Association of mate tea (Ilex paraguariensis) intake and dietary intervention and effects on oxidative stress biomarkers of dyslipidemic subjects

lable at ScienceDirect

Nutrition 28 (2012) 657–664

Contents lists avai

Nutrition

journal homepage: www.nutr i t ionjrnl .com

Applied nutritional investigation

Association of mate tea (Ilex paraguariensis) intake and dietary interventionand effects on oxidative stress biomarkers of dyslipidemic subjects

Brunna Cristina Bremer Boaventura M.Sc. a, Patr�ıcia Faria Di Pietro Ph.D. a,*, Aliny Stefanuto M.Sc. a,Graziela Alessandra Klein M.Sc. a, Elayne Cristina de Morais M.Sc. b, Fernanda de Andrade M.Sc. b,Elisabeth Wazlawik Ph.D. a, Edson Luiz da Silva Ph.D. a,b

aNutrition Postgraduate Program, Department of Nutrition, Health Sciences Center, Federal University of Santa Catarina, Florian�opolis, Santa Catarina, Brazilb Pharmacy Postgraduate Program, Department of Clinical Analyses, Health Sciences Center, Federal University of Santa Catarina, Florian�opolis, Santa Catarina, Brazil

a r t i c l e i n f o

Article history:Received 16 August 2011Accepted 28 October 2011

Keywords:Mate teaAntioxidant potentialOxidative stressDyslipidemiaPhenolic compoundsDietary interventionHumans

This study was supported by the Le~ao J�unior Co.Post-Graduation Program of Nutrition of the Federalrina, and a scholarship from the Brazilian FederalEvaluation of Graduate Education (CAPES).* Corresponding author. Tel.: þ55-48-3721-8014; fa

E-mail address: [email protected] (P. F. Di P

0899-9007/$ - see front matter � 2012 Elsevier Inc. Adoi:10.1016/j.nut.2011.10.017

a b s t r a c t

Objective: To evaluate the effect of long-term ingestion of mate tea, with or without dietaryintervention, on the markers of oxidative stress in dyslipidemic individuals.Methods: Seventy-four dyslipidemic volunteers participated in this randomized clinical trial.Subjects were divided into three treatment groups: mate tea (MT), dietary intervention (DI), andmate tea with dietary intervention (MD). Biochemical and dietary variables were assessed at thebeginning of the study (baseline) and after 20, 40, 60, and 90 d of treatment. Participants in the MTand MD groups consumed 1 L/d of mate tea. Those in the DI and MD groups were instructed toincrease their intake of fruit, legumes and vegetables and decrease their consumption of foods richin cholesterol and saturated and trans-fatty acids. Biomarkers of oxidative stress such as antioxi-dant capacity of serum (ferric reducing antioxidant potential assay), uric acid, reduced glutathione,paraoxonase-1 enzyme, lipid hydroperoxide (LOOH), and protein carbonyl were analyzed.Results: Participants in the DI group showed a significant decrease in total fat and saturated fattyacid intakes. Those in the DI and MD groups presented a significant increase in vitamin Cconsumption. For all groups, there was a significant increase in ferric reducing antioxidantpotential and reduced glutathione concentrations but no significant changes in LOOH, proteincarbonyl, and paraoxonase-1 values. The reduced glutathione concentration was positively corre-lated with the consumption of monounsaturated fatty acids, fiber, and vitamin C, whereas levels ofLOOH were inversely correlated with intakes of vitamin C and fiber. In addition, LOOH correlatedpositively with low-density lipoprotein cholesterol and inversely with high-density lipoproteincholesterol, which had a positive association with paraoxonase-1.Conclusion: The ingestion of mate tea independently of the dietary intervention increased plasmaand blood antioxidant protection in patients with dyslipidemia.

� 2012 Elsevier Inc. All rights reserved.

Introduction

Cellular and tissue damage induced by oxidative stress arerelated to the etiology of cardiovascular diseases (CVD),which areresponsible for approximately 17 million deaths annuallyworldwide [1,2]. Oxidative stress is a causal factor linking

(Curitiba, PR, Brazil), theUniversity of Santa Cata-Agency for Support and

x: þ55-48-3721-9542.ietro).

ll rights reserved.

the pathogenic processes of dyslipidemia and atherosclerosisbecause of its important role in low-density lipoprotein (LDL)oxidation in the early stages of atherosclerosis [3].

In addition to oxidized LDL, it has been shown that dyslipi-demic individuals and/or patients with CVD have an importantimbalance of various biomarkers of oxidative stress. For example,Serdar et al. [4] reported increased levels of lipid and proteinoxidation in patients with CVD. Moreover, the plasma levels ofreduced glutathione (GSH) are decreased in individuals withvascular disease [5]. Similarly, patients with hypercholesterol-emia and/or diabetes mellitus present with decreased activity ofthe antioxidant enzyme paraoxonase-1 (PON1) [6].

B. C. B. Boaventura et al. / Nutrition 28 (2012) 657–664658

Epidemiologic studies have shown an inverse associationbetween an intake of foods rich in antioxidant substances, suchas vitamins A, C, and E and phenolic compounds, and CVDmortality [7,8]. Therefore, there is growing interest in theprevention of CVD, with a focus on antioxidant intake fromdietary sources [9].

Mate tea, also known as toasted yerba mate, is an herbalinfusion made from dried leaves of the Ilex paraguariensis [10].This beverage is widely consumed by large numbers of people inSouth American countries, in particular in Uruguay, Argentina,Brazil, and Paraguay [11]. Given increased consumption, commer-cial availability, and reported potential health benefits,mate tea isbecoming popular in Europe and North America [10]. Yerba matecontains several classes of chemical constituents, mainly caffeoylderivatives (caffeic acid, chlorogenic acid, 3,4-dicaffeoylquinicacid, 3,5-dicaffeoylquinic acid, and 4,5-dicaffeoylquinic acid),catechins, amino acids, flavonoids (quercetin, kaempferol, andrutin), minerals (phosphate, iron, and calcium), and the vitaminsC, B1, and B2 [12–14].

Various studies have reported the high antioxidantpotential of aqueous extracts of yerba mate, including aninhibitory ability against plasma and LDL oxidation [15–19].Menini et al. [20] demonstrated that an acute ingestion ofyerba mate infusion inhibited high-density lipoprotein (HDL)oxidation induced by peroxyl radicals and preserved apoli-poproteins A-I and PON1 enzyme activity. Furthermore,Matsumoto et al. [21] showed that acute and 1-wkconsumptions of roasted mate tea by healthy womenpromoted a decrease in plasma lipid oxidation and increasesin plasma antioxidant capacity and gene expression of theantioxidant enzymes catalase, superoxide dismutase, andglutathione peroxidase.

Therefore, the purpose of this study was to verify whether thelong-term consumption of mate tea (90 d) would improve theantioxidant status and/or decrease the oxidative parameters ofdyslipidemic individuals who are at greater risk to developatherosclerosis. The effectiveness of the mate tea intake associ-ated with a qualitative dietary intervention based on therecommendation of increased intakes of fruit, legumes, andvegetables, which are rich in fiber and antioxidant compounds,and decreased intakes of foods containing cholesterol and satu-rated and trans-fatty acids, which are involved with dyslipide-mia, were also investigated.

Materials and methods

Subjects and study design

The present randomized intervention and clinical trial included dyslipidemic(male and female) volunteers older than 18 y. Dyslipidemia was characterizedaccording to the IV Brazilian Guidelines on Dyslipidemia and AtherosclerosisPrevention [22]: total cholesterol (TC) �200 mg/dL or LDL cholesterol �130mg/dL or HDL cholesterol �40 mg/dL for women or �50 mg/dL for men.

Exclusion criteriawere the use of hypolipidemic medicines, hepatic or kidneydisease, diabetesmellitus, cancer, thyroid disorders, angina, alcoholism, smoking,use of nutritional supplements, morbid obesity, and intense physical activity.

The baseline period consisted of 30 d of biochemical, anthropometric, anddietetic monitoring. After this period, individuals were randomly divided intothree different treatment groups: mate tea (MT; n¼ 23), dietary intervention (DI;n ¼ 26), or mate tea with the dietary intervention (MD; n ¼ 25). Volunteers wereevaluated according to the variables described earlier 20, 40, 60, and 90 d afterthe baseline period. Participants in all groups were evaluated and instructed bythe nutritionists. All volunteers were requested to not change their habituallifestyle. This information was monitored at every return visit.

This study was approved by ethics committee of human research at theFederal University of Santa Catarina (UFSC) and all participants gave written andinformed consent.

Anthropometric and dietetic procedures

Body weight was measured with a digital scale (Marte PP180, Santa Rita doSapucai, MG, Brazil) with a capacity of 180 kg and sensitivity of 0.1 kg. Bodyheight was measured with an anthropometer (Seca, Hamburg, Germany) witha maximum height of 220 cm and a 0.1-cm scale. Anthropometric measurementprocedures were in accordance with the techniques recommended by the WorldHealth Organization [23]. Body weight and height variables were used tocalculate the body mass index [23].

Participants in the MT and MD treatment groups received minced leaves ofcommercial roasted yerba mate and were instructed in preparing mate infusionswith boiling water at a proportion of 20 mg/mL. After a 10-min extraction, themixture should be filtered and consumed immediately with no sugar or anysugar-like substances. Volunteers were instructed to drink 330 mL of mate teathree times a day, totaling approximately 1 L, mainly with meals (breakfast,lunch, and dinner). The yerba mate was collected in Irati (PR, Brazil), anda specimen of the plant was identified as authentic I. paraguariensis at theBotanical Department of the UFSC. A sample of I. paraguariensiswas deposited inthe herbarium of the Botanical Department of the UFSC under the code FLOR37066. The chemical characterization by high-performance liquid chromatog-raphy of the yerba mate used in this study has been published previously [14].

Volunteers from the DI and MD groups received a dietary intervention basedon recommendations according to the IV Brazilian Guidelines on Dyslipidemiaand Atherosclerosis Prevention [22]. The dietary intervention consisted ofnutritional counseling to increase the consumption of fruits, legumes, andvegetables, which are rich in fiber and antioxidants. The volunteers had theoption of eating fruits and vegetables according to their food habits and prefer-ences. However, they were encouraged to consume seasonal fruits, which aremore nutritious, tasty, and accessible. Volunteers in the DI and MD groups werealso instructed to decrease their consumption of foods rich in cholesterol andsaturated and trans-fatty acids (meat fat, industrial products, butter) and toconsume low-fat versions of milk, cheese, and yogurt.

The nutritional composition of the normal dietary intake of all participantsduring each stage of the study was calculated from 3-d dietary records (1d during the weekend and 2 d during the week). All food records were checkedindividually by a nutritionist during every return visit. The food compositionwasdetermined using Avanutri 3.1.4 (Rio de Janeiro, RJ, Brazil). Dietary variablesconsidered in this analysis were energy, carbohydrates, proteins, total fat, satu-rated fatty acids (SFAs), polyunsaturated fatty acids, monounsaturated fatty acids(MUFAs), dietetic cholesterol, fiber, and vitamins A, C, and E.

Biochemical analysis

Blood samples were collected, after a 12- to 14-h fast, by vein puncture witha vacuum system (Vacutainer, BD, S~ao Paulo, SP, Brazil) into tubes with or withoutethylenediaminetetraacetic acid. All tubes were immediately centrifuged (1000� g, 15 min) to obtain serum or plasma. All biochemical assays were performedimmediately after the sample collection, except PON1 enzyme activity andprotein carbonyl, which were determined after sample storage at �70�C for nolonger than 30 d.

Serum lipid profile

The serum lipid profile was determined using commercial kits on automatedequipment (Dimension RxL; Siemens Healthcare Diagnostics, Inc., Deerfield, IL,USA). For the measurement of TC, all cholesterol esters present in 3 mL of serumwere hydrolyzed into free cholesterol and fatty acids by cholesterol esterase, ina total volume of 358 mL. In the presence of oxygen, the free cholesterol producedand the preexisting free cholesterol were oxidized in a reaction catalyzed bycholesterol oxidase to form cholest-4-ene-3-one and hydrogen peroxide (H2O2).The H2O2 thus formed is used to oxidize N,N-diethyl-aniline-HCl/4-aminoantipyrine to produce a chromophore that absorbs at 540 nm. The absor-bance from the oxidized N,N-diethyl-aniline-HCl/4-aminoantipyrine is directlyproportional to the TC concentration and was measured using a polychromatic(452-, 540-, 700-nm) endpoint technique. The analytical between-day precision,expressed as a coefficient of variation (CV) and determined from data obtained intriplicate over 20 separate days, was 2.14% using a commercial control.

The triacylglycerols present in 4 mL of serumwere hydrolyzed by lipoproteinlipase to produce free glycerol and fatty acids. The glycerol participates in a seriesof coupled enzymatic reactions. Glycerol kinase catalyzes the phosphorylation ofglycerol by adenosine-5-triphosphate to glycerol-3-phosphate. Glycerol-3-phosphate oxidase oxidizes glycerol-3-phosphate to dihydroxyacetonephosphate and H2O2. The catalytic action of peroxidase forms quinone-iminefrom H2O2, 4-aminoantipyrine, and 4-chlorophenol. The absorbance of quinone-imine is directly proportional to the total amount of glycerol and its precursors inthe sample and is measured using a bichromatic (510-, 700-nm) endpointtechnique. The CV was 3.36% using a commercial control.

B. C. B. Boaventura et al. / Nutrition 28 (2012) 657–664 659

Direct LDL cholesterol was measured using the homogeneous LDL cholesterolassay (Siemens ALDL). Briefly, this method involves a two-reagent format. Thefirst reagent is a detergent that solubilizes only non-LDL particles present in 3 mLof serum. The solubilized cholesterol is consumed in a non-color-forming reac-tion by cholesterol esterase and cholesterol oxidase. The second detergent is thenadded, which solubilizes the remaining LDL particles and links the enzymaticoxidation of LDL cholesterol by cholesterol esterase and cholesterol oxidase,forming cholestan-4-one and H2O2. The enzymatic action of peroxidase on H2O2

produces color in the presence of N,N-bis(4-sulfobutyl)-m-toluidine, disodiumsalt, and 4-aminoantipyrine. The colored product is measured using a bichro-matic (540-, 570-nm) endpoint technique. The color produced is directlyproportional to the amount of LDL cholesterol present in the sample. Between-day precision on the serum control demonstrated a CV of 2.18%.

The A-HDL homogeneous method (Dimension system, Siemens HealthcareDiagnostics) was used to directly measure HDL cholesterol levels in serum. Inbrief, the method has a two-reagent format and is based on accelerating thereaction of cholesterol oxidase with non-HDL unesterified cholesterol and dis-solving HDL selectively using a specific detergent. For the first reagent, non-HDLunesterified cholesterol is subjected to a cholesterol oxidase reaction and theH2O2 generated is consumed bya peroxidase reactionwith disodium salt, yieldinga colorless product. The second reagent consists of a detergent capable of solu-bilizing HDL specifically and a coupler for cholesterol esterase, disodium salt, and4-aminoantipyrine to develop color for the quantitative determination of HDLcholesterol using a bichromatic (600-, 700-nm) endpoint technique. The CVobtained in triplicate over 20 separate dayswas 2.85% using a commercial control.

Non-HDL cholesterol level was estimated by the difference between TC andHDL cholesterol.

The additional determination of biochemical (urea, creatinine, uric acid, andactivities of the enzymes alanine aminotransferase, aspartate aminotransferase,alkaline phosphatase, and g-glutamyl transferase) and hematologic (completeblood count) parameters were evaluated tomonitor the potential toxicity of matetea using commercial kits in automated equipment (Dimension RxL, SiemensHealthcare Diagnostics; and Sysmex Xe-2100D, Kobe, Japan) according to themanufacturer’s instructions.

Oxidative stress biomarkers

Seruj antioxidant capacity was determined according to the ferric reducingantioxidant potential (FRAP) assay as described by Benzie and Strain [24]. In thisprocedure, the antioxidants present in the serum were evaluated as reducers ofFe3þ to Fe2þ, which is chelated by 2,4,6-tris(2-pyridyl)-s-triazine (TPTZ) to forma Fe2þ–TPTZ complex with maximum absorbance at 593 nm. Ten microliters ofserum was mixed with 1 mL of reagent containing 1.7 mM FeCl3 and 0.8 mMTPTZ, prepared in 300 mM sodium acetate (pH 3.6). The samples were incubatedfor 15 min at 37�C and the absorbance was read at 593 nm (Spectrum SP-2000spectrophotometer; Spectrum Instruments, Shanghai, China). The 6-hydroxy-2,5,7,8-tetramethylchroman-2-carboxylic acid (trolox) was used as a standardand the FRAP values were expressed as trolox equivalents in micromoles per liter.The interassay CV using slopes of the trolox standard curves was 5.4% (n ¼ 20)and the average CV for duplicate variation was 3.8%.

Blood GSH was assessed according to the method proposed by Beutler et al.[25]. An aliquot of the total ethylenediaminetetraacetic acid-treated blood washemolyzed with cold water and the proteins were precipitated by the addition of30% trichloroacetic acid. Aliquots of 50 mL of the hemolyzed sample and 50 mL of10 mM 5,50-dithiobis-2-nitrobenzoic acid were mixed in tubes containing 0.8 mLof 200 mM phosphate buffer, pH 8.0. After 3 min, the absorbance of the thiolateanion was measured at 412 nm (Bioplus BIO 2000, Barueri, SP, Brazil). GSH wasused as a standard and the results were expressed in micromoles per liter. Theinterassay CV using 50 mM GSH was 4.8% (n ¼ 20) and the CV for the duplicatevariation was 3.9%.

The arylesterase activity of PON1 was analyzed as previously described [26]using phenyl acetate as a substrate. The reaction was started by adding 20 mL ofserum diluted by a factor of 20 in 2.0 mL of 20 mmol/L Tris HCl buffer, pH 8.0,containing CaCl2 1 mmol/L and phenyl acetate 1 mmol/L. The kinetic of phenolformation at 25�C was monitored for 5 min at 270 nm (Spectrum SP-2000).Values were expressed in arylesterase units corresponding to 1 mmol of phenylacetate hydrolyzed per milliliter of serum (units per milliliter). The interassay CVfor the pooled human serum obtained in duplicate over 20 separate days was8.9%, with a 6.5% variation between duplicates.

Lipid hydroperoxides (LOOH) present in the serumwere evaluated accordingto the ferric oxidation of the xylenol orange assay [27]. The method is based onthe fast oxidation of Fe2þ to Fe3þ in acid medium mediated by lipid peroxides. Inthe presence of xylenol orange, Fe3þ forms a complex (Fe3þ–xylenol orange),which is measured spectrophotometrically at 560 nm. Aliquots of serum, induplicate, were mixed with the FOX2 reagent (1.4 mL), containing 250 mMH2SO4, 4.4 mM butylated hydroxytoluene, 1 mM xylenol orange, and 2.5 mMiron, and ammonium sulfate in methanol. Ten microliters of 10 mM triphenyl-phosphine in methanol was added to two additional microtubes with serum to

reduce the LOOH before the addition of the FOX2 reagent, thereby generatinga blank sample. Subsequently, the mixtures were kept at room temperature for30 min, the tubes were centrifuged (1000 � g, 5 min), and the absorbance wasread at 560 nm (Spectrum SP-2000). The LOOH level in the plasmawas calculatedusing themean absorbance difference between the duplicates of the test samplesand the blank samples. N H2O2 curve was used to calculate the LOOH concen-tration and the values were expressed in micromoles per liter. The interassay CV,calculated by the measurement of 0.1 mM H2O2 on different days, was 9.5%(n ¼ 20). The average CV for the duplicate variation was 5.8%.

Plasma levels of protein carbonyl were determined spectrophotometricallyby the method proposed by Levine et al. [28], using 2,4-dinitrophenylhydrazine(2,4-DNPH). After precipitation of 0.1 mL of plasma proteins with 0.1 mL of 20%trichloroacetic acid, the pellet was resuspended in 0.6 mL of 10 mM DNPH in 2 NHCl. A blank reactionwas prepared by adding 0.6 mL of 2 N HCl without DNPH tothe plasma. All tubes were kept at room temperature for 1 h in the dark andvortex-mixed every 15 min. Afterward, 0.6 mL of 20% trichloroacetic acid wasadded, and after a 10-min incubation the tubes were centrifuged at 12 000� g for15 min at 4�C, and the pellets were washed three times with 0.8 mL of ethanol-ethyl acetate (1:1, v/v) to remove the free DNPH and lipid contaminants. The finalpellet was dissolved in 0.9 mL of 6 M guanidine hydrochloride, which wasprepared in 20 mM KH2PO4, and kept at 37�C for 1 h under continuous agitation.The tubes were centrifuged for 15 min at 12 000 � g and the carbonyl contentwas measured by reading the absorbance at 370 nm (Spectrum SP-2000). Themolar coefficient absorption (ε) of 22 000 L $ mol�1 $ cm�1 was used. The totalprotein content, measured in the blank reaction at 280 nm, was calculated usingbovine albumin as a standard, and the carbonyl levels were expressed in nano-moles per milligram of protein.

Statistical analysis

Continuous variables are presented as mean and standard error of the mean,and categorical variables are presented as absolute frequency. The Kolmogorov–Smirnov test was applied to verify the normality of the data. The differencesbetween treatment groups were verified by analysis of variance. The Student’s ttest or Wilcoxon test was used to evaluate the treatment effects within eachgroup for data with a normal or non-Gaussian distribution, respectively. Theassociation between changes in the biochemical parameters and changes of thefood intake variables and bodyweightwas evaluated by the Pearson or Spearmancorrelation. The Pearson correlation was applied to verify the associationbetween changes in the biomarkers of oxidative stress and changes of the foodintake. In addition, the association between changes in the lipid parameters andchanges in the oxidative stress biomarkers was evaluated by the Pearsoncorrelation. A level lower than 5% was considered statistically significant.SigmaStat 3.5 (SPSS Inc., Chigaco, IL, USA) was used for all statistical analyses.

Results

One hundred thirty-five individuals started the study. Therewere 38 individuals who gave up the treatment for personalreasons; 21 volunteers were excluded because they had baselinelipid profile values below those of the inclusion criteria; and 2participants had angina. Thus, 74 volunteers (17 men and 57women, average age 48.5 � 11.6 y) participated in the study.

In the baseline clinical and biodemographic characteristics ofthe participants, there was no significant difference betweentreatment groups in the number of individuals, age, weight, bodymass index, abdominal circumference, blood pressure, history ofcoronary artery disease, physical activity, or smoking habit(Table 1). In addition, no significant differences were found fordietary variables, serum lipid profile, and oxidative stressbiomarkers of participants in the different groups at the baselineperiod (Tables 2 and 3). The long-term consumption of mate teadid not show adverse effects in the participants (data not shown).

In relation to food consumption, there were no significantchanges in the dietary profile of volunteers in the MT groupthroughout the study (Table 2). However, as expected, theparticipants in the DI group showed a significant decrease in theingestion of total fat (24.5%) and SFAs (28.1%) after 90 d of inter-vention. Moreover, a significant increase in vitamin C intake wasverified after 40 d (90.3%), 60 d (119.7%; data not shown), and 90d (97.9%)of treatment comparedwith thebaselinevalues (Table2).

Table 1Baseline clinical and biodemographic characteristics of study participants

Characteristics MT DI MD

Women/men 15/8 21/5 21/4Age (y) 51.8 � 2.6 47.8 � 2.3 46.2 � 2.0Weight (kg) 71.9 � 2.8 71.2 � 2.6 69.8 � 3.0BMI (kg/m2) 27.5 � 0.8 27.1 � 0.7 27.4 � 0.8AC (cm) 95.2 � 10.1 91.3 � 8.3 93.4 � 9.9SBP (mmHg) 115.5 � 13.7 112.0 � 13.9 119.1 � 21.0DBP (mmHg) 72.9 � 7.7 73.1 � 7.9 75.7 � 8.6Family history of CAD 20 (86.9%) 20 (76.9%) 15 (60.0%)Physical activity (h/wk) 1.9 � 0.4 2.1 � 0.4 2.0 � 0.4Smoking 2 (8.7%) 0 (0.0%) 1 (4.0%)

AC, abdominal circumference; BMI, body mass index; CAD, coronary arterydisease; DBP, diastolic blood pressure; DI, dietary intervention group; MD, matetea plus dietary intervention group; MT, mate tea group; SBP, systolic bloodpressureResults are expressed as mean � SEM or number (percentage). No significantdifference was observed among treatment groups at baseline

B. C. B. Boaventura et al. / Nutrition 28 (2012) 657–664660

Volunteers in the MD group demonstrated a significantdecrease in SFA intake after 20 d (16.1%), 40 d (25.7%), and 60d (24.4%) of intervention (P < 0.05; data not shown). Further-more, an important decrease in the consumption of total fat(13.7%), dietary cholesterol (17.2%), and SFAs (20.7%) was detec-ted after 90 d of intervention, but these variations did not reachstatistical significance (Table 2). In addition, participants in theMD group presented a significant increase in vitamin C ingestionafter 40 d (50.1%), 60 d (58.8%; data not shown), and 90 d (54.5%)of treatment (P < 0.05; Table 2).

Regardless of the differences detected within each treatmentgroup, there were no significant intergroup variations for theconsumption of calories, carbohydrates, total fat, fiber, protein,cholesterol, polyunsaturated fatty acids, MUFAs, and vitamins A,C, and E by dyslipidemic subjects in the three groups throughoutthe study (Table 2).

Oxidative stress parameters indicated a significant increase inserum antioxidant capacity (based on FRAP levels) of partici-pants in the MT group after 20, 40, 60, and 90 d of treatmentcompared with baseline levels (P < 0.05; Table 3 and Fig. 1A).Dyslipidemic individuals in the DI group showed a significantincrease in FRAP levels by 6.7% after 20 d of treatment only (P <

0.05; Fig. 1A). Compared with the baseline values, significantincreases in the FRAP levels of 8.6%, 11.6%, and 8.1% were

Table 2Dietary intake of dyslipidemic individuals at baseline and after 90 days of interventio

MT DI

Baseline After 90 days Baseline

Energy (kcal) 1940 � 97 1894 � 139 2018 �Protein (g) 70.1 � 4.8 70.3 � 8.7 77.5 �Carbohydrate (g) 253.4 � 13.7 223.5 � 16.7 271.6 �Fiber (g) 11.4 � 0.7 11.4 � 1.5 13.6 �Total fat (g) 61.8 � 3.9 64.8 � 7.8 66.4 �Cholesterol (mg) 205.2 � 19.2 192.9 � 28.5 184.5 �SFA (g) 18.1 � 1.5 18.8 � 2.3 17.1 �PUFA (g) 12.8 � 1.1 14.4 � 2.8 10.8 �MUFA (g) 15.0 � 1.3 18.3 � 1.6 14.7 �Vitamin A (mg) 567.4 � 56.3 531.3 � 81.5 684.5 �Vitamin C (mg) 75.6 � 10.8 73.6 � 13.4 80.9 �Vitamin E (mg) 17.3 � 3.5 18.2 � 1.6 13.6 �

DI, dietary intervention group; MD, mate tea plus dietary intervention group; MT, macids; SFA, saturated fatty acidsResults are expressed as mean � SEM

* Significant difference (P < 0.05) compared with baseline.

observed after 20, 40, and 90 d, respectively, for participants inthe MD group (Fig. 1A). All three treatments studied promoteda significant increase in the blood GSH levels of participants, withmaximum increases of 21.7% after 90 d (MTgroup),15.6% after 60d (DI group), and 13.4% after 90 d (MD group; P< 0.05; Fig.1B). Incontrast, there were no significant changes in the levels of uricacid, PON1, LOOH, and protein carbonyl compared with baselinevalues after 20, 40, 60, and 90 d compared with the baselinelevels of participants in the three different groups (Table 3).

For the lipid profile, only participants in the MT group pre-sented a significant decrease in the LDL cholesterol fraction after90 d of intervention (Table 3). Individuals in all treatment groupsshowed no significant changes in TC, HDL cholesterol, HDLcholesterol, non-HDL cholesterol, and triacylglycerols (Table 3).

Associations between dietary intake and oxidative stresswere examined to demonstrate a potential influence of nutrientson the levels of oxidative stress biomarkers, particularly forsubjects who received a dietary intervention only (Table 4). Fordyslipidemic individuals in the DI group, changes in the levels ofblood GSH were directly correlated to the consumption of fiber(r ¼ 0.43, P < 0.001), MUFAs (r ¼ 0.28, P ¼ 0.01), and vitamin C(r ¼ 0.25, P ¼ 0.04). Approximately 50% of participants showedan increased intake of these nutrients and had, concomitantly,increased GSH levels. In contrast, the LOOH concentration wasinversely associated with; ingestion of fiber (r ¼ �0.34,P ¼ 0.003) and vitamin C (r ¼ �0.25, P ¼ 0.04). Almost 37.5% ofindividuals in this group presented a concomitant increase offiber and vitamin C intakes and decreasing levels of LOOH(Table 4). For individuals in the MD group, increased consump-tion of vitamin C was found to be associated with lower LOOHconcentrations (r ¼ �0.25, P ¼ 0.04; results not shown).

The correlations between lipid parameters and oxidativestress profile were also examined to determine a potentialinfluence of lipids on levels of oxidative stress biomarkers(Fig. 2). The LOOH concentration had a positive association withLDL cholesterol levels for individuals in the MT group (r ¼ 0.337,P< 0.005; Fig. 2A) and for volunteers in the MD group (r¼ 0.241,P < 0.04; results not shown). In addition, for individuals in theMT group, the serum level of LOOH was inversely correlated tothe HDL cholesterol concentration (r ¼ �0.309, P ¼ 0.009;Fig. 2B). Moreover, PON1 activity was found to be positivelyassociated with HDL cholesterol levels in participants in the MDgroup (r ¼ 0.263, P ¼ 0.016; Fig. 2C).

n

MD

After 90 days Baseline After 90 days

105 1860 � 138 1862 � 86 1936 � 924.6 87.7 � 6.5 70.0 � 4.6 60.7 � 7.118.3 267.3 � 22.5 231.3 � 11.8 236.5 � 20.71.2 16.0 � 1.6 11.1 � 1.0 12.2 � 1.44.2 50.1 � 5.2* 62.1 � 4.2 53.6 � 3.319.2 179.1 � 22.0 209.2 � 17.1 173.2 � 19.41.4 12.3 � 1.4* 16.9 � 1.3 13.4 � 1.50.9 8.4 � 1.4 12.9 � 1.2 14.1 � 1.31.4 13.6 � 1.8 15.2 � 1.4 15.2 � 1.670.1 785.9 � 100.1 560.5 � 63.7 631.5 � 73.013.22 160.1 � 13.8* 81,7 � 10.6 132,24 � 15.5*

3.2 11.7 � 1.4 13.5 � 4.2 17.1 � 1.6

ate tea group; MUFA, monounsaturated fatty acids; PUFA, polyunsaturated fatty

Table 3Lipid profile and oxidative stress markers of dyslipidemic individuals at baseline and after 90 days of intervention

MT DI MD

Baseline After 90 days Baseline After 90 days Baseline After 90 days

CT (mg/dL) 233.1 � 5.8 234.2 � 6.0 232.2 � 8.0 233.5 � 7.4 231.4 � 8.0 234.4 � 6.2LDL-C (mg/dL) 160.2 � 5.7 150.1 � 4.8* 156.5 � 5.2 149.2 � 5.4 153.2 � 6.0 152.2 � 5.7HDL-C (mg/dL) 52.2 � 2.8 52.8 � 2.5 50.9 � 1.7 48.8 � 2.0 49.1 � 1.6 47.1 � 1.8Non-HDL-C (mg/dL) 187.1 � 4.8 187.5 � 5.0 183.7 � 4.8 183.0 � 4.6 184.9 � 4.4 192.2 � 4.9TG (mg/dL) 131.1 � 12.0 129.0 � 11.2 126.3 � 10.0 123.3 � 10.7 146.4 � 14.6 156.3 � 13.5FRAP (mmol/L) 588.1 � 18.4 630.2 � 19.0* 562.6 � 14.4 574.8 � 15.9 564.8 � 21.2 615.19 � 24.2*

Uric acid (mg/dL) 4.5 � 0.3 4.3 � 0.4 4.1 � 0.2 4.1 � 0.3 4.3 � 0.2 4.3 � 0.3GSH (mmol/L) 83.8 � 2.1 102.0 � 3.1* 84.9 � 2.4 98.2 � 2.7* 85.1 � 2.9 96.5 � 3.1*

LOOH (mmol/L) 8.8 � 0.6 8.9 � 0.1 9.3 � 0.6 9.5 � 0.1 8.8 � 0.5 9.4 � 1.1PC (nmol/mg) 0.9 � 0.05 0.9 � 0.06 0.8 � 0.04 0.9 � 0.05 0.8 � 0.03 0.9 � 0.04PON1 (U/mL) 91.8 � 4.7 90.4 � 8.5 93.9 � 5.5 99.8 � 2.7 94.5 � 3.3 96.6 � 5.2

DI, dietary intervention group; FRAP, ferric reducing antioxidant potential; GSH, reduced glutathione; HDL-C, high density lipoprotein cholesterol; LDL-C, low densitylipoprotein cholesterol; LOOH, lipid hydroperoxide; MD, mate tea plus dietary intervention group; MT, mate tea group; PC, protein carbonyl; PON1, enzymeparaoxonase-1; TC, total cholesterol; TG, triacylglycerolsResults are expressed as mean � SEM

* Significant difference (P < 0.05) compared with baseline.

B. C. B. Boaventura et al. / Nutrition 28 (2012) 657–664 661

Discussion

This study demonstrated that daily ingestion of 1 L of matetea, a beverage rich in phenolic antioxidants, for 90 d increasedthe antioxidant protection of plasma and whole blood, therebylowering oxidative stress in individuals at increased risk of CVD.

0

550

600

650

700

*

A

MT

**

* *

*

**

MDDI

Baseline 20 days 40 days 60 days 90 days

Antio

xida

nt C

apac

ity o

f Ser

um(µ

mol

/L)

010

80

90

100

110B

**

*** *

** * *

*

*

*

MDDI

Baseline 20 days 40 days 60 days 90 days

MT

GSH

(µm

ol/L

)

Fig. 1. Serum antioxidant capacity (A) (ferric reducing antioxidant potential assay)and (B) blood GSH of dyslipidemic individuals at baseline and after 20, 40, 60, and90 d of intervention. Results are expressed as mean � SEM. * Significant difference(P < 0.05) compared with baseline. DI, dietary intervention group; GSH, reducedglutathione; MD, mate tea with dietary intervention group; MT, mate tea group.

Prolonged ingestion of mate tea, with or without dietaryintervention, promoted a significant increase in serum antioxi-dant capacity throughout the study. Recently, we demonstratedthat acute ingestion of mate infusion significantly increasedserum antioxidant capacity in healthy subjects [19]. Matsumotoet al. [21] also reported increased plasma antioxidant capacity inhealthy subjects after ingestion of roasted mate tea for 1 wk. It isworth noting that the present results are consistent with those ofother studies that have evaluated the effect of beverages rich inantioxidants. For example, daily consumption of green teacaused a significant increase in the total antioxidant capacity ofplasma after 45 and 90 d [29]. Similar results were found whenacute ingestion of green tea by individuals with dyslipidemiawasinvestigated [30].

Plasma antioxidant capacity is a function of the individual andsynergistic effects of numerous components such as uric acid,phenolic compounds, ascorbic acid, tocopherols, bilirubin, andalbumin, among others, and uric acid is responsible for approx-imately 60% of the antioxidant capacity in the FRAP assay [24,31].Considering that the ingestion of mate tea did not cause signif-icant changes in the serum levels of uric acid, it is suggested thatthe increased levels of FRAP may be promoted by the constitu-ents of mate tea, possibly phenolic compounds. In addition, nosignificant association was found between the FRAP values anduric acid (r ¼ 0.21, P ¼ 0.07) or between FRAP and GSH (r ¼�0.03, P ¼ 0.77). In fact, Gungor et al. [32] demonstrated thatFRAP is not associated with thiol antioxidants such as GSH.

In this study, we also demonstrated that ingestion of matetea significantly increased the GSH blood concentration in dys-lipidemic subjects, with a maximum increase of 21.7% onaverage after 90 d of mate tea consumption. This increase wasslightly higher than that found for individuals who received onlya dietary intervention (15.6%) or nutritional counseling withmate tea (13.4%). The phenolic antioxidants in mate tea mayhave contributed to the preservation of the GSH concentrationby lowering its oxidation and even favoring an increase in itssynthesis, as suggested by Lahouel et al. [33]. It is noteworthythat in subjects receiving dietary intervention, the GSHconcentration was directly associated with a higher consump-tion of foods rich in vitamin C, fiber, and MUFAs [34,35]. GSH isan important antioxidant that prevents free radical damage,promotes detoxification by combining with chemicals, and actsas a substrate for the antioxidant enzyme glutathione peroxi-dase [36]. It is interesting to note that the ingestion of mate tea

Table 4Correlation between biomarkers of oxidative stress and dietary variables in the dietary intervention group

FRAP Uric acid GSH PON1 LOOH PC

r P r P r P r P r P r P

Energy (kcal) �0.15 0.19 �0.05 0.63 0.22 0.06 0.14 0.23 �0.17 0.14 0.09 0.43Protein (g) 0.01 0.96 �0.01 0.92 0.21 0.08 0.14 0.23 0.09 0.41 �0.13 0.25Carbohydrate (g) �0.23 0.06 �0.11 0.31 0.16 0.14 �0.01 0.92 �0.22 0.06 0.11 0.36Fiber (g) 0.05 0.67 �0.02 0.80 0.43 <0.001 �0.02 0.82 �0.34 0.003 0.17 0.13Total fat (g) �0.18 0.12 0.04 0.71 0.07 0.55 0.19 0.11 �0.06 0.58 0.11 0.35Cholesterol (mg) 0.03 0.74 0.12 0.27 �0.02 0.84 �0.05 0.68 0.04 0.69 �0.03 0.79SFA (g) �0.13 0.27 0.01 0.87 0.08 0.47 0.09 0.42 0.09 0.42 �0.07 0.54PUFA (g) 0.00 0.99 �0.09 0.46 0.13 0.27 0.11 0.37 �0.18 0.12 0.21 0.08MUFA (g) 0.06 0.60 0.01 0.95 0.28 0.01 0.00 0.99 �0.07 0.50 0.06 0.58Vitamin A (mg) 0.13 0.27 0.00 0.94 0.06 0.60 0.00 0.95 �0.19 0.10 0.01 0.89Vitamin C (mg) 0.10 0.40 0.04 0.68 0.25 0.04 0.13 0.27 �0.25 0.04 0.09 0.45Vitamin E (mg) �0.06 0.58 0.10 0.38 �0.16 0.15 0.17 0.13 �0.03 0.75 0.23 0.06

FRAP, ferric reducing antioxidant potential; GSH, reduced glutathione; LOOH, lipid hydroperoxide; MUFA, monounsaturated fatty acids; PC, protein carbonyl; PON1,enzyme paraoxonase-1; PUFA, polyunsaturated fatty acids; SFA, saturated fatty acidsResults were obtained from the Pearson correlation test

B. C. B. Boaventura et al. / Nutrition 28 (2012) 657–664662

for 1 wk increased the gene expression of glutathione peroxi-dase [21], and that low concentrations of glutathione peroxidasecan be considered a risk factor for the development of CVD [37].

The PON1 enzyme is associated with plasma HDL, promotesthe inhibition of LDL oxidation, and decreases oxidized lipids inatherosclerotic lesions [38]. Moreover, its activity is inverselyassociated with CVD in hypercholesterolemic and diabeticpatients [6]. The activity of PON1 may also be positively modu-lated by diets rich in antioxidant compounds [39,40] and nega-tively by diets containing oxidized fatty acids [41]. Menini et al.[20] reported that an acute ingestion of Ilex paraguariensisinfusion increased PON1 activity in the plasma of healthy indi-viduals. However, in the present study, there was no change inPON1 activity after a prolonged ingestion of mate tea. Despite theabsence of a significant difference after 90 d of consuming matetea ingestion, approximately 50% of dyslipidemic subjectsshowed a significant increase of 23%, on average, in PON1activity, suggesting a potential protection against CVD in theseindividuals. As expected, we also observed a positive associationbetween PON1 activity and HDL cholesterol. However, althoughthe intervention proposed in this study promoted an increasedintake of foods rich in vitamin C, it was not sufficiently effectiveto increase plasma PON1 activity, suggesting that more stringentdietary approaches are needed to observe this effect.

The long-term consumption of mate tea did not lower theconcentration of serum lipid peroxidation in dyslipidemicsubjects as assessed by the measurement of LOOH. Although anintake of mate tea for 1 wk was reported to decrease theconcentration of thiobarbituric acid-reactive substances, indic-ative of lipid peroxidation [21], the prolonged consumption ofother antioxidant beverages, such as green tea, did not lowerLOOH levels in dyslipidemic subjects [29]. Despite the absence ofa significant decrease in the serum levels of LOOH, we observeda positive correlation between LOOH and LDL cholesterol (apossible oxidizable substrate) and an inverse association withHDL cholesterol (a carrier of PON1), indicating that, at least inpart, individuals can decrease lipid peroxidation with animprovement of the lipid profile, particularly by decreasing LDL-cholesterol and increasing HDL-cholesterol. Furthermore, weconfirmed that the long-term consumption of mate tea signifi-cantly lowered serum levels of LDL-cholesterol, as demonstratedin our previous study after mate tea ingestion for 40 d [14]. Thesaponins and phenolic compounds present in mate tea may beresponsible for the hypocholesterolemic effect of the beverage,

by forming complexes with bile salts and cholesterol, inhibitingintestinal absorption of steroids, and/or decreasing the endoge-nous synthesis of cholesterol [42,43].

The nutritional intervention applied in the present study didnot promote a decrease in the serum concentrations of LOOH.However, dyslipidemic individuals in the two groups whoreceived dietary intervention, associated or not with mate tea,showed an inverse correlation between LOOH and theconsumption of foods rich in vitamin C and fiber. Similar resultswere found for individuals with a high risk for developing CVDafter an increased consumption of foods rich in fiber, fruits, andvegetables [44].

In the present study, it was observed that dyslipidemic indi-viduals who received a dietary intervention had a significantdecrease in the consumption of total and saturated fats,accompanied by an increased intake of foods rich in vitamin C.Diet has been considered a major factor in the development ofCVD, and studies have reported that high intakes of cholesteroland saturated and trans-fatty acids are directly related to dysli-pidemia and CVD [45,46]. Conversely, these diseases can beprevented by the consumption of fruits and vegetables. Theprotective effect of these foods is mediated by nutrients such asvitamins and minerals [22,47]. In fact, studies have found aninverse association of a dietary or supplemented consumption ofvitamins A and C and polyphenols with CVD [9,47,48].

Conclusions

The results of this study demonstrated an increase in serumantioxidant capacity and GSH and a decrease in LDL-cholesterolin dyslipidemic individuals after long-term ingestion of mate tea,suggesting that this is an effective dietary supplement for theprevention of CVD in individuals following a free diet. Similarly,the dietary intervention recommended in this study increasedthe antioxidant capacity of serum and blood GSH levels andpromoted a decreased consumption of total and saturated fatsand an increased intake of foods rich in vitamin C, which wasdirectly associated with GSH and inversely correlated to LOOH. Acombination of a mate tea intake and a dietary intervention didnot promote an additional improvement of the oxidative stressbiomarkers. In the future, dietary interventions and long-termstudies should be conducted to evaluate the antioxidant effectof mate in the prevention of CVD.

-30 -20 -10 0 10 20 30

-100

-50

0

50

100

150

200

250

300

Cha

nge

in L

OO

H (%

of b

asel

ine)

Change in LDL-c (% of baseline)

R = 0.337P < 0.005

A

-40 -20 0 20 40 60

-100

-50

0

50

100

150

200

250

300

Cha

nge

in L

OO

H (%

of a

selin

e)

Change in HDL-c (% of baseline)

R = - 0.309P = 0.009

B

-30 -20 -10 0 10 20 30 40

-50

0

50

100

150

200

Cha

nge

in P

ON

1 (%

of b

asel

ine)

Change in HDL-c (% of baseline)

R = 0.263P = 0.016

C

Fig. 2. Correlation between LOOH concentration and (A) LDL-c or (B) HDL-c levelsof dyslipidemic subjects in the mate tea group and between PON1 activity and (C)HDL-c level of dyslipidemic participants in the group with mate tea and a dietaryintervention. HDL-c, high-density lipoprotein cholesterol; LDL-c, low-density lipo-protein cholesterol; LOOH, lipid hydroperoxide; PON1, enzyme paraoxonase-1.

B. C. B. Boaventura et al. / Nutrition 28 (2012) 657–664 663

Acknowledgments

The authors thank Le~ao J�unior Co. (Curitiba, PR, Brazil) forproviding the yerba mate leaves.

References

[1] Fang YZ, Yang S, Wu GY. Free radicals, antioxidants, and nutrition. Nutrition2002;18:872–9.

[2] World Health Organization. Cardiovascular diseases. Available at: www.who.int/cardiovascular_disease/en/. Accessed July, 2011.

[3] Stocker R, Keaney JF. Role of oxidative modifications in atherosclerosis.Physiol Rev 2004;84:1381–478.

[4] Serdar Z, Aslan K, Dirican M, Sarandol E, Yesilbursa D, Serdar A. Lipid andprotein oxidation and antioxidant status in patients with angiographicallyproven coronary artery disease. Clin Biochem 2006;39:794–803.

[5] Shimizu H, Kiyohara Y, Kato I, Kitazono T, Tanizaki Y, Kubo M, et al.Relationship between plasma glutathione levels and cardiovasculardisease in a defined populationdthe Hisayama study. Stroke2004;35:2072–7.

[6] Mackness MI, Harty D, Bhatnagar D, Winocour PH, Arrol S, Ishola M, et al.Serum paraoxonase activity in familial hypercholesterolemia and insulin-dependent diabetes-mellitus. Atherosclerosis 1991;86:193–9.

[7] Geleijnse JM, Launer LJ, van der Kuip DAM, Hofman A, Witteman JCM.Inverse association of tea and flavonoid intakes with incident myocardialinfarction: the Rotterdam Study. Am J Clin Nutr 2002;75:880–6.

[8] Osganian SK, Stampfer MJ, Rimm E, Spiegelman D, Manson JE, Willett WC.Dietary carotenoids and risk of coronary artery disease in women. Am J ClinNutr 2003;77:1390–9.

[9] N�u~nez-C�ordoba JM, Mart�ınez-Gonz�alez MA. Antioxidant vitamins andcardiovascular disease. Curr Topics Med Chem 2011;11:1861–9.

[10] de Mejia EG, Song YS, Heck CI, Ramirez-Mares M. Yerba mate tea (Ilexparaguariensis): phenolics, antioxidant capacity and in vitro inhibition ofcolon cancer cell proliferation. J Funct Foods 2010;2:23–34.

[11] Bracesco N, Sanchez AG, Contreras V, Menini T, Gugliucci A. Recentadvances on Ilex paraguariensis research: mini-review. J Ethnopharmacol2011;136:378–84.

[12] Carini M, Facino RM, Aldini G, Calloni M, Colombo L. Characterization ofphenolic antioxidants from mate (Ilex paraguayensis) by liquid chroma-tography mass spectrometry and liquid chromatography tandem massspectrometry. Rapid Commun Mass Spectrom 1998;12:1813–9.

[13] Bastos DHM, Fornari AC, Queiroz YS, Torres E. Bioactive compounds contentof Chimarrao infusions related to the moisture of yerba mate (Ilex para-guariensis) leaves. Braz Arch Biol Technol 2006;49:399–404.

[14] Morais EC, Stefanuto A, Klein GA, Boaventura BCB, de Andrade F,Wazlawik E, et al. Consumption of yerba mate (Ilex paraguariensis)improves serum lipid parameters in healthy dyslipidemic subjects andprovides an additional LDL-cholesterol reduction in individuals on statintherapy. J Agric Food Chem 2009;57:8316–24.

[15] Bastos DHM, Ishimoto EY, Marques MOM, Ferri AF, Torres E. Essential oiland antioxidant activity of green mate and mate tea (Ilex paraguariensis)infusions. J Food Composition Anal 2006;19:538–43.

[16] Gugliucci A, Stahl AJC. Low-density-lipoprotein oxidation is inhibited byextracts of Ilex paraguariensis. Biochem Mol Biol Int 1995;35:47–56.

[17] Gugliucci A. Antioxidant effects of Ilex paraguariensis: induction ofdecreased oxidability of human LDL in vivo. Biochem Biophys Res Commun1996;224:338–44.

[18] Bracesco N, Dell M, Rocha A, Behtash S, Menini T, Gugliucci A, et al. Anti-oxidant activity of a botanical extract preparation of Ilex paraguariensis:prevention of DNA double-strand breaks in Saccharomyces cerevisiae andhuman low-density lipoprotein oxidation. J Altern Complement Med2003;9:379–87.

[19] Silva EL, Neiva TJC, Shirai M, Terao J, Abdalla DSP. Acute ingestion of yerbamate infusion (Ilex paraguariensis) inhibits plasma and lipoprotein oxida-tion. Food Res Int 2008;41:973–9.

[20] Menini T, Heck C, Schulze J, de Mejia E, Gugliucci A. Protective action of Ilexparaguariensis extract against free radical inactivation of paraoxonase-1 inhigh-density lipoprotein. Planta Med 2007;73:1141–7.

[21] Matsumoto RLT, Bastos DHM, Mendonca S, Nunes VS, Bartchewsky W,Ribeiro ML, Carvalho PO. Effects of mate tea (Ilex paraguariensis) ingestionon mRNA expression of antioxidant enzymes, lipid peroxidation, and totalantioxidant status in healthy young women. J Agric Food Chem2009;57:1775–80.

[22] Department of Atherosclerosis. Brazilian Society of Cardiology. IV Brazilianguidelines on dyslipidemia and prevention of atherosclerosis. Arq BrasCardiol 2007;88:3–19.

[23] World Health Organization. Physical status: the use and interpretation ofanthropometry. WHO technical report, series 854. Geneva: World HealthOrganization; 1995.

[24] Benzie IFF, Strain JJ. The ferric reducing ability of plasma (FRAP) as ameasureof “antioxidant power”: the FRAP assay. Anal Biochem 1996;239:70–6.

[25] Beutler E, Kelly BM, Duron O. Improved method for determination of bloodglutathione. J Lab Clin Med 1963;61:882–8.

[26] Gan KN, Smolen A, Eckerson HW, Ladu BN. Purification of human serumparaoxonase arylesterasedevidence for one esterase catalyzing bothactivities. Drug Metab Dispos 1991;19:100–6.

[27] Nourooz-Zadeh J, Tajaddini-Sarmadi J, Wolff SP. Measurement of plasmahydroperoxide concentrations by the ferrous oxidation-xylenol orange

B. C. B. Boaventura et al. / Nutrition 28 (2012) 657–664664

assay in conjunction with triphenylphosphine. Anal Biochem1994;220:403–9.

[28] Levine RL, Garland D, Oliver CN, Amici A, Climent I, Lenz AG, et al. Deter-mination of carbonyl content in oxidatively modified proteins. MethodsEnzymol 1990;186:464–78.

[29] de Santana MB, Mandarino MG, Cardoso JR, Dichi I, Dichi JB, Camargo AEI,et al. Association between soy and green tea (Camellia sinensis) diminisheshypercholesterolemia and increases total plasma antioxidant potential indyslipidemic subjects. Nutrition 2008;24:562–8.

[30] Camargo AEI, Daguer DA, Barbosa DS. Green tea exerts antioxidant actionin vitro and its consumption increases total serum antioxidant potential innormal and dyslipidemic subjects. Nutr Res 2006;26:626–31.

[31] Wayner DDM, Burton GW, Ingold KU, Barclay LRC, Locke SJ. The relativecontributions of vitamin E, urate, ascorbate and proteins to the total per-oxyl radical-trapping antioxidant activity of human-blood plasma. BiochimBiophys Acta 1987;924:408–19.

[32] Gungor N, Ozyurek M, Guclu K, Cekic SD, Apak R. Comparative evaluationof antioxidant capacities of thiol-based antioxidants measured by differentin vitro methods. Talanta 2011;83:1650–8.

[33] Lahouel M, Boutabet K, Kebsa W, Alyane M. Polyphenolic fractions ofAlgerian propolis reverses doxorubicin induced acute renal oxidativestress. Afr J Pharm Pharmacol 2010;4:712–20.

[34] Vecera R, Skottova N, Vana P, Kazdova L, Chmela Z, Svagera Z, et al. Antiox-idant status, lipoprotein profile and liver lipids in rats fed on high-cholesterol diet containing currant oil rich in n-3 and n-6 polyunsaturatedfatty acids. Physiol Res 2003;52:177–87.

[35] Codoner-Franch P, Lopez-Jaen AB, Muniz P, Sentandreu E, Belles VV.Mandarin juice improves the antioxidant status of hypercholesterolemicchildren. J Pediatr Gastroenterol Nutr 2008;47:349–55.

[36] Hayes JD, Flanagan JU, Jowsey IR. Glutathione transferases. Annu RevPharmacol Toxicol 2005;45:51–88.

[37] Blankenberg S, Rupprecht HJ, Bickel C, Torzewski M, Hafner G, Tiret L, et al.Glutathione peroxidase 1 activity and cardiovascular events in patientswith coronary artery disease. N Engl J Med 2003;349:1605–13.

[38] Mackness B, Quarck R, Verreth W, Mackness M, Holvoet P. Humanparaoxonase-1 overexpression inhibits atherosclerosis in a mousemodel of metabolic syndrome. Arterioscler Thromb Vasc Biol2006;26:1545–50.

[39] Jarvik GP, Tsai NT, McKinstry LA, Wani R, Brophy VH, Richter RJ, et al.Vitamin C and E intake is associated with increased paraoxonase activity.Arterioscler Thromb Vasc Biol 2002;22:1329–33.

[40] Kleemola P, Freese R, Jauhiainen M, Pahlman R, Alfthan G, Mutanen M.Dietary determinants of serum paraoxonase activity in healthy humans.Atherosclerosis 2002;160:425–32.

[41] Sutherland WHF, Walker RJ, de Jong SA, van Rij AM, Phillips V, Walker HL.Reduced postprandial serum paraoxonase activity after a meal rich in usedcooking fat. Arterioscler Thromb Vasc Biol 1999;19:1340–7.

[42] Ferreira F, Vazquez A, Guntner C, Moyna P. Inhibition of the passivediffusion of cholic acid by the Ilex paraguariensis St Hil saponins. PhytotherRes 1997;11:79–81.

[43] Yeh YH, Lee YT, Hsieh HS, Hwang DF. Dietary caffeic acid, ferulic acid andcoumaric acid supplements on cholesterol metabolism and antioxidantactivity in rats. J Food Drug Anal 2009;17:123–32.

[44] Roberts CK, Ng C, Hama S, Eliseo AJ, Barnard RJ. Effect of a short-term dietand exercise intervention on inflammatory/antiinflammatory properties ofHDL in overweight/obese men with cardiovascular risk factors. J ApplPhysiol 2006;101:1727–32.

[45] O’Keefe JH, Gheewala NM, O’Keefe JO. Dietary strategies for improvingpost-prandial glucose, lipids, inflammation, and cardiovascular health. JAm Coll Cardiol 2008;51:249–55.

[46] Nicklas TA, Dwyer J, Feldman HA, Luepker RV, Kelder SH, Nader PR. Serumcholesterol levels in children are associated with dietary fat and fatty acidintake. J Am Diet Assoc 2002;102:511–7.

[47] Hu FB, Willett WC. Optimal diets for prevention of coronary heart disease.JAMA 2002;288:2569–78.

[48] Ford ES, Giles WH. Serum vitamins, carotenoids, and angina pectoris:findings from the National Health and Nutrition Examination Survey III.Ann Epidemiol 2000;10:106–16.