Boiled wild artichoke reduces postprandial glycemic and insulinemic responses in normal subjects but has no effect on metabolic syndrome patients

Available online at www.sciencedirect.com

Nutrition Research 27 (2007) 741–749www.elsevier.com/locate/nutres

Boiled wild artichoke reduces postprandial glycemic andinsulinemic responses in normal subjects but has no effect on

metabolic syndrome patientsTzortzis Nomikos, Paraskevi Detopoulou, Elisabeth Fragopoulou,

Emmanuel Pliakis, Smaragdi Antonopoulou⁎

Department of Science of Nutrition-Dietetics, Harokopio University, Athens 17671, Greece

Received 22 May 2007; revised 4 September 2007; accepted 12 September 2007

Abstract

Postprandial hyperglycemia is a common complication in patients with metabolic syndrome

⁎ Corresponding aE-mail address: an

0271-5317/$ – see frodoi:10.1016/j.nutres.2

(MS), and an improvement of their postprandial status could be beneficial for them. The aim of thisstudy was to assess for the first time the effect of a wild artichoke (Cynara cardunculus) meal onpostprandial glucose and insulin levels in patients with MS fulfilling the National CholesterolEducation Program Adult Treatment Panel III criteria (n = 19) and normal subjects (n = 8). We alsoevaluated the postprandial homocysteine levels because its metabolism seems to be affected byinsulin. A randomized, crossover design was followed in which all subjects consumed a wildartichoke meal (boiled wild artichoke, white bread, refined olive oil, and lemon juice) and a controlmeal that did not contain the boiled artichoke. Glucose, insulin, and homocysteine levels weremeasured postprandially for 4 hours. The boiled artichoke, as compared with the control meal,abolished the postprandial increase of glucose (P = .031) and attenuated the postprandial insulinresponse (P = .047) 0.5 hours after the meal in normal subjects, whereas it had no effect in patientswith MS. The consumption of the artichoke meal by both groups did not influence the postprandialhomocysteine levels compared with the control meal. In conclusion, the boiled wild artichoke couldnot improve postprandial glycemic profile in subjects with MS, whereas its hypoglycemic effectscould protect normal subjects from the noxious consequences of postprandial hyperglycemia.© 2007 Elsevier Inc. All rights reserved.

Keywords: Cynara cardunculus; Plant; Metabolic syndrome X; Postprandial period; Hyperglycemia; Homocysteine; Diet; Mediterranean; Humans

1. Introduction

The metabolic syndrome (MS) is characterized by adysregulation of the metabolic system defined by insulinresistance, atherogenic dyslipidemia (elevated levels ofserum triacylglycerols and small dense low-density lipopro-tein cholesterol [LDL-C] along with low levels of high-density lipoprotein cholesterol [HDL-C]), hypertension, and

uthor. Tel.: +30 2109549230; fax: +30 [email protected] (S. Antonopoulou).

nt matter © 2007 Elsevier Inc. All rights reserved.007.09.009

central obesity [1,2]. Among other etiological factors,lifestyle (especially a sedentary lifestyle and dietary habits)strongly correlates with obesity, insulin resistance, dyslipi-demia, and inflammation [3,4]. Therefore, the AmericanDiabetes Association and the National Cholesterol Educa-tion Program Adult Treatment Panel III have suggested thatthe starting point for the treatment of MS should betherapeutic lifestyle changes such as loss of excess bodyfat, increased physical activity, smoking cessation, and ofcourse dietary modifications [5,6].

Postprandial hyperglycemia and hyperlipidemia arecommon features of MS and early markers for diabetes

742 T. Nomikos et al. / Nutrition Research 27 (2007) 741–749

and atherosclerosis [7,8]. They can induce oxidative stress[9], activation of the coagulation system [10], endothelialdysfunction, and inflammation [11]. Laboratory data clearlyshow that insulin can alter the enzymes of homocysteinemetabolism, thus affecting its levels [12,13]. Acute hyper-insulinemia can regulate homocysteine levels only innondiabetic patients implying that insulin resistance maybe a cause of elevated homocysteine levels [14]. Becausehomocysteine is a putative independent risk factor forcardiovascular diseases [15], its postprandial variations maycontribute to the atherosclerotic process. Considering theWestern type of life, which is characterized by plentifulmeals at regular intervals, postprandial hyperglycemia isa daily feature of people with insulin resistance, whichmay have detrimental metabolic effects. Therefore, a dailyreduction of postprandial glucose excursions could bebeneficial for patients with MS.

Dietary patterns can play a major role to this directionbecause diets rich in high glycemic index foods areassociated with higher postprandial glucose and insulinlevels and increased demand for insulin secretion, whereasan increased consumption of low glycemic index foods andespecially fibers has the opposite results [16]. Wild greenplants, widely consumed throughout the Mediterraneanbasin [17], are a rich source of fiber [18] and therefore areputative hypoglycemic agents.

Artichoke, either globe artichoke (Cynara scolymus) orwild artichoke (Cynara cardunculus) is widely consumed,raw or cooked, throughout the Mediterranean region [17].It contains significant amounts of dietary fiber, folate,vitamin B6, b-carotene, magnesium, potassium, andcopper [19,20]. Organic or aqueous extracts of the rawplant, a rich source of polyphenols [21-23], demonstratestrong antimicrobial [24] and antioxidant activity in cellsystems [21,25], inhibit cholesterol biosynthesis inhepatocytes [26], and modulate vasomotor function[22,27]. Clinical studies with raw artichoke extractshave shown contradictory results, either having no effector moderately improving the lipemic status of hyperlipi-demic patients [28,29]. From a nutritional point of view,studies using raw extracts are of limited importancebecause these extracts represent only a small fraction ofthe plant. Moreover, thermal processes during cookingmay significantly alter the composition of the consumedplant and its metabolic effects in comparison with the rawplant or extracts.

C cardunculus (wild artichoke) is traditionally consideredby the Greeks as a plant with antidiabetic properties.However, neither long-term studies nor postprandial studies,investigating its putative antihyperglycemic action, havebeen conducted. Therefore, the aim of this study was toinvestigate the impact of boiled wild artichoke on thepostprandial excursions of glucose, insulin, and homocys-teine, induced by the garniture of a traditional wild artichokemeal, in patients with MS and normal volunteers. As faras we know, this is the first time that the health effects

of a cooked plant, incorporated in a real traditional meal,are studied.

2. Methods and materials

2.1. Subjects

Two groups of volunteers were recruited for the study.Initially, a pilot study with a small number of healthyvolunteers (8 male volunteers) was conducted, whereas themain study was carried out with 19 volunteers diagnosed ashaving MS. All volunteers lived in the region of Athens. Thediagnosis of the MS was made according to the criteria ofthe National Cholesterol Education ProgramAdult TreatmentPanel III [30] by the doctors of the EndocrinologyDepartment of the General Hospital of Athens and theDiabetology Department of the Nikea General Hospital ofAthens. Specifically, patients with MS should fulfill 3 of thefollowing criteria: waist circumference N102 cm for men orN88 cm for women, concentration of serum triacylglycerols≥1.695 mmol/L, concentration of HDL-C b1.036 mmol/Lfor men or b1.295 mmol/L for women, blood pressure≥130/85 mm Hg, and fasting glucose levels ≥6.1 mmol/L.Exclusion criteria were the presence of renal or liver disease,thyroid dysfunction, congestive heart failure, angina, cancer,retinopathy, diabetic neuropathy, peripheral vascular disease,known gastrointestinal diseases, use of supplements andsteroids, aversion to plant foods, heavy drinking, and the useof antibiotics and anti-inflammatory drugs at least 3 weeksbefore the intervention. Patients with MS treated with oralantidiabetic, hypolipidemic, and antihypertensive medicationdid not take any medication before the meal consumption.The nature and purpose of the study were explained to eachsubject, and all participants gave written informed consent.

2.2. The meals

The wild artichokes (C cardunculus) were collected inCrete during April. The plants were immediately delivered toHarokopio University where the seed heads were cleaned up,weighted, separated into equal portions of 300 g, scalded for5 minutes, and stored at approximately −20°C until the dayof the experiment. This is a common way by which Greekspreserve the wild plants for later use.

The volunteers with MS consumed on 2 separate days thecynara meal, consisting of the boiled green plant (100 g ofboiled plant), 3 slices of bread, 60 mL of refined oil, and7.5 mL of canned lemon juice, and a control meal thatincluded all of the above apart from the boiled green plant. Thenormal volunteers consumed smaller portions of each mealcomponent because of their lower body mass. Specifically,they ate 70 g of the boiled plant, 2 slices of bread, 40 mL ofrefined olive oil, and 5 mL of lemon juice. The 22.8% ofthe control meal's calories is derived from carbohydrates, the74.2% from fats, and the 3.7% from proteins. The respectivevalues for the artichoke meal are 25.5%, 67.8%, and 5.0%.The small difference in the percentage of calories deriving

Table 1Nutritional composition of the control and artichoke meals consumed by patients with MS and normal volunteers (in brackets)

Nutritionalcomposition

Control meal Artichoke meal

Refinedolive oil

White bread Lemon juice Total meal Refinedolive oil

White bread Lemon juice Boiledartichoke

Total meal

Weight (g) 56.8 (37.9) 75.0 (50.0) 7.50 (5.00) 139.4 (92.9) 56.8 (37.9) 75.0 (50.0) 7.5 (5.0) 100.0 (70.0) 239.4 (162.9)Carbohydrate (g) 0 (0) 40.8 (27.2) 0.49 (0.32) 41.29 (27.52) 0 (0) 40.8 (27.2) 0.49 (0.32) 9.3 (6.5) 50.59 (34.02)Fiber (g) 0 (0) 1.9 (1.3) 0 (0) 1.9 (1.3) 0 (0) 1.9 (1.3) 0 (0) 5.4 (3.8) 7.3 (5.1)Protein (g) 0 (0) 6.75 (4.50) 0.03 (0.02) 6.78 (4.52) 0 (0) 6.75 (4.50) 0.03 (0.02) 3.20 (2.24) 9.98 (6.76)Fats (g) 56.7 (37.8) 3.00 (2.00) 0.02 (0.01) 59.72 (39.81) 56.7 (37.8) 3.00 (2.00) 0.02 (0.01) 0.16 (0.11) 59.88 (39.92)SFA (g) 8.10 (5.40) 0.43 (0.29) 0.003 (0.002) 8.54 (5.69) 8.10 (5.40) 0.43 (0.29) 0.003 (0.002) 0.037 (0.026) 8.47 (5.72)MUFA (g) 41.70 (27.9) 0.597 (0.398) 0.00 (0.00) 42.30 (28.30) 41.70 (27.9) 0.597 (0.398) 0.00 (0.00) 0.01 (0.00) 42.30 (28.30)PUFA (g) 4.20 (2.80) 1.57 (1.04) 0.01 (0.00) 5.77 (3.85) 4.20 (2.80) 1.57 (1.04) 0.01 (0.00) 0.07 (0.05) 5.84 (3.90)Water (g) 0 (0) 22.8 (15.2) 6.93 (4.62) 29.73 (19.82) 0 (0) 22.8 (15.2) 6.93 (4.62) 81.9 (57.3) 111.63 (77.12)Energy (kJ) 2100 (1401) 920 (615) 8 (4) 3028 (2020) 2100 (1401) 920 (615) 8 (4) 293 (205) 3321 (2225)

Number of volunteers: n = 8 for normal subjects and n = 19 for subjects with MS. The values for the commercially prepared white bread and canned lemon juiceare based on the USDANational Nutrient Database for Standard Reference [20]. The values for the wild artichoke are based on the food database developed fromthe Department of Social Medicine of the University of Crete [19]. The values for the olive oil are based on the food database of Trichopoulou [31].SFA indicates saturated fatty acids; MUFA, monounsaturated fatty acids; PUFA, polyunsaturated fatty acids.

743T. Nomikos et al. / Nutrition Research 27 (2007) 741–749

from carbohydrates and proteins is due to the carbohydrateand protein content of the wild artichoke. All meals werefreshly prepared on the day of the intervention. Boiledwild green plants accompanied with olive oil, lemon juice,and bread is a traditional way by which these plants areconsumed in Greece and especially Crete [31]. The nutritionalcomposition of the wild artichoke and control meals is shownin Table 1. The calculation of the nutrient content of themeals was based on the following databases: US Departmentof Agriculture food database for the commercially preparedwhite bread and canned lemon juice [20], the food databasedeveloped at the Department of Social Medicine of theUniversity of Crete for the wild artichoke [19], and the fooddatabase of Trichopoulou for the olive oil [32].C cardunculusis also a source of polyphenols, such as apigenin, luteolin,dicaffeoylquinic, and chlorogrenic acid [22,23,27]. In parti-cular, stems, flowers, and receptacles of C cardunculuscoming from Italy contain 448.8 mg of polyphenols per gramof ethanolic extract, whereas leaf rachis of C cardunculuscoming from Spain contains 214.4 mg of polyphenols pergram of ethanolic extract [21].

2.3. Study design

The study had a randomized, double-masked, crossoverdesign in which each subject consumed a control meal(control group) and a wild artichoke meal (treatment group)in random order. The consumption of the 2 meals was spaced2 weeks for each subject.

At least 3 days before each meal, the volunteers weregiven detailed advice to follow a diet low in dietary fiber,vegetables, green plants, and generally antioxidant andphenolic compounds. They were not allowed to drink coffee,tea, or alcohol (washout period), and they were advised tokeep their usual physical activity. After a 12-hour fast period,the volunteers visited Harokopio University. Before the

administration of the meal, blood pressure was measured.A cannula with an Intima 20-gauge intravenous catheter(Becton Dickinson, Rutherford, NJ, USA) was insertedinto the brachial vein, and after a 15-minute rest, a fastingblood sample was drawn by registered personnel. Subjectsconsumed the entire meal within 15 minutes, and the timeof meal completion was considered time 0 for subsequentmeasurements. Blood was collected again at 0.5, 1, 2, 3,and 4 hours. During this time, anthropometric parameterswere measured. The volunteers also filled out detailedquestionnaires concerning the demographic, socioeco-nomic data, medical history, smoking, dietary habits, andphysical activity. Subjects remained supine for theduration of the study and were allowed to consumenothing but mineral water. The protocol and humansubject use were approved by the Bioethics Committee ofHarokopio University and were in accordance with theDeclaration of Helsinki (1989) of the World MedicalAssociation. A schematic representation of the protocol isdepicted in Fig. 1.

2.4. Blood sampling and handling

All blood samples were collected from the brachial vein ofthe volunteers. Venous blood samples, for the determinationof glucose, insulin, triacylglycerols, total cholesterol (TC),LDL-C, HDL-C, apolipoprotein A1, apolipoprotein B, andhomocysteine, were drawn into evacuated glass tubes (BDVacutainer Systems, Becton Dickinson, Plymouth, UK), andserum was collected after half an hour by centrifugation at1200 × g for 10 minutes at 10°C. For the determination offibrinogen, 1.8 mL of venous blood samples was drawn intoevacuated BD citrated vacutainers, immediately centrifugedat 800 × g for 10 minutes at 4°C, and the supernatant plasmawas collected. Plasma or serum was immediately aliquotedand stored at approximately −80°C.

Fig. 1. Schematic representation of the experimental protocol. Washout period: subjects followed a diet low in dietary fiber, vegetables, green plants, andgenerally antioxidant and phenolic compounds. They were not allowed to drink coffee, tea, or alcohol, and they were advised to keep their usual physical activity.Biochemical measurements: baseline levels of glucose, insulin, triacylglycerols, TC, LDL-C, HDL-C, apolipoprotein A1, apolipoprotein B, and homocysteinewere measured. Postprandially, we assessed the levels of glucose, insulin, and homocysteine. Anthropometric measurements: height, weight, waist, and hipcircumference were measured. From these measurements, BMI and waist-to-hip ratio were calculated.

744 T. Nomikos et al. / Nutrition Research 27 (2007) 741–749

2.5. Anthropometry and blood pressure measurement

Anthropometry was carried out on each visit for bothnormal subjects and subjects with MS according tostandardized procedures [33]. Weight was measured to thenearest 0.1 kg using a digital scale (Seca). Height wasmeasured to the nearest 0.1 cm using a stadiometer withhead in horizontal Frankfurt plane. Both measurements weretaken with the subject in light clothing and without shoes.Body mass index (BMI) was then calculated as weight (kg)

Table 2Anthropometric and biochemical characteristics of normal subjects and volunteers

Normal volunteers (n = 8)

Control meal Artich

Male/Female 6/0Age (y) 33.8 ± 12.5 (22-57)Smokers 2BMI (kg/m2) 25.9 ± 3.2 25.Waist (cm) 89.7 ± 10.9 89.Hip (cm) 99.8 ± 5.6 99.Waist-to-hip ratio 0.893 ± 0.068 0.89Systolic BP (mm Hg) 110 ± 12 11Diastolic BP (mm Hg) 77 ± 5 7Glucose (mmol/L) 5.439 ± 0.555 5.462Insulin (pmol/L) 39.586 ± 14.584 38.89HOMA-IR 1.38 ± 0.39 1.3Triacylglycerols (mmol/L) 0.944 ± 0.502 1.08TC (mmol/L) 5.322 ± 1.088 5.18LDL-C (mmol/L) 3.703 ± 0.932 3.60HDL-C (mmol/L) 1.175 ± 0.166 1.08Apolipoprotein A1 (g/L) 1.40 ± 0.15 1.4Apolipoprotein B (g/L) 0.76 ± 0.25 0.7Fibrinogen (μmol/L) 7.673 ± 1.999 6.97Homocysteine (μmol/L) 12.7 ± 2.2 12.

The measurements were made in fasting blood obtained before the consumption ogroup of normal volunteers was not matched with the MS group; therefore, no direcdifferences were observed within each group (control meal versus artichoke mealBP indicates blood pressure.

divided by height squared (m2). Waist circumference wasmeasured with a tape between the superior iliac crest andthe lower rib margin in the midaxillary line, after a moderateexpiration. Hip circumference was measured as the maximalhorizontal circumference at the level of the buttocks. Bothmeasurements were taken to the nearest 0.1 cm. Waist-to-hipratio was then calculated as waist circumference (cm)divided by hip circumference (cm). A physician measuredthe blood pressure on the right arm of the volunteers using astandard sphygmomanometer after at least 20 minutes of

with MS

MS volunteers (n = 19)

oke meal Control meal Artichoke meal

5/1456 ± 12 (35-72)

107 ± 3.1 34.7 ± 5.4 34.9 ± 5.61 ± 9.4 104.2 ± 14.9 104.6 ± 14.63 ± 6.4 114.6 ± 10.9 115.1 ± 11.21 ± 0.059 0.908 ± 0.009 0.908 ± 0.0861 ± 13 138 ± 15 136 ± 166 ± 7 82 ± 7 82 ± 63 ± 0.671 7.659 ± 2.775 7.326 ± 2.6082 ± 4.167 80.562 ± 45.142 77.089 ± 33.3364 ± 0.22 3.9 ± 2.3 3.5 ± 1.41 ± 0.523 2.113 ± 0.598 2.181 ± 0.6558 ± 1.100 5.905 ± 1.062 5.827 ± 0.9580 ± 1.191 4.014 ± 0.751 4.169 ± 0.9068 ± 0.083 1.088 ± 0.104 1.114 ± 0.1040 ± 0.18 1.42 ± 0.17 1.39 ± 0.147 ± 0.19 0.96 ± 0.18 0.99 ± 0.183 ± 1.899 11.231 ± 3.381 10.584 ± 2.2343 ± 3.2 13.6 ± 3.1 12.5 ± 2.5

f the control or the artichoke meal. Values are expressed as mean ± SD. Thet comparisons between the 2 groups can be made. No statistically significant) as assessed by a paired t test (level of significance was set at P b .05).

Fig. 2. (A) Postprandial serum glucose concentrations and (B) postprandial serum insulin concentrations after the consumption of the control and the artichokemeal by the normal volunteers (n = 8). Values are expressed as mean ± SEM. †Significant difference of postprandial levels between the control and wildartichoke meal as assessed with RM-ANOVA adjusted for age, sex, and BMI (P b .05). *Significant difference from baseline value as assessed with pairedt test (P b .05).

745T. Nomikos et al. / Nutrition Research 27 (2007) 741–749

seated rest and before the meal consumption. For eachsubject, 3 measurements were taken at 1-minute intervals,and an average of the last 2 was used in the analysis [34].

2.6. Biochemical measurements

Serum glucose, triacylglycerols, TC, LDL-C, and HDL-Cwere determined enzymatically with an ACE biochemicalanalyzer (Schiapparelli Biosystems, Inc, NJ, USA) usingreagents from AlfaWassermann (Woerden, The Netherlands)[35-38]. Serum insulin was determined by an immunoenzy-mometric assay on an AIA 600 II enzyme immunoassaysystem (Tosoh Corporation, Tokyo, Japan). Serum apolipo-protein A1 and apolipoprotein B were determined by animmunoturbidometric assay [39]. Specifically, they formeda precipitate with a specific antiserum (sheep antiserumfor apolipoprotein A and rabbit antiserum for apolipopro-tein B), which was determined turbidimetrically at 340 nmby using the Cobas Integra 700 biochemical analyzer(Roche Diagnostics, GmbH, Mannheim, Germany). Serumhomocysteine concentrations were measured by high-performance liquid chromatography with fluorometric detec-

tion [40]. For the fibrinogen levels evaluation, citratedplasma was immediately loaded on a Coatron 2 analyser(Biotechnology, Greece), and fibrinogen was measured usingTeclot Fib. (W. Kaolin, TECO Medical In, G) [41]. Home-ostasis model assessment-insulin resistance (HOMA-IR)formula was calculated using the following formula:HOMA-IR = fasting insulin (μU/mL) × fasting glucose(mmol/L) / 22.5 [42].

2.7. Statistical analysis

Normal distribution of variables was checked using theKolmogorov-Smirnov criterion before further analysis.Differences in the baseline characteristics of subjects alongwith differences between the baseline and the postprandialvalues were assessed by a paired t test [43]. A repeated-measures analysis of variance (RM-ANOVA) was used tocompare the postprandial curves of the biochemical para-meters between the control and the cynara meal [43]. In theRM-ANOVA analysis, the artichoke consumption was set asbetween-subjects factor, and potential confounding variableswere used as covariates (age, sex, BMI). Data are expressed

Fig. 3. (A) Postprandial serum glucose concentrations and (B) postprandial serum insulin concentrations after the consumption of the control and the artichokemeal by the patients with MS (n = 19). Values are expressed as mean ± SEM. No significant difference was found concerning the postprandial glucose and insulinlevels between the control and the wild artichoke meal as assessed with RM-ANOVA. *Significant difference from baseline value as assessed with paired t test(P b .05).

746 T. Nomikos et al. / Nutrition Research 27 (2007) 741–749

as mean ± SD at the tables and as mean ± SEM in the figures.The level of statistical significance was set at 5% for allcomparisons. The Statistical Package for the Social Sciences(SPSS 13.0 for Windows, 2005, Chicago, IL) was used for allthe analysis.

Table 3Percentage change of the postprandial homocysteine values from thebaseline values

Timepoints(h)

% change of homocysteine from baseline

Normal subjects Patients with MS

Control meal Artichoke meal Control meal Artichoke meal

0 100 a 100 b 100 c 100 d

2 94.0 ± 5.8 97.4 ± 2.0 98.2 ± 5.8 97.2 ± 4.43 96.9 ± 9.3 97.5 ± 1.8 97.4 ± 6.0 97.7 ± 4.14 94.9 ± 5.3 99.6 ± 2.1 101.0 ± 6.5 99.4 ± 7.5

The group of normal volunteers (n = 8) was not matched with the MS group(n = 19); therefore, no direct comparisons between the 2 groups can be made.No statistically significant differences were observed within each group(control meal vs artichoke meal) as assessed by RM-ANOVA (level ofsignificance was set at P b .05). Values are expressed as mean ± SD.

a Baseline value 12.7 ± 2.2 μmol/L.b Baseline value 12.3 ± 3.4 μmol/L.c Baseline value 13.6 ± 3.1 μmol/L.d Baseline value 12.5 ± 2.5 μmol/L.

3. Results

3.1. Anthropometric and biochemical characteristicsof volunteers

The baseline anthropometric and biochemical profiles ofthe normal volunteers and patients with MS are shown inTable 2. As expected, most MS volunteers were character-ized by abdominal obesity. Moreover, the biochemicalprofile of the volunteers, determined at fasting state, wascharacterized by quite high levels of serum glucose andinsulin, abnormally high levels of TC, LDL-C, triacylglycer-ols and fibrinogen, and low levels of HDL-C, indicating ageneral dysregulation of the metabolic system of thesevolunteers. No significant differences were found concern-ing the fasting levels of the biochemical and anthropometricparameters of the volunteers between the 2 meals. The group

of normal volunteers was recruited for the pilot study of thiswork, and it is not age-, sex-, or BMI-matched with theMS group; therefore, no direct comparisons between the2 groups can be made.

747T. Nomikos et al. / Nutrition Research 27 (2007) 741–749

3.2. Effect of wild artichoke on postprandial hyperglycemiaand hyperinsulinemia of normal subjects

Fig. 2A and B show the postprandial serum glucose andinsulin concentrations after the consumption of the control orthe artichoke meal by normal volunteers. No significantdifferences in baseline plasma glucose and insulin valuesacross treatments were found. As it was expected, maximumglucose concentration was reached at 0.5 to 1 hour after thecontrol meal, and the same postprandial pattern was observedfor insulin. On the other hand, the addition of the boiledartichoke to the control meal (artichoke meal) abolished thepostprandial increase of glucose (P = .079) especially at0.5 hour after the meal (P = .031). A lower postprandialinsulin increase was also observed after the artichoke mealcompared with the postprandial insulin increase of the controlmeal, which reached statistical significance only at 0.5 hourafter the meals (P = .047).

3.3. Effect of wild artichoke on postprandial hyperglycemiaand hyperinsulinemia of subjects with MS

The postprandial serum glucose and insulin changes afterthe consumption of the control meal by the subjects with MSare represented in Fig. 3A and B. A delayed and moresustained increase of postprandial glucose was observedafter the consumption of the control meal, but no changes ofthe postprandial values of glucose after the consumption ofthe control meal and the artichoke meal were found againstto what we observed in the case of normal subjects. Althoughthe postprandial insulin levels of the artichoke meal wereslightly higher than the respective values of the control meal,RM-ANOVA analysis showed no statistically significantdifference between the 2 curves.

3.4. Effect of wild artichoke on postprandial levels ofhomocysteine for normal subjects and subjects with MS

The postprandial levels of homocysteine were measured2, 3, and 4 hours after the meals. The percent changes fromthe baseline values are shown in Table 3. A minor decreaseof homocysteine levels after the consumption of the controlmeal from the normal subjects was found; however, thischange was not significantly different from the baselinevalues. Artichoke meal caused a smaller decrease of thepostprandial homocysteine levels compared with the controlmeal. The difference between the 2 curves was marginallysignificant (P = .065). A nonsignificant 2% decrease of thehomocysteine levels was found 2 and 3 hours after theconsumption of the control meal by the patients with MS.The consumption of the artichoke meal, as compared withthe control meal, by the same patients did not influence thepostprandial homocysteine levels.

4. Discussion

Most postprandial studies usually use plant extracts givenin combination with meals of high caloric, glucose, and lipid

content [44-46]. In our study, we provided a traditional mealwith boiled artichoke, as during cooking, several ingredientsof the plant can decompose or be extracted by the water.Aside from the plant, the meal also contained bread, lemonjuice, and olive oil, a combination of foods that areconsumed by the Greeks for a light lunch or supper [31].Refined olive oil was used instead of virgin olive oil becausethe latter is a rich source of polyphenols and the othermicronutrients may affect or mask the postprandial actionsof wild artichoke.

The main outcome of this study was the different responseof normal subjects and subjects with MS to the wildartichoke meal. In normal volunteers, boiled artichokecompletely suppressed the postprandial glucose increase,although it contained carbohydrates. However, only a smallreduction of insulin levels was observed. A possibleexplanation for this discrepancy may lie on the concomitantfat-induced secretion of the incretins glucagon-like peptide-1and gastric inhibitory polypeptide, which can induce insulinsecretion from the β cells [47]. Therefore, even if artichoke'sdietary fiber can slower gastric emptying and keeppostprandial glucose levels low, the aforementioned peptidescan still stimulate insulin secretion.

As far as patients with MS are concerned, the artichokehad no effect on the postprandial levels of glucose ascompared with the control meal. It seems that the artichokecould not attenuate the absorption of glucose in contrast towhat we observed with the normal subjects. Probably, theslow absorption of glucose due to artichoke's fiber incombination with the relatively low amount of providedcarbohydrates and the proper function of insulin acted assupportive factors in the study of postprandial glucosemetabolism. On the other hand, the inability of artichoke tolower postprandial hyperglycemia in patients with MS couldbe attributed to the impaired function of insulin in peripheraltissue and the increased absorption of monosaccharidesdue to the up-regulation of glucose transporters in theintestine [48].

In both normal subjects and subjects with MS, nosignificant reductions of postprandial homocysteine levelswere observed after the consumption of the control meal.Similar studies have also shown that homocysteine levelsdecrease after breakfast [49], although in one study thereduction was statistically significant [50]. It seems that thepostprandial insulin increase was combined with a smallreduction of the homocysteine. We must mention that theglycemic load of the control meal was rather low comparedwith other meals offered in similar studies, causing a modestincrease of insulin postprandially. At this point, it isquestionable whether the postprandial levels of insulinobserved in our study were elevated enough to affecthomocysteine levels. Artichoke had no effect on postpran-dial homocysteine levels of both normal subjects andpatients with MS. This result can be explained by theinability of artichoke to alter postprandial insulin levelsmainly in patients with MS. As far as we know, no studies

748 T. Nomikos et al. / Nutrition Research 27 (2007) 741–749

concerning the effect of artichoke extracts on homocysteinemetabolism and synthesis have been made. Therefore, we donot know whether any artichoke ingredient can affecthomocysteine levels postprandially.

In conclusion, the presence of boiled wild artichoke in atraditional, mildly lipemic and glycemic meal had impressivepostprandial hypoglycemic effects in healthy subjects,whereas it had no effect on the postprandial fluctuations ofglucose and insulin in patients with MS. Wild artichoke hadno significant effect on the postprandial levels of homo-cysteine in both groups. Our findings imply that theconsumption of wild artichoke from healthy subjects mayprotect from hyperglycemia and insulin resistance, but theseeffects cannot be generalized because they may be related topopulation-specific characteristics. Moreover, long-termregular consumption of wild plants such as C cardunculusmay be related to beneficial health outcomes.

Acknowledgment

We wish to thank Zacharias Kypriotakis for theidentification and collection of the wild artichoke plants,Demosthenes Panagiotakos for his useful comments on thestatistical analysis, and Margaret Christea for her excellenttechnical support. This work was supported by the EuropeanCommission grant (QLK1-CT-2001-00173).

References

[1] Isomaa B. A major health hazard: the metabolic syndrome. Life Sci2003;73:2395-411.

[2] Eckel RH, Grundy SM, Zimmet PZ. The metabolic syndrome. Lancet2005;365:1415-28.

[3] Stone NJ, Saxon D. Approach to treatment of the patient withmetabolic syndrome: lifestyle therapy. Am J Cardiol 2005;96:15E-21E.

[4] Scott CL. Diagnosis, prevention, and intervention for the metabolicsyndrome. Am J Cardiol 2003;92:35i-42i.

[5] Pritchett AM, Foreyt JP, Mann DL. Treatment of the metabolicsyndrome: the impact of lifestyle modification. Curr Atheroscler Rep2005;7:95Q102.

[6] Stone NJ. Focus on lifestyle change and the metabolic syndrome.Endocrinol Metab Clin North Am 2004;33:493-508.

[7] Ceriello A. Postprandial hyperglycemia and diabetes complications: isit time to treat? Diabetes 2005;54:1-7.

[8] Tushuizen ME, Diamant M, Heine RJ. Postprandial dysmetabolismand cardiovascular disease in type 2 diabetes. Postgrad Med J 2005;81:1-6.

[9] Rebolledo OR, Actis Dato SM. Postprandial hyperglycemia andhyperlipidemia-generated glycoxidative stress: its contribution to thepathogenesis of diabetes complications. Eur Rev Med Pharmacol Sci2005;9:191-208.

[10] Duttaroy AK. Postprandial activation of hemostatic factors: role ofdietary fatty acids. Prostaglandins Leukot Essent Fatty Acids 2005;72:381-91.

[11] Burdge GC, Calder PC. Plasma cytokine response during thepostprandial period: a potential causal process in vascular disease?Br J Nutr 2005;93:3-9.

[12] Tessari P, Coracina A, Kiwanuka E, VedovatoM, Vettore M, Valerio A,et al. Effects of insulin on methionine and homocysteine kinetics intype 2 diabetes with nephropathy. Diabetes 2005;54:2968-76.

[13] Fonseca V, Dicker-Brown A, Ranganathan S, Song W, Barnard RJ,Fink L, et al. Effects of a high-fat-sucrose diet on enzymes inhomocysteine metabolism in the rat. Metabolism 2000;49:736-41.

[14] Fonseca VA, Mudaliar S, Schmidt B, Fink LM, Kern PA, Henry RR.Plasma homocysteine concentrations are regulated by acute hyper-insulinemia in nondiabetic but not type 2 diabetic subjects. Metabolism1998;47:686-9.

[15] Lentz SR. Mechanisms of homocysteine-induced atherothrombosis.J Thromb Haemost 2005;3:1646-54.

[16] Delzenne NM, Cani PD. A place for dietary fibre in the management ofthe metabolic syndrome. Curr Opin Clin Nutr Metab Care 2005;8:636-40.

[17] Rivera D, Obon C, Heinrich M, Inocencio C, Verde A, Fajardo J.Gathered Mediterranean food plants—ethnobotanical investigationsand historical development. Forum Nutr 2006;59:18-74.

[18] Zeghichi S, Kallithraka S, Simopoulos AP, Kypriotakis Z. Nutritionalcomposition of selected wild plants in the diet of Crete. World RevNutr Diet 2003;91:22-40.

[19] Department of Social Medicine of the University of Crete. Composi-tion table of Greek foods. http://nutrition.med.uoc.gr/GreekTables/Main/main.htm.

[20] US Department of Agriculture. USDA National Nutrient Database forStandard Reference Release 18. http://www.nal.usda.gov/fnic/food-comp, 2005.

[21] The Local Food-Nutraceuticals Consortium. Understanding localMediterranean diets: a multidisciplinary pharmacological and ethno-botanical approach. Pharmacol Res 2005;52:353-666.

[22] Grande S, Bogani P, de Saizieu A, Schueler G, Galli C, Visioli F.Vasomodulating potential of Mediterranean wild plant extracts. J AgricFood Chem 2004;52:5021-6.

[23] Sevcikova P, Glatz Z, Slanina J. Analysis of artichoke (Cynaracardunculus L.) extract by means of micellar electrokinetic capillarychromatography. Electrophoresis 2002;23:249-52.

[24] Zhu X, Zhang H, Lo R. Phenolic compounds from the leaf extractof artichoke (Cynara scolymus L.) and their antimicrobial activities.J Agric Food Chem 2004;52:7272-8.

[25] Schaffer S, Eckert GP, Muller WE, Llorach R, Rivera D, Grande S,et al. Hypochlorous acid scavenging properties of local Mediterra-nean plant foods. Lipids 2004;39:1239-47.

[26] Gebhardt R. Inhibition of cholesterol biosynthesis in primarycultured rat hepatocytes by artichoke (Cynara scolymus L.) extracts.J Pharmacol Exp Ther 1998;286:1122-8.

[27] Rossoni G, Grande S, Galli C, Visioli F. Wild artichoke prevents theage-associated loss of vasomotor function. J Agric Food Chem 2005;53:10291-6.

[28] Pittler MH, Thompson CO, Ernst E. Artichoke leaf extract for treat-ing hypercholesterolaemia. Cochrane Database Syst Rev 2002;CD003335.

[29] Lupattelli G, Marchesi S, Lombardini R, Roscini AR, Trinca F,Gemelli F, et al. Artichoke juice improves endothelial function inhyperlipemia. Life Sci 2004;76:775-82.

[30] Expert Panel on Detection, Evaluation, and Treatment of HighBlood Cholesterol in Adults. Executive Summary of The ThirdReport of The National Cholesterol Education Program (NCEP) ExpertPanel on Detection, Evaluation, And Treatment of High BloodCholesterol In Adults (Adult Treatment Panel III). JAMA 2001;285:2486-9247.

[31] Kafatos A, Verhagen H, Moschandreas J, Apostolaki I, Van WesteropJJ. Mediterranean diet of Crete: foods and nutrient content. J Am DietAssoc 2000;100:1487-93.

[32] Trichopoulou A. Composition tables of food and Greek dishes. Athens:Parisianos SA; 2004.

[33] Heyward V, Stolarczyk L. Applied body composition assessment.Leeds, United Kingdom: Human Kinetics; 1996.

[34] Pickering T, Hall J, Appel L, Falkner B, Graves J, Hill M, et al.Recommendations for blood pressure measurement in humans andexperimental animals: part 1: blood pressure measurements in humans:

749T. Nomikos et al. / Nutrition Research 27 (2007) 741–749

a statement of professionals from the subcommittee of professional andpublic education of the American Heart Association council on highblood pressure research. Hypertension 2005;45:142-61.

[35] Klotzsch SG, McNamara JR. Triglyceride measurements: a review ofmethods and interferences. Clin Chem 1990;36:1605-13.

[36] Allain CC, Poon LS, Chan CS, Richmond W, Fu PC. Enzymaticdetermination of total serum cholesterol. Clin Chem 1974;20:470-5.

[37] Friedewald WT, Levy RI, Fredrickson DS. Estimation of theconcentration of low-density lipoprotein cholesterol in plasma,without use of the preparative ultracentrifuge. Clin Chem 1972;18:499-502.

[38] Izawa S, Okada M, Matsui H, Horita Y. A new direct method formeasuring HDL-cholesterol which does not produce any biased values.J Med Pharm Sci 1997;37:1385-8.

[39] Brewer Jr HB, Gregg RE, Hoeg JM, Fojo SS. Apolipoproteins andlipoproteins in human plasma: an overview. Clin Chem 1988;34:B4QB8.

[40] Ueland PM, Refsum H, Stabler SP, Malinow MR, Andersson A, AllenRH. Total homocysteine in plasma or serum: methods and clinicalapplications. Clin Chem 1993;39:1764-79.

[41] Clauss A. Rapid physiological coagulation method in determination offibrinogen. Acta Haematol 1957;17:237-46.

[42] Matthews DR, Hosker JP, Rudenski AS, Naylor BA, Treacher DF,Turner RC. Homeostasis model assessment: insulin resistance andbeta-cell function from fasting plasma glucose and insulin concentra-tions in man. Diabetologia 1985;28:412-9.

[43] Sprinthall RC. Basic statistical analysis. Boston: Allyn and Bacon;2000.

[44] Heacock PM, Hertzler SR, Williams JA, Wolf BW. Effects of amedical food containing an herbal alpha-glucosidase inhibitor onpostprandial glycemia and insulinemia in healthy adults. J Am DietAssoc 2005;105:65-71.

[45] Madar Z, Abel R, Samish S, Arad J. Glucose-lowering effect offenugreek in non-insulin dependent diabetics. Eur J Clin Nutr 1988;42:51-4.

[46] Gregersen S, Jeppesen PB, Holst JJ, Hermansen K. Antihyperglycemiceffects of stevioside in type 2 diabetic subjects. Metabolism 2004;53:73-6.

[47] Gautier JF, Fetita S, Sobngwi E, Salaun-Martin C. Biological actions ofthe incretins GIP and GLP-1 and therapeutic perspectives in patientswith type 2 diabetes. Diabetes Metab 2005;31:233-42.

[48] Dyer J, Wood IS, Palejwala A, Ellis A, Shirazi-Beechey SP. Expressionof monosaccharide transporters in intestine of diabetic humans. Am JPhysiol Gastrointest Liver Physiol 2002;282:G241-8.

[49] Guttormsen AB, Schneede J, Fiskerstrand T, Ueland PM, Refsum HM.Plasma concentrations of homocysteine and other aminothiol com-pounds are related to food intake in healthy human subjects. J Nutr1994;124:1934-41.

[50] Ubbink JB, Vermaak WJ, van der Merwe A, Becker PJ. The effect ofblood sample aging and food consumption on plasma totalhomocysteine levels. Clin Chim Acta 1992;207:119-28.