ARTICLE IN PRESS

0952-3278/$ - se

doi:10.1016/j.pl

�Correspondfax: +216 73 46

E-mail addr

mohamed.hamm

Prostaglandins, Leukotrienes and Essential Fatty Acids 74 (2006) 323–329

www.elsevier.com/locate/plefa

Association between dietary fat and antioxidant status of Tunisiantype 2 diabetic patients

M. Smaouia, N. Koubaaa, S. Hammamib, N. Abidc, M. Fekic, R. Chaabaa, N. Attiaa,M. Abidc, M. Hammamia,�

aLaboratory of Biochemistry, UR ‘Nutrition Humaine et desordres metaboliques’, Faculty of Medicine, Avicenne Street, Monastir 5019, TunisiabDepartment of Internal Medicine, CHU F, Bourguiba, Monastir

cDepartment of Endocrinology-Diabetology, CHUH, Chaker, Sfax, Tunisia

Received 13 November 2005; received in revised form 11 February 2006; accepted 14 February 2006

Abstract

Background: The antioxidant enzymes: superoxide dismutase (Cu/Zn SOD) and glutathione peroxidase (GSH-Px) provide a

defense against the damage of cells by reactive oxygen species, which increased in diabetic state. It was demonstrated that dietary

treatment could improve the antioxidant status in patients with type 2 diabetes mellitus. This study was undertaken to determine if

erythrocyte Cu/Zn SOD and GSH-Px activities correlate with dietary nutrients in 35 selected type 2 diabetic patients (21 women and

14 men) without diabetic complications.

Results: We found that erythrocyte Cu/Zn SOD was diminished in patients with poor controlled diabetes and GSH-Px activity was

significantly decreased in obese compared with non-obese type 2 diabetic patients (1.0770.87 and 2.3671.99U/ml, respectively;

P ¼ 0.024). Both erythrocyte Cu/Zn SOD and GSH-Px activities were positively correlated to erythrocyte o3-polyunsaturated fatty

acids (PUFA). In non-obese diabetic patients, only GSH-Px activity was correlated negatively to the fraction of linoleic acid

(18:2o6) and arachidonic acid (20:4o6) in erythrocytes phospholipids.

Conclusions: The data of this study reveal that activities of erythrocyte antioxidant enzymes were altered in type 2 diabetic patients.

Further studies are needed to determine if diet supplemented with o3-PUFA is required to improve antioxidant defense system in

diabetic state.

r 2006 Elsevier Ltd. All rights reserved.

1. Introduction

Diabetes mellitus (DM) is recognized as one ofthe leading causes of morbidity and mortality in theworld [1]. It is by far the most common seriousmetabolic disorder, with a world-wide prevalenceestimated between 1% and 5% [2]. Patients with type2 DM are at increased risk of developing cardiovasculardiseases (CVD). Assessment of conventional risk factorssuch lipids, lipoproteins and hypertension only partlyaccount for this excessive risk [3]. Our previous findings

e front matter r 2006 Elsevier Ltd. All rights reserved.

efa.2006.02.003

ing author. Tel.: +216 73 462 200;

0 737.

esses: m_smaoui@yahoo.com (M. Smaoui),

ami@fmm.rnu.tn (M. Hammami).

showed that the increased risk for CVD in Tunisiandiabetic population may be related to dyslipidemia andother ‘‘novel’’ factors like Lipoprotein(a) [4] andtransfer of cholesteryl esters mediated by CholesterylEster Transfer Protein (CETP) [5].

There has been currently great interest in the potentialcontribution of the enhanced oxidative stress to thedevelopment of various complications seen in diabetes[6,7]. Under diabetic conditions, persistent hyperglycemiamay cause high production of reactive oxygen species viaglucose auto-oxidation and/or protein glycation invarious tissues [8]. The biological effects of free radicalsare controlled by various cellular defense mechanismsconsisting of enzymatic and non-enzymatic scavengercomponents [9,10]. Among antioxidant enzymes, super-oxide dismutase (SOD) catalyzes dismutation of the

ARTICLE IN PRESSM. Smaoui et al. / Prostaglandins, Leukotrienes and Essential Fatty Acids 74 (2006) 323–329324

superoxide anion (O2d�) into H2O2 [11] and glutathione

peroxidase (GSH-Px) detoxifies H2O2 into H2O byconverting reduced glutathione (GSH) into oxidizedglutathione (GSSG) [12]. It has been proposed that poorglycemic control is associated with the depletion ofprotective serum antioxidant enzyme activities includingSOD and GHS-Px in type 2 DM patients [13]. Asknown, in diabetes, basic treatment is intensive dietarytreatment. Studies focused on the support of antioxidantsystems of the sufferers, have demonstrated the bene-ficial effect of dietary treatment [14]. The fatty acidprofile of red blood cell (RBC) phospholipids is affectedby the relative abundance of the different fatty acidsrequired for their synthesis. This in turn depends onthe composition of the dietary fat and on fatty acidbioconversion in the body [15].

The purpose of the present study was to determine therelationship between the erythrocyte activity of SODand GSH-Px and nutrient intakes in Tunisian free-livingpatients with type 2 DM.

2. Methods

2.1. Subjects

We studied 35 free-living type 2 diabetic patients (21women and 14 men) recruited from Endocrine out-patient Clinic at H. Chaker hospital of Sfax (Tunisia).Informed consent was obtained from all the individualsafter the purpose of the study had been explained.Inclusion criteria were as follows: (i) type 2 DMdiagnosis as defined by the criteria of the AmericanDiabetes Association [16], (ii) age less than 70 years;(iii) body mass index (BMI) less than 35 kg/m2. Themean duration of diabetes was 8.7877.01 years and73.6% of patients had less than 10 years of diabetes.Diabetic patients had no diabetes complications (cardi-ovascular disease, diabetic nephropathy or retinopathy)and they were treated with diet alone or with thecombination of diet plus oral hypoglycemic drugs:sulfonylureas (Gilbenclamide 5–15mg/day) and/or Met-formin (850–2550mg/day). The abdominal obesitymeasured by waist circumference was present in almost90% of patients. The patients using lipid lowering drugs,treated with insulin or multivitamins including anyantioxidant, having renal or liver failure or thyroiddisease and who were alcohol consumer 3 days prior tothe study or less, were not included to the study.Postmenopausal women had no hormone replacementtherapy.

2.2. Biochemical determinations

Venous blood samples were drawn in empty and inEDTA vacuum tubes after an overnight fast. Plasma and

erythrocytes (RBC) were separated by centrifugation at3000 rpm for 10min at room temperature. RBC werewashed three times with 0.9% NaCl solution, and thenhemolyzed with four volumes of cold distilled water.Plasma samples and erythrocyte lysates were stored at�70 1C until assayed. Blood glucose, and plasma lipidsand apolipoproteins A1 and B were determined asdescribed elsewhere [4]. Glycemic control was assessedby measuring glycated hemoglobin (Hb A1c). Erythro-cyte activity of GSH-Px and SOD was estimated onhemolysates. GSH-Px estimation was based on themethod of Paglia and Valentine [17] by measuringthe decrease of NADPH absorbance at 340nm. Thecoefficient of variability between assays was 4%. Theactivity of SOD was measured at 500nm by testingthe inhibition degree of tetrazolium salt oxidationreaction [18]. The coefficient of variability betweenassays was 4.2%. For analysis of fatty acid compositionof erythrocyte phospholipids, lipids were extracted fromthe hemolyzed material according to Folch et al. [19].After extraction with hexane and methanolysis, themethyl esters-fatty acids were analyzed by capillary-gasliquid chromatography (CGL) under chromatographicconditions described previously [20].

2.3. Dietary survey

Food intakes were estimated by two dietitians usingan open-ended, interview-administered diet history.Subjects were asked on their daily diet over a weekperiod. They were also asked on amounts, frequenciesand variations in consumption. Energy and nutrientintakes were calculated by using the software Nutri-tionist IV Computer Analysis Program [21].

2.4. Statistical analysis

The data were analyzed using the Statistical Packagefor Social Sciences (SPSS 10.0 for Windows). Means ofall measurements are presented with standard deviation(mean7SD) or as percentage (%). Spearman’s rankcorrelation evaluated relationships between oxidantdefense enzyme (SOD and GSH-Px activities) andbiochemical parameters. The criterion for significancewas Po0.05.

3. Results

In the group of 35 type 2 diabetic patients studiedhere, 71.4% of patients were at poor controlled diabetesand 15 subjects (42.8%) were obese (BMI 430 kg/m2)(Table 1). A surprising finding is that diabetics hadnormal lipid and lipoprotein profile as general Tunisianpopulation. It seems that diabetic dyslipidemia did not

ARTICLE IN PRESSM. Smaoui et al. / Prostaglandins, Leukotrienes and Essential Fatty Acids 74 (2006) 323–329 325

take place in these patients like demonstrated inTunisian diabetic patients living in Monastir region [4].

We focused on the relationships between somemarkers of the oxidative defense system and dietaryintakes in these patients with type 2 diabetes. Ourprevious results showed that erythrocyte SODactivity was decreased and GSH-Px activity was similarin type 2 DM patients compared to controls (Data notpublished). The diabetic patients were divided into

Table 1

Clinical and biochemical characteristics of type 2 diabetic patients

(n ¼ 35)

Parameter Type 2 diabetic patients (n ¼ 35)

Age (years) 52.0979.75 (39–70)

Sex-ratio (male/female) 14/21

Physical activity (%)

Low 60

Moderate 22.9

High 17.1

Waist circumference (cm) 104.33710.08 (91–122)

Hip circumference (cm) 111.6179.70 (100–140)

Waist to hip ratio (WHR) 0.9370.078 (0.82–1.17)

Body mass index (BMI) (kg/m2) 28.5574.38 (20.02–40.16)

Obesitya (%) 42.8

Fasting glucose (mM) 9.6974.60 (4.20–18)

HbA1c (%) 8.3871.30 (5.27–11.10)

Poor controlled diabetes (%) 71.4

Diabetes duration (years) 8.7877.01

o10 years (%) 73.6

Smoking (%) 21.4

Postmenopausal women (%) 62

Total cholesterol (mmol/l) 5.2471.17 (3.43–7.74)

Total triglyceride (mmol/l) 1.8371.14 (0.52–5.94)

HDL-C (mmol/l) 1.0970.31 (0.48–1.87)

LDL-C (mmol/l) 3.3171.19 (1.12–5.89)

ApoA1 (g/l) 1.3570.21 (1.01–1.81)

ApoB (g/l) 1.0370.28 (0.55–1.70)

Data are expressed as mean7SD or as %. Range of values is shown in

parenthesis.aAbdominal obesity measured by waist circumference was present in

90% of patients (n ¼ 32).

Table 2

Erythrocyte activity of glutathione peroxidase (GSH-Px) and superoxide

according to physical activity, glycemic control and presence or not of obes

GSH-Px ac

Physical activity

Low (n ¼ 21) 1.6071.82

Moderate or high (n ¼ 14) 1.9671.36

Glycemic control

HbA1C o7% (n ¼ 10) 2.0872.02

HbA1C 47% (n ¼ 25) 2.1271.85

Obesity

Non-obese (n ¼ 20) 2 .3671.99

Obese (n ¼ 15) 1.0770.87

Data are expressed as mean7SD.

*Po0.05.

different subgroups and compared for erythrocyteSOD and GSH-Px activities: those with poor control(HbA1c47%) versus those with good control ofdiabetes (o7%), those with low physical activity versusthose with moderate or high physical activity and obeseversus lean subjects. As shown in Table 2, erythrocyteGSH-Px activity was significantly higher in lean than inobese diabetic patients (2.3671.99 vs. 1.0770.87U/ml,P ¼ 0.024) and erythrocyte SOD activity was lower indiabetic patients with poor glycemic control than inthose with good glycemic control (58.90758.41 vs.117.07758.15U/ml, respectively, P ¼ 0.04). Further-more, correlation analysis revealed a significant negativecorrelations as well as between GSH-Px activity andwaist circumference (r ¼ �0.71, P ¼ 0.019) or totaltriglycerides (r ¼ �0.48, P ¼ 0.049) and between SODactivity and total cholesterol (r ¼ �0.58, P ¼ 0.027) intype 2 diabetic patients.

In Table 3, given the correlation coefficients betweenerythrocyte SOD and GSH-Px activities and dietaryintakes in type 2 diabetic patients. There was asignificant correlation between SOD activity and thepercentage of energy derived from protein (r ¼ 0.63,P ¼ 0.039) and fat (r ¼ �0.38, P ¼ 0.040) in diabeticpatients. While no correlation was found between GSH-Px and these macronutrients (Table 3). However, theactivity of GSH-Px was correlated negatively withdietary cholesterol (r ¼ �0.50, P ¼ 0.009) and positivelywith the mean intake of calcium (r ¼ 0.34, P ¼ 0.045)and phosphorus (r ¼ 0.35, P ¼ 0.043).

The analysis of fatty acid composition of erythrocytemembrane in diabetic patients showed that the highestproportion was presented by polyunsaturated fatty acids(PUFA) with predominance of o3-fatty acids. Signifi-cant correlations between fasting blood glucose and o3-PUFA (r ¼ �0.66, P ¼ 0.002) (Fig. 1) and o6-PUFA(r ¼ 0.38, P ¼ 0.044) (Fig. 2) were detected. Thediabetic patients were evaluated for the relationshipbetween antioxidant enzyme activities of erythrocyteSOD or GSH-Px and RBC fatty acid composition

dismutase (SOD) in different subgroups of type 2 diabetic patients

ity of type 2 diabetic patients

tivity (U/ml) SOD activity (U/ml)

85.88771.81

60.59737.03

117.07758.15

58.90758.41*

57.87737.34

* 93.16773.48

ARTICLE IN PRESS

Table 3

Correlation coefficients between erythrocyte antioxidant activity of glutathione peroxidase (GSH-Px) or superoxide dismutase (SOD) and nutrient

intakes of type 2 diabetic patients (n ¼ 35)

Nutrient Mean7SD Correlation coefficients with nutrient intake

GSH-Px activity (U/ml) SOD activity (U/ml)

Kcalories 2171.557494.85 0.19 �0.27

Protein (%) 11.3872.08 0.11 0.63*

Carbohydrate (%) 52.8578.69 0.10 0.29

Fat (%) 35.8179.43 �0.13 �0.38*

SFA 8.7372.39 �0.05 �0.17

MUFA 20.8177.53 �0.22 �0.32

PUFA 6.3573.88 0.07 �0.25

P/S 0.7970.48 0.09 �0.07

Cholesterol (mg) 198.957120.26 �0.50* �0.13

Fiber (g) 17.2075.01 0.43* �0.08

Calcium (mg) 449.147151.80 0.34* �0.20

Vitamin B1 (mg) 0.4970.12 0.19 �0.15

Potassium (mg) 2372.737631.01 0.06 0.08

Phosphorus (mg) 8757181.60 0.35* �0.15

Magnesium (mg) 241.17759.75 0.27 �0.18

Zinc (mg) 971.97 0.04 0.12

Folates (mg) 150.03777.09 0.20 �0.23

Vitamin C (mg) 88.61756.70 0.03 �0.02

Vitamin E (mg) 7.2973.34 �0.05 �0.29

SFA: Saturated fatty acids, MUFA: Monounsaturated fatty acids, PUFA: Polyunsaturated fatty acids and P/S: PUFA to SFA ratio

*Po 0.05.

Fasting blood glucose (mmol/l)

Red

blo

od c

ell w

3-P

UFA

(ar

ea %

)

22

26

30

34

38

42

4 6 8 10 12 14 16 18 20

Fig. 1. Correlation between RBC o3-polyunsaturated fatty acids

(PUFA) and fasting blood glucose in type 2 diabetic patients

(r ¼ �0.66; Po0.001).

Fasting blood glucose (mmol/l)

Red

blo

od c

ell w

6-P

UFA

(ar

ea %

)

0

4

8

12

16

20

24

4 6 8 10 12 14 16 18 20

RégressionIC à 95%

Fig. 2. Correlation between RBC o6-polyunsaturated fatty acids

(PUFA) and fasting blood glucose in type 2 diabetic patients (r ¼ 0.44;

P ¼ 0.04).

M. Smaoui et al. / Prostaglandins, Leukotrienes and Essential Fatty Acids 74 (2006) 323–329326

(Table 4). Erythrocyte GSH-Px activity was correlatedpositively with a-linolenic acid (18:3o3) and docosa-hexaenoic acid (DHA) (r ¼ 0.34, P ¼ 0.042 andr ¼ 0.43, P ¼ 0.009, respectively). Besides, SOD activitywas correlated positively to eicosapentaenoic acid (EPA)(r ¼ 0.49, P ¼ 0.006) and C22:5o3 (r ¼ 0.45, P ¼

0.012) but negatively to palmitic acid (r ¼ �0.51,P ¼ 0.004). It follows from this fatty acids compositionof erythrocyte membrane that only SOD activity wassignificantly correlated to o3-PUFA (Table 4). In non-obese diabetic patients, significant correlations werefound between erythrocyte GSH-Px activity and linoleic

acid (18:2o6) (r ¼ �0.46, P ¼ 0.034), arachidonic acid(20:4o6) (r ¼ �0.56, P ¼ 0.008) and also with the totalof o6-PUFA (r ¼ �0.48, P ¼ 0.026).

4. Discussion

The erythrocyte activity of Cu/Zn SOD which servesas a major antioxidant enzyme in RBC [9] was altered indiabetic patients with uncontrolled type 2 DM. Possibleexplanations for mechanisms are several. In diabetes,the decrease in activity of antioxidant enzymes may

ARTICLE IN PRESS

Table 4

Correlation coefficients between erythrocyte antioxidant activity of glutathione peroxidase (GSH-Px) or superoxide dismutase (SOD) and

erythrocyte membrane fatty acids of type 2 diabetic patients

Fatty acid Mean (%)7SD Correlation coefficients with fatty acid weight (%)

GSH-Px activity (U/ml) SOD activity (U/ml)

C16:0 23.7978.39 �0.06 �0.51a

C18:0 14.2474.04 0.23 0.25

C18:1 o9 6.3273.99 0.32 �0.28

C18:2o6 3.7173.78 �0.22 �0.13

C18:3 o3 0.9071.63 0.34a 0.26

C20:2o6 1.0671.12 0.25 0.57a

C20:3 o6 1.1070.56 �0.17 0.07

C20:4o6 3.3372.92 �0.25 �0.23

C20:5o3 (EPA) 2.2171.41 �0.04 0.49a

C22:0 0.5770.86 0.23 0.18

C22:1 0.7371.43 0.35a 0.41a

C22:4o6 1.6071.41 �0.007 0.53a

C22:5 o3 1.8772.25 0.05 0.45a

C22:6 o3 (DHA) 15.1376.03 0.43a 0.29

C24:0 0.9170.94 0.057 0.24

C24:1 1.8371.44 0.07 0.03PSFA 42.5476.61 0.03 �0.25

PMUFA 13.1575.27 0.13 �0.23

PPUFA 44.3078 �0.13 0.46aPo3-PUFA 33.4778.10 0.25 0.58a

Po6-PUFA 10.8375.60 �0.28 �0.08

EPA/DHA 0.1670.11 �0.54a �0.32

o3/o6 4.2572.81 0.40a 0.22

P/S 1.0870.35 �0.05 0.35a

PSFA, sum of saturated fatty acids;

PMUFA, sum of monounsaturated fatty acids;

PPUFA, sum of polyunsaturated fatty acids; o3/o6,

Po3-

PUFA toP

o6-PUFA ratio and P/S,P

PUFA toP

SFA ratio.aExpressed as % of total fatty acids.

M. Smaoui et al. / Prostaglandins, Leukotrienes and Essential Fatty Acids 74 (2006) 323–329 327

reflect the known sensitivity of these enzymes to radical-induced inactivation [22]. Superoxide anion radicals arecontinuously generated by the oxidation of hemoglobin[23]. Hydrogen peroxide (H2O2) accumulated in diabetesand it may inhibit SOD activity [24]. A negativecorrelation of Cu/Zn SOD activity and glycosylatedhemoglobin was reported by Tho et al. [25] indicatingthat progressive glycation of erythrocyte SOD is linkedto the low active form. An other study enrolled byKotake et al. demonstrated that the decrease in activityof Cu/Zn SOD in erythrocyte of diabetics is consistentwith glycosylation of the active site of Cu/Zn SOD,without any effect of hyperglycemia on the concentra-tion of Cu/Zn SOD [26]. Another cause that mayproduce a decrease in the amount or activity of Cu/ZnSOD is zinc deficiency [27] like observed in our studygroup (Table 3). In fact, the mean of daily zinc intakewas above the recommended daily allowance [28].

Erythrocyte GSH-Px activity in type 2 diabeticpatients had been shown to be not different from thecontrols in some investigations [27–29]. Aaccordingly,we found unchanged GSH-Px activity between diabeticand control group, probably due to sufficient seleniumintake of our patients. The status of selenium which the

antioxidant role is directly linked to its essentialinvolvement in the glutathione peroxidase enzymesystem (Se-GSH-Px) [30] seemed to be not altered inTunisian patients with type 2 DM [29]. However,erythrocyte Se-GSH-Px activity in obese diabetic sub-jects was decreased compared to non-obese patients(Table 2). This result suggest that this antioxidantenzyme could not cope with the excess production offree radicals mostly by glucose auto-oxidation due tomarked insulin resistance or insulin deficiency [31].A depletion of the crucial protector glutathione (GSH)which is an essential cofactor of GSH-Px, was reportedin a group of overweight type 2 diabetic patients [32] indiabetic state.

Sekeroglu and co-workers demonstrated that stan-dard dietary treatment in type 2 diabetes produced asubstantial improvement in erythrocyte antioxidantstatus and reduced serum and erythrocyte lipid per-oxidation [14]. To evaluate the effects of diet habits onthe same markers of antioxidant system, nutrients intakeof patients consuming their usual diet, were evaluatedby dietary questionnaire. Although the limitations of themethod used, our results showed that the percentagecontribution of protein and carbohydrate were similar

ARTICLE IN PRESSM. Smaoui et al. / Prostaglandins, Leukotrienes and Essential Fatty Acids 74 (2006) 323–329328

and comparable to the recommended values for thediabetic diet [28]. The dietary fat, however, was underthe recommended daily allowance of 30–35% of totalcalories, with exceeded polyunsaturated to saturatedfatty acids (40.7) ratio [28]. The high intake ofmonounsaturated fatty acids (MUFA) reflected higholive oil consumption of our population. The benefits ofMediterranean-style diet on lipid and lipoproteinmetabolism was well documented [33].

Negative correlations between dietary fat and Cu/ZnSOD activity on one hand and between cholesterolintake and GSH-Px activity on other hand, wereobserved (Table 3). So, particular attention wasaddressed to the fatty acid composition of erythrocytemembranes. In fact, this fatty acid composition isaffected by the relative abundance of the different fattyacids required for their synthesis over a period ofmonths [15]. In turn, relative abundance of fatty acidsreflects dietary intake and on fatty acid biosynthesis inthe body [34]. Our results showed a negative correlationbetween erythrocyte palmitic acid and erythrocyte SODactivity which is in line with the recommendation thatsaturated fatty acids intake should be limited [28]. Theinteresting finding is that both GSH-Px and SOD werepositively correlated to erythrocyte o3-PIFA. It is wellknown that o3-fatty acids have a wide range ofbiological effects, including benefits on lipoproteinmetabolism, platelet function, endothelial function andvascular reactivity, inflammatory markers, cytokineproduction, coagulation, and fibrinolysis [35,36]. So,increased o3-fatty acids may be of particular benefit topatients with type 2 DM [37]. Recently, associationsbetween insulin resistance and long-chain o6-PUFA innon-obese type 2 diabetics were described by Rodriguezet al. [38]. Based on our results, we put forward thehypothesis that conditions, reduced intake of o6/o3-polyunsaturated fatty may be beneficial for antioxidantsystem in type 2 diabetic patients.

In conclusion, our results showed a decline ofendogenous antioxidant defense system capability thatmay contribute to oxidative stress in Tunisian type 2diabetic patients. Erythrocyte activities of Cu/Zn SODand GSH-Px were correlated positively to o3-PUFAin RBC phospholipids. A clear relationship betweenglycemic control and the content in o3- and o6-PUFAof erythrocyte suggested needed individual dietaryadvice for a large portion of diabetic patients in viewof their poor glycemic control and obesity.

Acknowledgments

This research was supported by a grant from the‘‘Ministere de l’Enseignement Superieur (DGRST;USCR-SM)’’ and the ‘‘Ministere de la RechercheScientifique de la Technologie et du Developpement

des Competences’’ (UR: ‘‘Nutrition Humaine et deso-rdres metaboliques’’). We would like to thank Dr.Noureddine Gazzah, Mrs. Faiza Hakim and Mrs. NejlaKilani for their expert technical support.

References

[1] W.B. Kannel, D.L. Mc Gee, Diabetes and cardiovascular disease.

The Framingham Study, J. Am. Med. Assoc. 241 (1979)

2035–2038.

[2] P. Froguel, G. Velho, Les marqueurs genetiques du diabete de

type 2, Rev. Prat. 46 (1996) 1813–1817.

[3] J.M. Hayden, P.D. Reaven, Cardiovascular disease in diabetes

mellitus type 2: a potential role for novel cardiovascular risk

factors, Curr. Opin. Lipidol. 11 (2000) 519–528.

[4] M. Smaoui, S. Hammami, R. Chaaba, et al., Lipids and

lipoprotein(a) concentrations in Tunisian type 2 diabetic patients :

relationship to glycemic control and coronary heart disease,

J. Diabetes Complicat. 18 (2004) 258–263.

[5] R. Chaaba, S. Hammami, N. Attia, et al., Association of plasma

cholesteryl ester transfer protein activity and polymorphism with

coronary artery disease extent in Tunisian type II diabetic

patients, Clin. Biochem. 38 (2005) 373–378.

[6] R.K. Sundaram, A. Bhaskar, S. Vijayalingham, M. Viswanatham,

R. Mohan, K.R. Shanmugasundaram, Antioxidant status and

lipid peroxidation in type II diabetes mellitus with and without

complications, Clin. Sci. 90 (1996) 255.

[7] J.W. Baynes, S.R. Thorope, Role of oxidative stress in diabetic

complications, Diabetes 48 (1999) 1–9.

[8] J.V. Hunt, C.C. Smith, S.P. Wolff, Auto oxidative glycosylation

and possible involvement of peroxides and free radicals in LDL

modification by glucose, Diabetes 39 (1990) 1420–1421.

[9] S.A. Wohaieb, D.V. Godin, Alterations in free radical tissue-

defense mechanism in streptozotocin-induced diabetes in rat,

Diabetes 36 (1987) 1014–1018.

[10] J.M. Mates, C. Perez-Gomez, I. Nunez de Castro, Antioxidant

enzymes and human diseases, Clin. Biochem. 32 (1999) 595–603.

[11] K. Brawn, I. Fridovich, Superoxide radical and superoxide

dismutase: threat and defense, Acta Physiol. Scand. 492 (1980)

9–18.

[12] L. Guemouri, Y. Artur, B. Herbeth, Biological variability of

superoxide dismutase, glutathione peroxidase, and catalase in

blood, Clin. Chem. 37 (1991) 1932–1937.

[13] S.R.J. Maxwell, H. Thomason, D. Sandler, et al., Poor glycaemic

control is associated with reduced serum free radical scavenging

(antioxidant) activity in non-insulin-dependent diabetes mellitus,

Ann. Clin. Biochem. 34 (1997) 638.

[14] R.M. Sekeroglu, H. Sahin, H. Dulger, E. Algun, The effect of

dietary treatment on erythrocyte lipid peroxidation, superoxide

dismutase, glutathione peroxidase, and serum lipid peroxidation

in patients with type 2 diabetes mellitus, Clin. Biochem. 33 (2000)

647–669.

[15] M.B. Katan, J.P. Deslypere, A.P.J.M. Van Birgelen, M. Pender,

M. Zegwaard, Kinetics of the incorporation of dietary fatty acids

into serum cholesteryl esters, erythrocyte membranes and adipose

tissue: an 18-month controlled study, J. Lipid Res. 38 (1997)

2012–2022.

[16] The expert committee on the diagnosis and classification of

diabetes mellitus. Report of the expert committee on the diagnosis

and classification of diabetes mellitus. Diabetes Care 20 (1997)

1183–1197.

[17] D.E. Paglia, W.N. Valentine, Studies on the quantitative and

qualitative characterization of erythrocyte glutathione peroxidase,

J. Lab. Clin. Med. 70 (1967) 158–169.

ARTICLE IN PRESSM. Smaoui et al. / Prostaglandins, Leukotrienes and Essential Fatty Acids 74 (2006) 323–329 329

[18] Y. Sun, L.W. Oberley, Y. Li, A simple method for clinical assay

of superoxide dismutase, Clin. Chem. 34 (1988) 497–500.

[19] J. Folch, M. Lees, S.G.M. Sloane Stanley, A simple method for

the isolation and purification of total lipids from animal tissues,

J. Biol. Chem. 226 (1995) 497–509.

[20] M. Smaoui, S. Hammami, N. Attia, et al., Modulation of

cholesterol ester transfer protein activity by plasma unsaturated

fatty acids in Tunisian type 2 diabetic women, Nutr. Metab.

Cardiovasc. Dis. 16 (2006) 44–53.

[21] Nutritionist IV Computer Analysis Program, 1994, Version 3.1,

N2 Computing, Hearst Corp., Salem, OR.

[22] A. Ceriello, D. Giugliano, A. Quatraro, P. Dello Russo, P.J.

Lefebvre, Metabolic control may influence the increased super-

oxide generation in diabetic serum, Diabetic Med. 8 (1991)

540–542.

[23] A.C. Maritim, R.A. Sanders, J.B. Watkins, Diabetes, oxidative

stress and antioxidants: a review, J. Biochem. Mol. Toxicol. 17

(2003) 24–38.

[24] S. Narayana, Free radical involvement in complications of

diabetes, Radiat. Res. Lab. 77 (1992) 308.

[25] L.L. Tho, J.K. Candlish, A.C. Thai, Correlates of diabetes

markers with erythrocyte enzymes decomposing reactive oxygen

species, Ann. Clin. Biochem. 25 (1988) 426–431.

[26] M. Kotake, R. Shinohara, K. Kato, et al., Reduction of activity,

but not decrease in concentration of erythrocyte Cu, Zn-super-

oxide dismutase by hyperglycemia in diabetic patients, Diabetic

Med. 15 (1998) 668–671.

[27] L.H. Chen, Y.D. Osio, J.W. Anderson, Blood antioxidant defense

system and dietary survey of elderly diabetic men, Arch.

Gerontol. Geriatr. 28 (1999) 65–83.

[28] L. Monnier, G. Slama, B. Vialettes, O. Ziegler, ALFEDIAM

recommandations. Nutrition et Diabete,, Diabetes Metab. 21

(1995) 207–217.

[29] R.A. Anderson, A.M. Roussel, N. Zouari, S. Mahjoub, J.M.

Matheau, A. Kerken, Potential antioxidant effects of zinc and

chromium supplementation in people with type 2 diabetes

mellitus, J. Am. Coll. Nutr. 20 (2001) 212–218.

[30] C. Hadrzynski, Diabetes and trace elements, J. Trace Elem. Exp.

Med. 12 (1999) 367–374.

[31] J.V. Silha, M. Krsek, J.V. Skrha, P. Sucharda, B.L. Nyomba, L.J.

Murphy, Plasma resistin, adiponectin and leptin levels in lean

and obese subjects: correlations with insulin resistance, Eur.

J. Endocrinol. 149 (2003) 331–335.

[32] J. Aaseth, S. Birketvedt, Glutathione in overweight patients with

poorly controlled type 2 diabetes, J. Trace Elem. Exp. Med. 13

(2000) 105–111.

[33] F.M. Sacks, Katan, Randomised clinical trials on the effects of

dietary fat and carbohydrate on plasma lipoproteins and

cardiovascular disease, Am. J. Med. 113 (2002) 13–24.

[34] J.E. Brown, R.M. Lindsay, R.A. Riemersma, Linoleic

acid metabolism in the spontaneously diabetic rat: D6desaturase activity vs. product/precursor ratios, Lipids 35 (2000)

1319–1323.

[35] T.A. Mori, L.J. Beilin, n-3 Fatty acids, blood lipids and

cardiovascular risk reduction, Curr. Opin. Lipidol. 12 (2001)

11–17.

[36] S.L. Connor, W.E. Connor, Are fish oils beneficial in the

prevention and treatment of coronary artery disease?, Am.

J. Clin. Nutr. 66 (1997) 1020–1031.

[37] C.E. Friedberg, M.J. Janssen, R.J. Heine, D.E. Grobbee, Fish oil

and glycemic control in diabetes. A meta-analysis, Diabetes Care

21 (1998) 494–500.

[38] Y. Rodriguez, A.B. Christophe, Long chain o6 polyunsaturaed

fatty acids in erythrocyte phospholipids are associated with

insulin resistance in non-obese type 2 diabetics, Clin. Chim. Acta

354 (2005) 195–199.