lable at ScienceDirect

Atherosclerosis 235 (2014) 649e658

Contents lists avai

Atherosclerosis

journal homepage: www.elsevier .com/locate/atherosclerosis

Extra virgin olive oil blunt post-prandial oxidative stress via NOX2down-regulation

Roberto Carnevale a, 1, Pasquale Pignatelli a, 1, Cristina Nocella a, Lorenzo Loffredo a,Daniele Pastori a, Tommasa Vicario a, Andreina Petruccioli b, Simona Bartimoccia a,Francesco Violi a, *

a Department of Internal Medicine and Medical Specialties, Sapienza University of Rome, Rome, Italyb AFC Patrimonio Servizi e furniture UO ristorazioni, Policlinico Umberto I, Rome, Italy

a r t i c l e i n f o

Article history:Received 8 October 2013Received in revised form30 April 2014Accepted 27 May 2014Available online 19 June 2014

Keywords:Extra virgin olive oilOxidative stressNOX2Mediterranean diet

* Corresponding author. Divisione I Clinica MedicRoma 00161, Italy. Tel.: þ39 0 64461933; fax: þ39 0

E-mail address: francesco.violi@uniroma1.it (F. Vio1 Drs Carnevale and Pignatelli equally contributed

http://dx.doi.org/10.1016/j.atherosclerosis.2014.05.9540021-9150/© 2014 Elsevier Ireland Ltd. All rights rese

a b s t r a c t

Objective: Olive oil protects against cardiovascular disease but the underlying mechanism is still unclear.We speculated that olive oil could inhibit oxidative stress, which is believed to be implicated in theatherosclerotic process.Methods and results: Post-prandial oxidative stress and endothelial dysfunction were investigated intwenty-five healthy subjects who were randomly allocated in a cross-over design to a Mediterranean dietadded with or without extra virgin olive oil (EVOO, 10 g) (first study, n ¼ 25) or Mediterranean diet withEVOO (10 g) or corn oil (10 g) (second study, n ¼ 25). Oxidative stress biomarkers including plateletreactive oxidant species (ROS) and 8-iso-PGF2a-III, activity of NOX2, the catalytic sub-unit of NADPHoxidase, as assessed in platelets and serum, serum vitamin E and endothelial dysfunction, were measuredbefore and 2 h after lunch.In the first study a significant increase of platelet ROS, 8-iso-PGF2a-III, NOX2 activity, sE-selectin,sVCAM1 and a decrease of serum vitamin E were detected in controls but not when EVOO wasincluded in the Mediterranean diet; oxidative stress and endothelial dysfunction increase were alsoobserved in the second study in subjects given corn oil. A significant correlation was found betweenNOX2 activity and platelet oxidative stress. In vitro study demonstrated that EVOO but not corn oilsignificantly decreased platelet and PMNs oxidative stress and NOX2 activity.Conclusion: The study provides the first evidence that post-prandial oxidative stress may be triggered byNOX2 up-regulation. EVOO but not corn oil, is able to counteract such phenomenon suggesting thataddition of EVOO to a Mediterranean diet protects against post-prandial oxidative stress.

© 2014 Elsevier Ireland Ltd. All rights reserved.

1. Introduction

Oxidative stress is believed to play a major role in the athero-sclerotic process via facilitating LDL oxidation within artery wall,which ultimately leads to foam cell formation and atheroscleroticplaque [1]. Thus, evidence is accumulating that oxidative stress isassociated with classic atherosclerotic risk factors reflecting aprocess of systemic inflammation that predisposes to atheroscle-rosis. However, recent data focus also on the possibility of an abruptincrease of oxidative stress occurring after meals [2]. Thus,

a, Viale del Policlinico 155,649970103.li).to this work.

rved.

exaggerated post-prandial increase of lipid and glucose is associ-ated with a burst of reactive oxidant species (ROS), which elicitsystemic inflammation and endothelial dysfunction [3e6]. Suchchanges have been suggested to facilitate the occurrence of car-diovascular disease [7].

There is consolidated clinical evidence that Mediterranean dietis associated with a lower risk of cardiovascular events includingmyocardial infarction, stroke and cardiovascular death [8]. It isunclear, however, if Mediterranean diet as a whole or its specificcomponents such as fresh fruit, vegetables, legumes, cereals areimplicated in protecting from cardiovascular events. In a recentlarge study in general population Mediterranean diet supple-mented with extra virgin olive oil protected from vascular diseasesuggesting a key role for olive oil as an anti-atherosclerotic nutrient[9]. As olive oil is usually assumed with meals, it could be possiblethat it protects from the abrupt increase of post-prandial oxidative

R. Carnevale et al. / Atherosclerosis 235 (2014) 649e658650

stress. Studies which addressed the question as to whether olive oilprotects from post-prandial oxidative stress provided equivocaldata as inhibition or no effect on oxidative stress have been re-ported [10,11]. Furthermore is still unclear the mechanism poten-tially accounting for the antioxidant activity of extra virgin olive oil.Several enzymatic pathways concur to the cellular production ofROS including NADPH oxidase, myeloperoxidase, xanthine oxidaseand uncoupled eNOS [12,13]. Among them, NADPH oxidase plays amajor role as shown by the almost complete cellular ROS sup-pression in patients with hereditary deficiency of NOX2, the cata-lytic sub-unit of the enzyme [14]. We speculated that extra virginolive could lower oxidative stress by functionally interfering withthe activity of NOX2. Thus, we explored if in healthy subjects atypical Mediterranean diet is associated with changes in the acti-vation of NOX2 and if the supplementation of extra virgin olive oil isable to blunt post-prandial oxidative stress generated by thisenzyme.

2. Materials and methods

2.1. Interventional study with oil

Twenty-five healthy subjects (males 12, females 13 age35.9± 2.1 years) (see Table 1, Supplementary Appendix), working inour research Institute, gave informed consent to participate in theinterventional study, which was performed between January andMarch 2013. In the two weeks before the study all the volunteerswere instructed to follow a Mediterranean-type lunch, which wasconsumed in a dedicated building of the hospital. None of theparticipant took aspirin or vitamin E supplements or other anti-oxidant in the month preceding the study that was divided in twoparts. In the first part 25 subjects were randomized to receive atypical Mediterranean lunch including or not 2 spoons of olive oilcorresponding to 10 g extra virgin olive oil (EVOO) (see Table 2,Supplementary Appendix) in a cross-over design; there was aninterval of 30 days between the two phases of the study. After onemonth from the end of the first study the same subjects (n ¼ 25)participated to a second study in which they were randomly allo-cated to receive a lunch with 2 spoons of EVOO (10 g) or 3 spoons ofcorn oil (10 g) (see Table 2, Supplementary Appendix) in a cross-over design. There was an interval of 30 days between the twophases of the study. For each phase of the study a blood sample wastaken before (at 13 pm) and 2 h after the lunch; in each bloodsample oxidative stress and vitamin E were measured. Every blooddetermination was performed blind.

2.2. Randomization and blinding

A person, not involved in the study, assigned codes to the studytreatments, randomly allocated the selected participants to atreatment sequence with no EVOO, EVOO or corn oil, and kept thekey in a sealed envelope. The randomization was carried out by aprocedure based on a random numeric sequence. The authors wereunaware of treatment allocation.

2.3. Laboratory analyses

All materials were from Sigma Aldrich unless otherwise speci-fied. Samples, obtained from patients after supine rest for atleast10 min, were taken into tubes with 3.8% sodium citrate andcentrifuged at 300� g for 10 min to obtain supernatant. Plasma andserum samples were immediately stored at �80 �C until use withthe antioxidant butylated hydroxytoluene (BHT) at final concen-tration of 20 mmol/l.

2.4. In vitro study

Experimental study was conducted in vitro to see if addition ofEVOO, corn oil or a-tocopherol (vitamin E) to platelets or PMNsaffected oxidative stress. To this purpose platelets and PMNs fromhealthy subjects (males 3, females 2 age 34.7 ± 1.1) were incubatedwith liposomes and stimulated with arachidonic acid (AA) orPhorbol Myristate Acetate (PMA). Analysis of cells ROS production,8-iso-PGF2a-III formation and NOX2 activation was carried out asbelow reported.

2.5. Platelet preparation and activation

Platelet-rich plasma (PRP) was obtained according to Pignatelliet al. [15]. Cell suspensionwas incubated30minat37 �Cunder stirringcondition inpresence or not of LPC/EVOO, LPC/corn oil, scalar doses ofvitamin E (1e20 mM) or a scalar dose of oleuropein (0.1e10 mM) andstimulated with 0.5 mM AA. Cells were separated from the superna-tant by centrifugation (5 min, 300 g) and stored until analysis.

2.6. Human polymorphonuclear leukocyte preparation andactivation

Polymorphonuclear leukocytes (PMNs) were isolated fromfreshly taken EDTA-blood from healthy volunteers by dextranenhanced sedimentation of red blood cells. Cell suspension wasincubated 30min at 37 �Cunder stirring condition inpresenceor notof LPC/EVOO, LPC/corn oil or scalardoses of vitaminE (1e20mM)andstimulated with 10 mM PMA. Cells were separated from the super-natant by centrifugation (5 min, 300 g) and stored until analysis.

2.7. Preparation of liposomes

Liposomes were composed of dimyristoylphosphatidylcholine(LPC) and either EVOO or corn oil. Using a rotary evaporator, theliposomes were dried by removing (all of the) chloroform atapproximately 30 �C. The swelling solution volume was alwayschosen such that the final phospholipid concentration in theaqueous dispersion was 10 mM, EVOO was 10 mg/ml and corn oil10 mg/ml. The amount of liposomes that was added to each samplewas adjusted to approximately 50 nM of phospholipids, 0.01mg/mlof EVOO and 0.01 mg/ml of corn oil.

The amount of liposomes that was added to each sample wasadjusted to approximately 50 nM of phospholipids and 10 ml ofEVOO (ratio buffer/oil 5:1; final concentration of vitamin E ¼ 7 mM)or 50 nM of phospholipids and corn oil (ratio buffer/oil 5:1; finalconcentration of vitamin E ¼ 0.4 mM).

2.8. Cell ROS production

Cells were treated with control medium, LPC/EVOO, LPC/cornoil, or a scalar dose of vitamin E (1e20 mM) for 10 min at 37 �C.Samples were incubated with 2',7'-dichlorofluorescin diacetate5 mM for 15 min at 37 �C and treated with or without AA (0.5 mM)for platelets or PMA (10 mM) for PMNs. Then 10 ml of each samplewas diluted with 1 ml of PBS and analyzed by flow cytometry.Values were expressed as mean fluorescence (M.F.).

2.9. Platelet and PMNs sNOX2-dp levels and 8-iso-PGF2a-IIIformation

Extracellular levels of soluble NOX2-derived peptide (sNOX2-dp), a marker of NADPH oxidase activation, were detected byELISA method as previously described by Pignatelli et al. [15].Values were expressed as pg/ml; intra-assay and inter-assay

R. Carnevale et al. / Atherosclerosis 235 (2014) 649e658 651

coefficients of variation were 5.2% and 6%, respectively. Cells iso-prostanes (8-iso-PGF2a-III) were measured by EIA assay method(Tema Ricerca, Italy) and expressed as pmol/L. Intra- and inter-assay coefficients of variation were 5.8% and 5.0%, respectively.

2.10. Detection of endothelial dysfunction

EIA assay method (Tema Ricerca, Italy) was used to determine thesVCAM1 and sE-selectin concentration, twomolecules of endothelialdysfuncion [16]. Values were expressed, as ng/ml, intra-assay andinter-assay coefficients of variation were 5.2% and 6% respectively.

2.11. Serum vitamin E

Serumoroil samples (100mL)were supplementedwith tocopherylacetate (internal standard), deproteinized by the addition of ethanol,and extracted with hexane. Phase separation was achieved by

Fig. 1. Interventional study: platelet ROS production (Panel A), platelet sNOX2-dp release (Pand serum vitamin E (Panel E) before and after 2 h of a 900-kcal mixed-meal with or with

centrifugation. The collected upper phase was evaporated andanalyzed using an Agilent 1200 Infinity series high-performanceliquid chromatography system equipped with an Eclipse Plus C18column (4.6 � 100 mm) [17]. Serum levels were given as the ratio(mmol/mmol) between serum a-tocopherol concentration (mmol/L)and serum total cholesterol concentration (mmol/L), which betterexpress the circulating levels of vitamin E [18].

2.12. Serum sNOX2-dp

Serum levels of sNOX2-dp were detected by ELISA method aspreviously described [15].

2.13. Oil total polyphenol content

Total polyphenol content in EVOO and corn oil was determinedby a modified FolineCiocalteu colorimetric method [19]. Briefly,

anel B), platelet 8-iso-PGF2a-III formation (Panel C), serum sNOX2-dp release (Panel D)out extravergin olive oil (EVOO), *p < 0.01, **p < 0.05.

R. Carnevale et al. / Atherosclerosis 235 (2014) 649e658652

0.1 mL of FolineCiocalteu reagent was added to 0.02 mL of oils.After 5 min at room temperature, 0.05 mL of 20 g L�1 sodiumcarbonate was added and the reaction mixture was incubated at37 �C for 20 min. The absorbance at 765 nmwas measured using anAsys UVM 340 microplate reader (Biochrom, Holliston, USA) andcompared with a gallic acid calibration curve (100e1000 mgL�1).Results were expressed as mg/ml of gallic acid equivalent (GAE). Allexperiments were performed in triplicate.

2.14. Sample preparation and polyphenols content

For HPLC analysis, samples were prepared according to Schro-eter et al. [20]. Briefly, samples weremixed with twice their volumeof acidified methanol (100% vol/vol, e20 �C, internalstandard¼ 30EthylEC) and centrifuged at 17,000 g for 15min at 4 �C.The supernatant was collected and the pellet was resuspendedtwice in methanol and centrifuged at 17,000 g for 15 min at 4 �C.After centrifugation, the supernatant was collected and the sol-vents were evaporated to dryness under nitrogen.

The determination of epicatechin, catechin, Epigallocatechin-3-gallate (ECGC) and oleuropein were carried out using an Agilent1200 Infinity series HPLC system equipped with an Eclipse plus C18column (4.6 � 100 mm). All determination were undertaken at25 �C. Separation of analytes were accomplished by using anacetonitrile gradient in 50 mM methanolic (4% vol/vol) sodiumacetate (pH 4.4) with a flow rate of 0.8 ml/min. Acetonitrile con-centrations were linearly increased from 0% to 10% between mi-nutes 5 and 20. Thereafter, acetonitrile concentrations were furtherincreased in linear segments (20e28 min, 10e12%; 28e34 min,

Fig. 2. Interventional study: sVCAM1 levels (Panel A) and sE-selectin levels (Panel B) before*p < 0.01. Interventional study: sVCAM1 levels (Panel C) and sE-selectin levels (Panel D) befo*p < 0.01.

12e20%; 34e41min, 20e30%; 41e45min, 30e71%) and held at 71%for 10 min. We used the UV detection at 280 nm for sample anal-ysis. Chromatographic peaks of analytes were identified bycomparing their retention times with those of the standard.

2.15. Sample size

As regards the interventional cross-over study, we computedthe minimum sample size with respect to a two-tailed one-sampleStudent's t-test, considering as (i) platelet sNOX2-dp variation to bedetected between extra virgin oil and germ oil treatment jdj � 9 pg/ml, (ii) standard deviation of the paired differences SD ¼ 6.0 pg/ml,(iii) type I error probability a ¼ 0.05 and power 1�b ¼ 0.90. Thisresulted in a minimum sample size of 10 per group.

2.16. Statistical methods

Categorical variables are reported as counts (percentage) andcontinuous variables as means ± SD unless otherwise indicated.Independence of categorical variables was tested by c2-test. Com-parisons between groups were carried out by Student's t-test andwere replicated as appropriate with nonparametric tests (Kolmo-goroveSmirnov (z) test in case of non-homogeneous variances asverified by Levene's test).

The cross-over study data were analyzed for the assessment oftreatment and period effects, by performing a split-plot ANOVAwith one between-subject factor (treatment sequence) and twowithin-subject factors (period 1 vs. 2; pre- vs. post-treatment) [21].The full model was considered, allowing for the assessment of all

and after 2 h of a 900-kcal mixed-meal with or without extravergin olive oil (EVOO),re and after 2 h of a 900-kcal mixed-meal with extarvergin olive oil (EVOO) or corn oil,

R. Carnevale et al. / Atherosclerosis 235 (2014) 649e658 653

main effects and interactions. Pairwise comparisons were correctedby the Bonferroni test; results were expressed as means ± SE.

Bivariate analysis was performed with Pearson's linear regres-sion test.

A value of p < 0.05 was considered statistically significant. Allanalyses were carried out with SPSS-13.0 software (SPSS Inc.).

3. Results

3.1. Interventional study

Clinical characteristics of healthy subjects who participated in thestudy are depicted in Table 1 (Supplementary Appendix). Total andsingle polyphenol content was significantly higher in EVOOcompared to corn oil (Table 3, Supplementary Appendix) (p < 0.001).

In the first study, a significant difference for treatments (mealcontaining vs not containing EVOO) was found with respect to

Fig. 3. Interventional study: platelet ROS production (Panel A), platelet sNOX2-dp release (Pand serum vitamin E (Panel E) before and after 2 h of a 900-kcal mixed-meal with extarve

serum vitamin E (F ¼ 8.4, p ¼ 0.006; Fig. 1E), platelet ROS (F ¼ 5.4,p ¼ 0.03; Fig. 1A), platelet sNOX2-dp release (F ¼ 8.2, p ¼ 0.007;Fig. 1B), platelet 8-iso-PGF2a-III formation (F ¼ 4.5, p ¼ 0.04;Fig. 1C) and serum sNOX2-dp release (F ¼ 7.1, p ¼ 0.006; Fig. 1D),sVCAM1 levels (F ¼ 18.8, p < 0.001; Fig. 2A) and E-selectin levels(F ¼ 17.1, p < 0.001; Fig. 2B) from the ANOVA performed on cross-over study data.

In particular, compared to baseline, when a meal not containingEVOO was given, platelet ROS generation increased significantly by27%, platelet sNOX2-dp release by 26% and 8-iso-PGF2a-III forma-tion by 45% (Fig. 1AeC); furthermore, serum sNOX2-dp releasesignificantly increased and vitamin E serum levels reduced by 30%(Fig. 1D and E). When a meal not containing EVOO was given,sVCAM1 levels increased significantly by 48% and E-selectin levelsby 47% (Fig. 2A and B).

Conversely, compared to baseline, in subjects given a mealcontaining EVOO, a non-significant increase of platelet ROS

anel B), platelet 8-iso-PGF2a-III formation (Panel C), serum sNOX2-dp release (Panel D)rgin olive oil (EVOO) or corn oil, *p < 0.01.

R. Carnevale et al. / Atherosclerosis 235 (2014) 649e658654

generation, platelet sNOX2-dp release and 8-iso-PGF2a-III forma-tion, serum sNOX2-dp release and a not significant reduction ofvitamin E were detected (Fig. 1AeE). Same results were observedfor endothelial dysfunction; sVCAM1 levels and E-selectin levels(Fig. 2A and B).

An overall correlation analysis demonstrated a direct correlationbetween sNOX2-dp and 8-iso-PGF2a-III (R ¼ 0.646, p < 0.001),sNOX2-dp and ROS production (R¼ 0.541, p < 0.001) and an inversecorrelation between vitamin E and 8-iso-PGF2a-III (R ¼ �0.393,p < 0.001) and vitamin E and ROS production (R¼ 0.431, p < 0.001).

The second study showed a significant difference for treatmentsin subjects given EVOO or corn oil with respect to platelet ROS(F ¼ 2.9, p < 0.001; Fig. 3A), platelet sNOX2-dp release (F ¼ 3.8,p < 0.001; Fig. 3B), platelet 8-iso-PGF2a-III formation (F ¼ 4.0,p < 0.001; Fig. 3C), serum sNOX2-dp release (F ¼ 6.4, p ¼ 0.007;Fig. 3D) and serum vitamin E (F ¼ 11.9, p < 0.001; Fig. 3E), sVCAM1

Fig. 4. Platelet (Panel A) and PMNs (Panel B) ROS production, platelet (Panel C) and PMNformation in samples added with or without dimyristoylphosphatidylcholine/extra virgin o

levels (F ¼ 38.1, p < 0.001; Fig. 2C) and E-selectin levels (F ¼ 20.9,p < 0.001; Fig. 2D) from the ANOVA performed on cross-over studydata.

In particular, after 2 h from lunch, in subjects given a mealcontaining corn oil, platelet ROS generation significantlyincreased by 38%, platelet sNOX2-dp release by 48%, 8-iso-PGF2a-III formation by 34%, serum sNOX2-dp release by 33%,while vitamin E serum levels decreased by 36% (Fig. 3AeE).Moreover in subjects given a meal containing corn oil, sVCAM1levels significantly increased by 58% and E-selectin levels by 42%(Fig. 2C and D).

Conversely, in subjects given a meal containing EVOO, plateletROS generation, platelet sNOX2-dp release, 8-iso-PGF2a-III forma-tion, serum sNOX2-dp release and vitamin E serum levels did notsignificantly change (Fig. 3AeE). Same results were observed forsVCAM1 levels and E-selectin levels (Fig. 2C and D).

s (Panel D) sNOX2-dp release, platelet (Panel E) and PMNs (Panel F) 8-iso-PGF2a-IIIlive oil (LPC/EVOO) in healthy subjects (n ¼ 5).

R. Carnevale et al. / Atherosclerosis 235 (2014) 649e658 655

At the end of the study markers of oxidative stress and endo-thelial dysfunction were significantly lower in subjects given EVOOcompared to those given corn oil (Fig. 3AeE).

3.2. In vitro study

AA-stimulated platelets and PMA-stimulated PMNs enhancedROS production (Fig. 4A and B), sNOX2-dp release (Fig. 4C and D)and 8-iso-PGF2a-III formation (Fig. 4E and F). Such changes weresignificantly reduced by cell incubation with EVOO (17 mM ofvitamin E content). In particular the production of ROS wasdecreased by 41% for platelet and by 52% for PMNs, sNOX2-dprelease by 54% for platelet and by 71% and the formation of 8-iso-PGF2a-III by 52% for platelet and by 58% for PMNs(Fig. 4AeF).

Conversely, platelets incubationwith corn oil (0.4 mMof vitamin Econtent) resulted in a trend toward a reduction of oxidative stress,

Fig. 5. Platelet (Panel A) and PMNs (Panel B) ROS production, platelet (Panel C) and PMNformation in samples added with or without dimyristoylphosphatidylcholine/corn oil (LPC/

which, however, was not significant (Fig. 5). In particular the pro-duction of ROS was decreased by 6% for platelet and by 11 for PMNs,sNOX2-dp release by 14% for platelet and by 17% for PMNs and 8-iso-PGF2a-III formationby13% forplatelet andby9% forPMNs (Fig. 5AeF).

Furthermore, we investigated whether the vitamin E content ofEVOO influenced platelet and PMNs oxidative stress. Oxidativestress elicited by AA-stimulated platelets or PMA-stimulated PMNswas dose-dependently inhibited by vitamin E with a significantreduction when 20 mM was used (Fig. 6).

In particular 20 mMvitamin E, which was close to that present inEVOO, inhibited ROS production by 34% for platelets and 57% forPMNs (Fig. 6B), sNOX2-dp release by 30% for platelets and by 45%for PMNs and 8-iso-PGF2a-III by 36% for platelets and 8-iso-PGF2a-III by 37% for PMNs (Fig. 6AeF).

Finally, we performed an in vitro study with a scalar dose ofoleuropein (0.1e10 mM), one of major antioxidant polyphenols ofolive oil.

s (Panel D) sNOX2-dp release, platelet (Panel E) and PMNs (Panel F) 8-iso-PGF2a-IIIcorn oil) in healthy subjects (n ¼ 5).

R. Carnevale et al. / Atherosclerosis 235 (2014) 649e658656

Compared to un-stimulated platelets, AA-stimulated plateletinduced an increased ROS production, (Fig. 7A), sNOX2-dp levels(Fig. 7B) and 8-iso-PGF2a-III formation (Fig. 7C). AA-stimulatedplatelets incubated with oleuropein showed a dose-dependentreduction of ROS production, sNOX2-dp levels and 8-iso-PGF2a-III formation (Fig. 7AeC).

4. Discussion

The study shows that in healthy subjects a typical Mediterra-nean lunch is associated with an increase of ROS-derived NOX2activation along with a reduction of vitamin E serum levels; extravirgin olive oil but not corn oil is capable of blunting this phe-nomenon by down-regulating NOX2 activation and counteractingpost-prandial serum vitamin E decrease.

Analysis of cellular and systemic markers of oxidative stressconsistently showed that a lipid- or glucose-rich meal is associated

Fig. 6. Platelet (Panel A) and PMNs (Panel B) ROS production, platelet (Panel C) and PMNformation in samples added with or without a scalar doses of vitamin E (1e20 mM) in hea

with enhanced oxidative stress [22,23]. It is still undefined, how-ever, if specific enzymatic pathways are implicated in post-prandialROS over-production.

In the present study we performed a pragmatic cross-over studywhich aimed at exploring if an usual Mediterranean lunch isassociated with oxidative stress. In accordance with the abovementioned reports we found a significant increase of oxidativestress 2 h after food intake. To explore the mechanism accountingfor post-prandial oxidative stress we measured NOX2 by an assaywhich allows to investigate the activity of NOX2 in blood [15]. Thus,upon activation, NOX2 is released in the blood and maximally re-flects the activity of platelets andwhite cells [15]. The novelty of thepresent study is in the demonstration that after food intake theactivity of NOX2, as assessed by platelet, PMNs and serum NOX2, isup-regulated. Post-prandial NOX2 activation was coincident withcellular over-production of ROS and 8-iso-PGF2a-III suggesting thatthis enzymatic pathway is implicated in post-prandial oxidative

s (Panel D) sNOX2-dp release, platelet (Panel E) and PMNs (Panel F) 8-iso-PGF2a-IIIlthy subjects (n ¼ 5).

Fig. 7. Platelet ROS production (Panel A), sNOX2-dp release (Panel B) and 8-iso-PGF2a-III formation (Panel C) in samples added with or without a scalar doses of oleuropein(0.1e10 mM) in healthy subjects (n ¼ 5).

R. Carnevale et al. / Atherosclerosis 235 (2014) 649e658 657

stress; in accordance with this hypothesis NOX2 activity wassignificantly correlated with cellular production of either ROS or 8-iso-PGF2a-III.

Coincidentally with the increase of NOX2-derived oxidativestress, vitamin E serum levels significantly decreased suggesting aconsumption of this antioxidant likely as a consequence ofenhanced oxidative stress. This hypothesis was supported by theinverse correlation between vitamin E serum levels and markers ofoxidative stress.

While there are many studies which analyzed the effect ofantioxidant supplementation on post-prandial oxidative stress wefound very fewwhich explored the antioxidant effect of extra virginolive oil in human [24,25]. One prospective study which investi-gated the effect of extra virgin olive oil rich or poor in phenolsfound no changes in biomarkers of oxidative stress as assessed byLDL susceptibility to in vitro oxidation, plasma hydro-peroxides ormalondialdehyde or total antioxidant blood capacity [26]. Inanother study Bogani et al. analyzed the acute effect of extra virginolive oil versus olive oil or corn oil on markers of oxidative stressand total antioxidant capacity. After 2 h from a morning meal theyfound a significant increase of total antioxidant capacity only insubjects given extra virgin olive oil with no change of urine H2O2, amarker of oxidative stress [10]. Despite this study had similaritywith the goal of our study, i.e analysis of extra virgin olive oil onpost-prandial oxidative stress, differences in study design andlaboratory variables does not permit to adequately compare the

two studies. Thus, we could investigate subjects given or not extravirgin olive oil and demonstrated a protective effect of extra virginolive oil on several markers of oxidative stress; also, we showed forthe first time that extra virgin olive oil down-regulated NOX2 ac-tivity, pointing to this enzymatic pathway as a mechanism ac-counting for the antioxidant activity of extra virgin olive oil. It isnoteworthy that extra virgin olive oil also blunted the decrease ofvitamin E observed in subjects given a meal without extra virginolive oil so providing indirect evidence, in accordance with Boganiet al. [10], that extra virgin olive oil enhances the body antioxidantactivity. Even if such effect is likely to be attributable to the highvitamin E content of extra virgin olive oil, hindered vitamin Econsumption secondary to reduced oxidative stress could also playa role. In contrast with the data achieved with olive oil, corn oil wasunable to counteract oxidative stress as documented by the stillpost-prandial significant increase of biomarkers of oxidative stress,endothelial dysfunction and NOX2 activation and serum vitamin Edecrease. Together these findings indicated that extra virgin oliveoil, but not corn oil, counteracts post-prandial oxidative stress byreducing ROS-derived NOX2 activation, endothelial dysfunctionand vitamin E lowering.

In order to further clarify the mechanism through which extravirgin olive oil exerted an antioxidant effect we performed in vitrostudy using platelets as a cellular source of ROS-derived NOX2activation. The study demonstrated that platelets and PMNs incu-bated with extra virgin olive oil significantly inhibited oxidative

R. Carnevale et al. / Atherosclerosis 235 (2014) 649e658658

stress as documented by reduced cell ROS and 8-iso-PGF2a-IIIformation and NOX2 regulation reinforcing the concept that extravirgin olive oil specifically affects ROS-derived NOX2. However theexperiment did not clarify if olive oil exerted this effect in virtue ornot of its vitamin E content; therefore, we conducted a furtherin vitro experiment by incubating platelets and PMNs with vitaminE alone, in a range of concentration comparable to that detected inextra virgin olive oil. In accordance with our previous study [27]vitamin E reduced platelet and PMNs oxidative stress but the in-hibition rate was lower than that observed when extra virgin oliveoil was used. This finding suggests that the antioxidant effect ofextra virgin olive oil is only partly attributable to its vitamin Econtent and that other components including polyphenols arelikely to contribute to inhibit oxidative stress. In fact, the differenteffect of EVOO and corn oil on post-prandial oxidative stress is infavor of the hypothesis that polyphenol content may be alsoresponsible for this effect as EVOO is richer in polyphenolcompared to corn oil. Accordingly, total and single polyphenolcontent was much higher in EVOO compared to corn oil. Further-more, in vitro experiments with oleuropein, the most importantpolyphenol with antioxidant properties of EVOO, showed thispolyphenol dose-dependently inhibits ROS production, NOX2 ac-tivity and 8-iso-PGF2a-III in AA-stimulated platelets.

The study has implications and limitations. The study indicatesthat evenwith a traditional Mediterranean lunch a burst of ROS canoccur and that addition of extra virgin olive oil mitigates it. Thisfinding may offer a novel interpretation of a recent trial showingthat extra virgin olive oil is a key component accounting for theprotective effect of Mediterranean diet against cardiovascularevents [9].

A limitation of the study is that our findings cannot be extrap-olated to other lunch typology, therefore further study is necessaryto establish if changes in lipid or glucose content of the mealdifferently affects post-prandial oxidative stress. In this context itworthwhile underscoring that we measured the activity of NOX2but we cannot exclude that other enzymatic pathways may beimplicated in post-prandial oxidative stress.

In conclusion we provide the first evidence that aMediterranean-type lunch is associated with up-regulation of ROS-derived NOX2 and that the extra virgin olive oil is able to counteractpost-prandial oxidative stress.

Conflicts of interest

The authors report no relationships that could be construed as aconflict of interest.

Appendix A. Supplementary data

Supplementary data related to this article can be found at http://dx.doi.org/10.1016/j.atherosclerosis.2014.05.954.

References

[1] Carnevale R, Pignatelli P, Lenti L, Buchetti B, Sanguigni V, Di Santo S, et al. LDLare oxidatively modified by platelets via GP91(phox) and accumulate in hu-man monocytes. FASEB J 2007;21:927e34.

[2] Sies H, Stahl W, Sevanian A. Nutritional, dietary and postprandial oxidativestress. J Nutr 2005;135:969e72.

[3] Gregersen S, Samocha-Bonet D, Heilbronn LK, Campbell LV. Inflammatory andoxidative stress responses to high carbohydrate and high-fat meals in healthyhumans. J Nutr Metab 2012;2012:1e8.

[4] Watanabe K, Oba K, Suzuki T, Ouchi M, Suzuki K, Futami-Suda S, et al. Oralglucose loading attenuates endothelial function in normal individuals. Eur JClin Invest 2011;41:465e73.

[5] Ceriello A, Taboga C, Tonutti L, Quagliaro L, Piconi L, Bais B, et al. Evidence foran independent and cumulative effect of postprandial hypertriglyceridemiaand hyperglycemia on endothelial dysfunction and oxidative stress genera-tion: effects of short- and long-term simvastatin treatment. Circulation2002;106:1211e8.

[6] Klein R, Myers CE, Cruickshanks KJ, Gangnon RE, Danforth LG,Sivakumaran TA, et al. Markers of inflammation, oxidative stress, and endo-thelial dysfunction and the 20-year cumulative incidence of early age-relatedmacular degeneration: the Beaver Dam Eye Study. JAMA Ophthalmol 2014.http://dx.doi.org/10.1001/jamaophthalmol.2013.7671.

[7] O’Keefe James H, Bell David SH. Postprandial hyperglycemia/hyperlipidemia(postprandial dysmetabolism) is a cardiovascular risk factor. Am J Cardiol2007;100:899e904.

[8] Gardener H, Wright CB, Gu Y, Demmer RT, Boden-Albala B, Elkind M, et al.Mediterranean-style diet and risk of ischemic stroke, myocardial infarction,and vascular death: the Northern Manhattan Study. Am J Clin Nutr 2011;94:1458e64.

[9] Estruch R, Ros E, Salas-Salvad�o J, Covas MB, Corella D, Ar�os F, et al., PREDIMEDStudy Investigators. Primary prevention of cardiovascular disease with aMediterranean diet. N Engl J Med 2013;368:1279e90.

[10] Bogani P, Galli C, Villa M, Visioli F. Postprandial anti-inflammatory and anti-oxidant effects of extra virgin olive oil. Atherosclerosis 2007;190:181e6.

[11] Pacheco YM, Bermudez B, Lopez S, Abia R, Villar J, Muriana FJ. Minor com-pounds of olive oil have postprandial anti-inflammatory effects. Br J Nutr2007;98:260e3.

[12] Cross AR, Jones OTG. Enzymatic mechanism of superoxide production. Bio-chim Biophys Acta 1991;1057:281e8.

[13] Stamler JS, Singel DJ, Loscalzo J. Biochemistry of nitric oxide and its redox-activated forms. Science 1992;258:1898e902.

[14] Violi F, Sanguigni V, Carnevale R, Plebani A, Rossi P, Finocchi A, et al. Hered-itary deficiency of gp91(phox) is associated with enhanced arterial dilatation:results of a multicenter study. Circulation 2009;120:1616e22.

[15] Pignatelli P, Carnevale R, Cangemi R, Loffredo L, Sanguigni V, Stefanutti C, et al.Atorvastatin inhibits gp91phox circulating levels in patients with hypercho-lesterolemia. Arterioscler Thromb Vasc Biol 2010;30:360e7.

[16] Thorand B, Baumert J, Chambless L, Meisinger C, Kolb H, Doring Angela, et al.,MONICA/KORA Study Group. Elevated markers of endothelial dysfunctionpredict type 2 diabetes mellitus in middle-aged men and women from thegeneral population. Arterioscler Thromb Vasc Biol 2006 Feb;26(2):398e405.

[17] Bieri JG, Tolliver TJ, Catignani GL. Simultaneous determination of alpha-tocopherol and retinol in plasma or red cells by high pressure liquid chro-matography. Am J Clin Nutr 1979;32:2143e9.

[18] Traber MG, Jialal I. Measurement of lipid-soluble vitamins-further adjustmentneeded? Lancet 2000;355:2013e4.

[19] Mosca L, De Marco C, Visioli F, Cannella C. Enzymatic assay for the determi-nation of olive oil polyphenol content: assay conditions and validation of themethod. J Agric Food Chem 2000;48(2):297e301.

[20] Schroeter H, Heiss C, Balzer J, Kleinbongard P, Keen CL, Hollenberg NK, et al.(-)-Epicatechin mediates beneficial effects of flavanol-rich cocoa on vascularfunction in humans. Proc Natl Acad Sci U S A 2006 Jan 24;103(4):1024e9.

[21] Loffredo L, Marcoccia A, Pignatelli P, Andreozzi P, Borgia MC, Cangemi R, et al.Oxidative-stress-mediated arterial dysfunction in patients with peripheralarterial disease. Eur Heart J 2007;28:608e12.

[22] Ursini F, Sevanian A. Postprandial oxidative stress. Biol Chem 2002;383:599e605.

[23] Lacroix S, Rosiers CD, Tardif JC, Nigam A. The role of oxidative stress inpostprandial endothelial dysfunction. Nutr Res Rev 2012;25:288e301.

[24] Yubero-Serrano EM, Delgado-Casado N, Delgado-Lista J, Perez-Martinez P,Tasset-Cuevas I, Santos-Gonzalez M, et al. Postprandial antioxidant effect ofthe Mediterranean diet supplemented with coenzyme Q10 in elderly men andwomen. Age (Dordr) 2011;33:579e90.

[25] Weinbrenner T, Fit�o M, de la Torre R, Saez GT, Rijken P, Tormos C, et al. Oliveoils high in phenolic compounds modulate oxidative/antioxidative status inmen. J Nutr 2004;134:2314e21.

[26] Vissers MN, Zock PL, Wiseman SA, Meyboom S, Katan MB. Effect of phenol-rich extra virgin olive oil on markers of oxidation in healthy volunteers. EurJ Clin Nutr 2001;55:334e41.

[27] Pignatelli P, Pulcinelli FM, Lenti L, Gazzaniga PP, Violi F. Vitamin E inhibitscollagen-induced platelet activation by blunting hydrogen peroxide. Arte-rioscler Thromb Vasc Biol 1999;19:2542e7.