononuclear cells fromhealthy men. Am J Clin Nutr 2004;80:1487–91.

10] Human Nutrition Information Service, Department of AgricultureComposition of foods. Agriculture handbook, no. 8. Washington, DC:US Government Printing Office; 1987.

11] Varela G. Food composition tables (Tablas de composicion de alimen-tos). Madrid: Instituto de Nutricion, CSIC; 1980.

12] Allain CC, Poon LS, Chang CSG, et al. Enzymatic determination oftotal serum cholesterol. Clin Chem 1974;20:470–5.

13] Bucolo G, David H. Quantitative determination of serum triglyceridesby use of enzymes. Clin Chem 1973;19:476–82.

14] Assmann G, Schierwer H, Schmitz G, et al. Quantification of high

density lipoprotein cholesterol by precipitation with phosphotungsticacid-MgCl2. Clin Chem 1983;29:2026–30.

15] Riepponem P, Marniemi J, Rautaoja T. Immunoturbidimetric determi-nation of apolipoproteins A-I and B in serum. Scand J Clin Lab Invest1987;47:739–44.

e eroscle

[

[

146 P. Perez-Martinez et al. / Ath

16] Friedewald WT, Levy RI, Fredrickson DS. Estimation of theconcentration of low-density lipoprotein cholesterol in plasma with-

out use of a preparative ultracentrifuge. Clin Chem 1972;18:499–502.

17] Hernandez-Presa M, Bustos C, Ortego M, et al. Angiotensin-convertingenzyme inhibition prevents arterial nuclear factor-�B activation, mono-cyte chemoattractant protein-1 expression, and macrophage infiltration

[

[

rosis 194 (2007) e141–e146

in a rabbit model of early accelerated atherosclerosis. Circulation1997;95:1532–41.

18] Wahle KW, Caruso D, Ochoa JJ, et al. Olive oil and modulation of cellsignaling in disease prevention. Lipids 2004;39:1223–31.

19] De Caterina R, Madonna R, Massaro M. Effects of omega-3 fattyacids on cytokines and adhesion molecules. Curr Atheroscler Rep2004;6:485–91.