Fast Track Paper

CONSUMPTION OF TOMATO PRODUCTS WITH OLIVE OIL BUT NOTSUNFLOWER OIL INCREASES THE ANTIOXIDANT ACTIVITY OF PLASMA

ALISON LEE, DAVID I. THURNHAM, and MRIDULA CHOPRA

Northern Ireland Centre for Diet and Health, School of Biomedical Sciences, University of Ulster, Coleraine,Northern Ireland, UK

(Received5 June2000;Revised15 August2000;Accepted7 September2000)

Abstract—Health benefits of lycopene from tomato products have been suggested to be related to its antioxidantactivity. Dietary fat may influence the absorption and hence the plasma levels and antioxidant activity of lycopene. Inthe present study, we have compared the effect of consumption of tomato products with extra-virgin olive oil vs. tomatoproducts plus sunflower oil on plasma lycopene and antioxidant levels. Results show that the oil composition does notaffect the absorption of lycopene from tomato products because similar levels of plasma lycopene (mean6 SD) wereobtained on feeding tomatoes (providing approximately 46 mg lycopene/d) for 7 d with either olive oil (0.666 0.26 vs1.20 6 0.20 mmol/l, p , .002) or sunflower oil (0.676 0.27 vs. 1.14mmol/l, p , .001). However, consumption oftomato products with olive oil significantly raised the plasma antioxidant activity (FRAP) from 9306 150 to 11186184mmol/l, p , .01) but no effect was observed when the sunflower oil was used. The change (supplementation minusstart values) in FRAP following the consumption of tomato products with oil was significantly higher for olive oil(1906 101) than for sunflower oil (29.66 99,p , .005). In conclusion, the results of the study show that consumptionof tomato products with olive oil but not with sunflower oil improves the antioxidant activity of the plasma. © 2000Elsevier Science Inc.

Keywords—Tomato, Lycopene, Antioxidant, FRAP, Olive oil, Sunflower oil, Free radicals

INTRODUCTION

The protective effect of the Mediterranean diet againstthe risk of cancer and cardiovascular disease has beenattributed to various dietary components, e.g., carote-noids from fruits and vegetables, flavonoids from the redwine and monounsaturated and polyphenolic compo-nents of the olive oil [1,2]. In vivo supplementationstudies on human subjects with “Mediterranean foods”have shown protective effects on several risk factorslinked to cardiovascular disease [3–6]. In case of caro-tenoids, it has been suggested that the dietary fat caninfluence the absorption and also the antioxidant effectsof these compounds [7,8]. Lycopene, a major carotenoidpresent in tomatoes, has been reported to show antioxi-dant activity both in vitro and in vivo [9,10]. Heating

tomatoes with fat is suggested to increase the bioavail-ability of lycopene [8] but it is not known whether thecomposition of fat can affect the bioavailability andplasma antioxidant activity when consumed with tomatoproducts. Both mono- and polyunsaturated fats are con-sidered “healthier” than saturated fat since the latter isassociated with higher blood cholesterol concentrations[11]. In addition, in human supplementation studies,monounsaturated fatty acids have been shown to reduce[6] and polyunsaturated fatty acids to increase the oxi-dation of low-density lipoprotein (LDL) [12,13].

The present study was conducted to determinewhether cooking of tomatoes with different oils (olive oilvs. sunflower oil) affects the plasma lycopene concen-tration and total plasma antioxidant activity.

METHODS

Subjects and study design

Eight healthy subjects (5 female, 3 male), average age22 years (range 20–24) were recruited for the study.

Address correspondence to: Dr. Mridula Chopra, NICHE, School ofBiomedical Sciences, University of Ulster, Coleraine BT52 1SA,Northern Ireland, UK; Tel:144 (0) 2870 324057/324870; Fax:144(0) 2870 324965; E-Mail: m.chopra@ulst.ac.uk; mridulachopra@hotmail.com.

Free Radical Biology & Medicine, Vol. 29, No. 10, pp. 1051–1055, 2000Copyright © 2000 Elsevier Science Inc.Printed in the USA. All rights reserved

0891-5849/00/$–see front matter

PII S0891-5849(00)00440-8

1051

Subjects’ lipid profile, blood count, and liver functionwere within the normal range. Average weight and bodymass index of the subjects were 606 2.1 Kg and 2262.6 Kg/m2, respectively. Ethical approval for the studywas obtained from the University of Ulster Ethical Com-mittee and all subjects gave signed informed consentprior to participation in the study. Two subjects withdrewat the beginning of the study: one did not turn up for thebaseline blood sample and a second subject withdrew forpersonal reasons. Six subjects (5 female and 1 male)completed the study. Subject number was consideredsufficient for the study as several studies have shown asignificant effect of treatment with a smaller number(5–7) of subjects [5,14,15]. Subjects were allowed tocontinue their normal dietary habits but asked to incor-porate an additional intake of tomato products with eitherolive or sunflower oil.

The study period was 5 weeks. Blood samples wereobtained from subjects following an overnight fast of atleast 10 h. At the baseline, 15 ml blood sample (10 ml inheparinized, 5 ml in plain tubes) was obtained withVacutainers (Becton Dickinson Vacutainer Systems, Ce-dex, France) and subjects were asked to consume 200 gof tomato soup (Heinz Co. Ltd., Uxbridge, UK) provid-ing approximately 33 mg lycopene, and 230 g cannedtomatoes (Safeway, Middlesex, UK) providing approxi-mately 13 mg lycopene, with 20 ml of olive oil (Greekextra-virgin olive oil, Safeway, Middlesex, UK) everyday for 7 d. Tomato soup was heated with oil prior toconsumption and most subjects either consumed it on itsown or with bread. Canned tomatoes were also con-sumed after heating with oil and mixed with other food(i.e., meat, pasta, lasagne, vegetables, etc.). At the end ofthe first supplementation period, a second blood samplewas obtained. Subjects were asked to return after a 3week washout period and were supplemented for a fur-ther 7 d with the similar amounts of tomato products plus20 ml sunflower oil (Flora sunflower oil, Crawley, UK)and blood samples were obtained before and after thesupplementation. Fatty acid composition and vitamin Econtent of the oils is shown in Table 1.

Subjects were asked to complete food diaries 2 d priorto the supplementation and during the supplementation

weeks. These food diaries were used to check compli-ance and to calculate the exact intake of lycopene usingthe database on lycopene content of tomato-based prod-ucts [16,17]. All subjects consumed the specifiedamounts of tomato products and oils, and other dietaryhabits of subjects remained consistent during the courseof the study.

Blood collection and analysis

The following analyses were done on the blood: Se-rum samples were analyzed for triglycerides (TG) andcholesterol (total, LDL, and HDL) at the local healthauthority laboratory. Li-heparin blood was used to mea-sure plasma lycopene levels using high performanceliquid chromatography (HPLC) [18], and the antioxidantactivity of the plasma was measured on COBAS FARAcentrifugal analyzer using the test to assay ferric-reduc-ing ability of the plasma (FRAP, total antioxidant activ-ity) [19].

Lipid analysis and antioxidant activity measurementswere done on the fresh samples. For the plasma lycopeneanalysis, all samples were stored at280°C and theanalysis of the whole batch of samples was done on thesame day and completed within 6 weeks of completionof the study.

A pooled plasma sample stored at280°C was used tomeasure the precision of the HPLC and FRAP assay.Within-assay precision calculated from five measure-ments of the same sample was 0.18% for the FRAP assayand 6% for the plasma lycopene levels. Between-assayprecision calculated from the analysis of the storedplasma at each blood sampling time was found to be 4%for the FRAP assay.

Statistical analysis

The data was found to be skewed. The statisticalanalysis was therefore done on the log10 transformeddata with the level of significance set atp , .05. Fol-lowing a significant effect of time with ANOVA re-peated measures analysis, within-subject comparisonswere done using pairedt-test to detect which time peri-ods were different. The results are shown as the geomet-ric mean (antilog of logarithm data) and standard devi-ation (antilog [log mean plus standard deviation]2geometric mean).

RESULTS AND DISCUSSION

In our previous study [17,20] we have shown that anincrease in the dietary intake of carotenoid to 30 mg/d forone week increases the resistance of LDL to oxidation. A

Table 1. Fatty Acid Composition of the Oil

Concentration/100 ml

Florasunflower oil

Greek-extra-virginolive oil

Saturated fatty acid 12.0 g 2.0 gMonounsaturated fatty acid — 10.0 gPoly-unsaturated fatty acid 63.0 g 1.1 gVitamin E 65.0 mg 4.7 mg

1052 A. LEE et al.

supplementation period of 7 d was selected in the presentstudy because in our previous study we did not find asignificant difference in plasma carotenoids between d 7and 14 of supplementation. In the present study, themean dietary intake of lycopene was increased from, 5mg/d to 46 mg/d during the supplementation period. Theconsumption of similar amounts of tomato products witholive oil and sunflower oil produced respectively approx-imately 80% and 70% increase in the plasma lycopenelevels, but the percent increase was not significantlydifferent between the two oils used during the supple-mentation period. Although both olive oil and sunfloweroil contained vitamin E (Table 1), no change in theplasmaa-tocopherol was observed during the supple-mentation period. Polyunsaturated fatty acid rich oil isreported to have a larger impact on reducing triglycerideconcentration than monounsaturated fatty acid rich oil[21]. However, in the present study a reduction in trig-lycerides was observed only following olive oil supple-mentation (p 5 0.01). Supplementation with both oilsfailed to show an effect on the plasma cholesterol levels.Most studies that show a lipid lowering effect of sun-flower or olive oil have included a supplementation pe-riod of $ 2 weeks [21,22]. It is therefore possible that thesize of the group and/or duration of the experiment werenot sufficient to show a major change in blood lipidlevels. Nevertheless, the change in lipids when comparedfor olive oil and sunflower oil treatment periods wasfound to be significantly different for LDL cholesterolbetween the two treatment periods (Table 3).

Day to day variability in antioxidant status is reportedto be low in unsupplemented subjects [23,24]. In the

present study, following supplementation of tomatoproducts with extra-virgin olive oil a significant increasein the plasma antioxidant activity (FRAP) was observed(p 5 .004, pairedt-test on log-transformed data), but theFRAP activity tended to decrease when tomato productswere fed with sunflower oil (Table 2). It is known that adiet rich in monounsaturated fatty acids inhibits [6] andpolyunsaturated fatty acids increases [13] the oxidationof LDL. The results of the present study suggest that thecomposition of oil used for cooking tomato products mayaffect the antioxidant activity of plasma. Other antioxi-dant carotenoids, i.e.,b-carotene, lutein, andb-cryptox-anthin were not changed during the study period (resultsnot shown). Also, since the plasma tocopherol levelswere not changed during the supplementation period, it isunlikely that the changes in antioxidant activity seenwere related to the tocopherol present in the oils. How-ever, extra-virgin olive oil also contains high amounts ofphenolic derivatives [25]. Since plasma phenolic com-pounds were not measured, it is difficult to assesswhether changes in plasma antioxidant activity could berelated to the phenolic acid constituents of the olive oil.

The feeding of tomato products with olive oil for 1week was followed by a 3 week washout period and afurther supplementation with sunflower oil and tomatoproducts. The plasma lycopene returned to the baselineafter the washout period but the plasma antioxidant ac-tivity remained significantly higher than the baseline(p 5 .01, Table 1). It has been reported that oil supple-mentation can have prolonged effects [26] and effects ofinduced fatty acid modifications remain for a long timein the biological systems [27,28]. Unfortunately it was

Table 2. Plasma Lycopene, FRAP, and Lipid Levels Before and After Supplementation with Tomato Products Plus Different Oils

Tomato products with olive oilMean6 SD

Tomato products with sunflower oilMean6 SD

Baseline Week 1 Week 4 Week 5

Dietary lycopenemg/d ,5 466 10 ,3 466 12

Plasma lycopenemmol/l 0.666 0.26a 1.206 0.20b 0.676 0.27a 1.146 0.35b

Plasma a-tocopherolmmol/l 19.06 3.6ab 19.46 3.8a 19.56 5.0ab 21.06 4b

Triglyceridesmmol/l 0.886 0.13a 0.686 0.27b 0.976 0.33ab 0.926 0.18a

Total cholesterolmmol/l 3.786 0.53 3.916 0.35 3.906 0.66 3.786 0.50

LDL cholesterolmmol/l 1.856 0.54 2.106 0.37 1.856 0.62 1.686 0.65

HDL cholesterolmmol/l 1.496 0.20 1.476 0.18 1.556 0.23 1.626 0.21

Total: HDL ratio 2.536 0.47 2.666 0.47 2.526 0.59 2.336 0.57FRAP

mmol/l 9306 150a 11186 184b 10496 186bc 10096 181ac

Effect of supplementation was measured using paired-t-test on log10 transformed data. Results shown are geometric means and standard deviation.Those not sharing a common superscriptabc were significantly different atp , .05.

1053Tomato and oil intervention in humans

not possible to do fatty acid and polyphenol analysis toconfirm whether the effects seen following the washoutperiod were related to the oil consumption. The con-sumption of tomato products with sunflower oil pro-duced no further increase in the FRAP and if anything,plasma antioxidant activity tended to decrease and wassignificantly lower than that measured following the ol-ive oil supplementation (p 5 .007, pairedt-test). Thechange in plasma FRAP (supplementation minus startvalues) was also significantly different when tomatoproducts were consumed with olive oil or sunflower oil(p 5 .002, Table 3).

In conclusion, the preliminary results indicate that theoil composition (olive oil vs. sunflower oil) does notaffect the bioavailabilty of lycopene from tomato prod-ucts. The oil composition however, may affect the anti-oxidant activity of the plasma. Further studies are neededto provide clarification on whether the antioxidant effectsseen in the present study were related to the olive oil orthe combination of olive oil and tomato products.

REFERENCES

[1] Kushi, L. E.; Lenart, E. B.; Willett, W. C. Health implications ofMediterranean diets in the light of contemporary knowledge. I.Plant foods and dairy products.Am. J. Clin. Nutr.61(Suppl.6):1407S–1415S; 1995.

[2] Kushi, L. E.; Lenart, E. B.; Willett, W. C. Health implications ofMediterranean diets in the light of contemporary knowledge. II.Meat, wine, fats and oils.Am. J. Clin. Nutr.61(Suppl. 6):1416S–1427S; 1995.

[3] Wise, J. A.; Morin, R. J.; Sanders, R.; Blum, K. Changes inplasma carotenoid, alpha-tocopherol and lipid peroxide levels inresponse to supplementation with concentrated fruit and vegetableextracts: a pilot study.Curr. Ther. Res. Clin. Exp.57:445–461;1996.

[4] Hinninger, I.; Chopra, M.; Thurnham, D. I.; Laporte, F.; Richard,M.-J.; Favier, A.; Roussel, A.-M. Effect of increased fruit andvegetable intake on the susceptibility of lipoprotein to oxidationin smokers.Eur. J. Clin. Nutr.51:601–606; 1997.

[5] Nigdikar, S. V.; Williams, N.; Griffin, B. A.; Howard, A. N.Consumption of red wine polyphenols reduces the susceptibilityof low-density lipoproteins to oxidation in vivo.Am. J. Clin. Nutr.68:258–265; 1998.

[6] Aviram, M.; Eias, J. K. Dietary olive oil reduces low-density

lipoprotein uptake by macrophages and decreases the susceptibil-ity of the lipoprotein to undergo lipid peroxidation.Ann. Nutr.Metab.37:75–84; 1993.

[7] Dimitrov, N. V.; Meyer, C.; Ullrey, D. E.; Chenoweth, W.;Michelakis, A.; Malone, W.; Boone, C.; Fink, G. Bioavailabilityof b-carotene in humans.Am. J. Clin. Nutr.48:298–304; 1988.

[8] Sies, H.; Stahl, W. Lycopene: antioxidant and biological effectsand its bioavailability in the human.Proc. Soc. Exp. Biol. Med.218:121–124; 1998.

[9] Clinton, S. K. Lycopene: chemistry, biology and implications forhuman health and disease.Nutr. Rev.56:35–51; 1998.

[10] Agarwal, S.; Venketeshwer, R. Tomato lycopene and low-densitylipoprotein oxidation: a human dietary intervention study.Lipids33:981–984; 1998.

[11] Woollett, L. A.; Spady, D. K.; Dietschy, J. M. Saturated andunsaturated fatty acids independently regulate low-density li-poprotein receptor activity and production rate.J. Lipid Res.33:77–88; 1992.

[12] Reaven, P.; Parthasarathy, S.; Grasse, B.; Miller, E.; Steinberg,D.; Witztum, J. L. Effect of oleate-rich and linoleate-rich diets onthe susceptibility of low-density lipoproteins to oxidative modi-fication in mildly hypercholesterolemic subjects.J. Clin. Invest.91:668–676; 1993.

[13] Louheranta, A. M.; Porkkala-Sarataho, E. K.; Nyyssonen, M. K.;Salonene, R. T.; Salonen, J. M. Linoleic acid intake and suscep-tibility of very low-density lipoproteins to oxidation in man.Am. J. Clin. Nutr.63:698–703; 1996.

[14] van Golde, P. H. M.; Sloots, L. M.; Vermeulen, W. P.; Wielders,J. P. M.; Hart, H. C.; Bouma, B. N.; van de Wiel, A. The role ofalcohol in the anti low-density lipoprotein oxidation activity ofred wine.Atherosclerosis147:365–370; 1999.

[15] Riso, P.; Pinder, A.; Santangelo, A.; Porrini, M. Does tomatoconsumption effectively increase the resistance of lymphocyteDNA to oxidative damage?Am. J. Clin. Nutr.69:712–718; 1999.

[16] Hart, D. J.; Scott, K. J. Development and evaluation of an HPLCmethod for the analysis of carotenes in vegetables and fruits in theUK. Food Chem.54:101–111; 1995.

[17] Increased fruit and vegetable consumption within the EC: poten-tial health benefits. Agro-Industry including Fisheries Programme2-CT93-0888 (Directorate-General XII, Science, Research andDevelopment) Report; 1997.

[18] Thurnham, D. I.; Smith, E.; Flora, P. S. Concurrent liquid-chro-matographic assay of retinol, alpha-tocopherol, beta-carotene,alpha-carotene, lycopene and beta-cryptoxanthin in plasma withtocopherol acetate as internal standard.Clin. Chem.34:377–381;1988.

[19] Benzie, I. F.; Strain, J. J. The ferric reducing ability of plasma(FRAP) as a measure of “Antioxidant Power”: the FRAP Assay.Anal. Biochem.239:70–76; 1996.

[20] Chopra, M.; McLoone, U.; O’Neill, M.; Williams, N.; Thurnham,D. I. Fruit and vegetable supplementation—Effect on ex vivoLDL oxidation in humans. In: Kumpulaine, J. T.; Salonen, J. T.,

Table 3. Change in Plasma Lycopene, Lipids, and FRAP Following the Two Treatment Periods

Change following consumption oftomato products with

Pairedt-testp 5Olive oil Sunflower oil

Plasma lycopene (mmol/l) 0.5336 0.15 0.4716 0.20 nsPlasma a-tocopherol (mmol/l) 0.6406 2.06 1.386 2.58 nsPlasma FRAP (mmol/l) 1906 101 29.616 99 .002Serum total cholesterol (mmol/l) 0.1176 0.31 20.1376 0.43 .06Serum LDL cholesterol (mmol/l) 0.2226 0.29 20.1756 0.40 .002Serum TG (mmol/l) 20.1736 0.16 20.0716 0.23 ns

The change in variables obtained during the supplementation of tomato products with olive oil was compared with that obtained during treatmentwith tomato products and sunflower oil using pairedt-test. Change is the difference between the treatment and respective start values.p values areshown where they were less than .1.

1054 A. LEE et al.

eds.Natural antixodant and food quality in atherosclerosis andcancer prevention.Cambridge, UK: The Royal Society of Chem-istry; 1996:151–155.

[21] Mensink, R. P.; Katan, M. B. Effect of dietary fatty acids onserum lipids and lipoproteins—a meta-analysis of 27 trials.Arte-rioscler. Thromb.12:911–919; 1992.

[22] Carmena, R.; Ascaso, J. F.; Camejo, G.; Varela, G.; Hurt-Camejo,E.; Ordovas, J. M. Effect of olive and sunflower oils on lowdensity lipoprotein level, composition, size, oxidation and inter-action with areterial proteoglycans.Atherosclerosis125:243–255;1996.

[23] van de Vijver, L. P. L.; van Duyvenvoorde, W.; Buytenhek, R.;van der Laarse, A.; Kardinaal, A. F. M.; van den Berg, H.;Princen, H. M. G. Seasonal variation in low density lipoproteinoxidation and antioxidant status.Free. Radic. Res.27:89–96;1997.

[24] Chopra, M.; Kavanagh, S.; Fitzsimons, P.; Thurnham, D. I. Vari-ations in the serum cholesterol, ferric reducing ability of plasma

and LDL lagphase in unsupplemented human subjects.Proc.Nutr. Soc.59:21A (abstr.); 2000.

[25] Litridou, M.; Linssen, J.; Schols, H.; Bergmans, M.; Pstrhumus,M.; Tsimdou, M.; Boskou, D. Phenolic compounds in virgin oliveoils: fractionation by solid phase extraction and antioxidant ac-tivity assessment.J. Sci. Food Agric.74:169–174; 1997.

[26] Brown, A. J.; Pang, E.; Roberts, D. C. Persistent changes in thefatty acid composition of erythrocyte membranes after moderateintake of n-3 polyunsaturated fatty acids: study design implica-tions.Am. J. Clin. Nutr.54:668–673; 1991.

[27] Prisco, D.; Filippini, M.; Francalanci, I.; Paniccia, R.; Genisini,G. F.; Abbate, K.; Neri Serneri, G. G. Effect of n-3 polyunsatu-rated fatty acid intake on phospholipid fatty acid composition inplasma and erythrocytes.Am. J. Clin. Nutr.63:925–932; 1996.

[28] Vicario, I. M.; Malkova, D.; Lund, E. K.; Johnson, I. T. Olive oilsupplementation in healthy adults: effects in cell membrane fattyacid composition and platelet function.Ann. Nutr. Metab.42:160–169; 1998.

1055Tomato and oil intervention in humans