Original Contribution

OXIDATIVE DAMAGE, PLASMA ANTIOXIDANT CAPACITY, ANDGLUCEMIC CONTROL IN ELDERLY NIDDM PATIENTS

F. AGUIRRE,*† I. MARTIN,*† D. GRINSPON,*† M. RUIZ,* A. HAGER,† T. DE PAOLI,† J. IHLO,† H. A. FARACH,‡

and C. P. POOLE, JR.‡

*Hospital de Clı´nicas ‘‘Jose´ de San Martı´n,’’ Universidad de Buenos Aires, Argentina,†Catedra de Fı´sica, Facultad de Farmacia yBioquımica, and LANAIS-RLBM, Universidad de Buenos Aires, Argentina, and‡Department of Physics and Astronomy,

University of South Carolina, Columbia, SC, USA

(Received13 March 1997;Revised19 July 1997;Accepted11 August1997)

Abstract—A study of oxidative damage was made in elderly noninsulin-dependent diabetes mellitus (NIDDM) patients.A statistically significant increase in glucose and fructosamine was found in fasting NIDDM patients, as well as anincrease in the oxidation induced bytert-butyl hydroperoxide. The Total Reactive Antioxidant Potential (TRAP) of theplasma was much reduced (p , .02) and the uricemia was unchanged. The erythrocytes of diabetic patients show greaterbasal oxidation products (p , .05), and the susceptibility of the diabetic erythrocytes to oxidation injury was also shownto increase in the oxidation induced byt-BOOH (p , .05). Linear regression studies showed that TRAP was associateddirectly with uric acid (p , .05) and inversely with fructosamine and with glucose (p , .03 andp , .05 respectively)in patients with NIDDM, but not in the controls. The levels of fructosamine were found to be related to the basal damageof the red blood cells (direct correlation,p , .001). This study suggest an useful approach to diabetic oxidative stressfor clinical settings. © 1998 Elsevier Science Inc.

Keywords—Free radical, Lipid peroxidation, Diabetes mellitus, Erythrocyte, Low-density lipoproteins

INTRODUCTION

Oxidative damage is considered to play an important rolein the pathogenesis of several diseases. Much recentwork has focused on the role of oxidative disturbances inthe development of chronic complications of diabetesmellitus.1–8 Mechanisms that are thought to be involvedin the increased oxidative stress in diabetes include notonly oxygen free radical generation due to nonenzymaticprotein glycosylation (glycation), autooxidation of gly-cation products, ascorbic acid, and glucose, but alsochanges in the tissue content and activity of antioxidantdefense systems.9–12

In healthy subjects, significant oxidative damage isprevented by a very complex and efficient antioxidantsystem, consisting of a number of interrelated antioxi-dant compounds and enzymes. Several studies have eval-uated plasma antioxidants in diabetic patients.10–13How-ever, most of these studies evaluated individual antioxi-dants that act cooperatively in vivo to provide greater

protection to the organism against radical damage thancould be provided by any single antioxidant acting alone.The evaluation of individual antioxidant defense systemsmight yield contradictory results when considering oxi-dative susceptibility and concomitant damage, and thus,the global plasma antioxidant capability provides inter-esting clinical information. The determination of theTotal Reactive Antioxidant Potential (TRAP) measuredas the ability to inhibit induced lipid peroxidation in redblood cell suspensions provides this class of informationabout the system.14 The overproduction of precursors toreactive oxygen radicals and decreased efficiency of in-hibitory and scavenger systems may lead to an increasedplasma lipoprotein oxidation. It has been suggested thatlow density lipoprotein (LDL) oxidation contributes toLDL accumulation and, therefore, the accelerated devel-opment of atherosclerosis in diabetes.15–17 Recent stud-ies have shown that the modification of LDL resultingfrom glycation and subsequent oxidation alters macro-phage ceroid accumulation, a mechanism occurring inhuman arterosclerotic plaque.18

Most research of oxidative stress in diabetic patientsAddress correspondence to: H. A. Farach, University of South Caro-

lina, Department of Physics and Anatomy, Columbia, SC 29208.

Free Radical Biology & Medicine, Vol. 24, No. 4, pp. 580–585, 1998Copyright © 1998 Elsevier Science Inc.Printed in the USA. All rights reserved

0891-5849/98 $19.001 .00

PII S0891-5849(97)00293-1

580

has concentrated on the investigation of the individualoxidation parameters that are involved with the highestmethodological level. To acquire an overall prospect,and because these studies are beyond our capabilities, wedesigned the present prospect, providing informationabout the possible relations between the oxidative phe-nomena and the variations in metabolism characteristicof the disease.

The aim of this study was to evaluate the status andinterrelationships of oxidative damage, total reactive an-tioxidant potential, and glycation products level in bloodsamples from elderly noninsulin-dependent diabetesmellitus (NIDDM) patients.

MATERIAL AND METHODS

Patients

The studies were performed on 25 male patientswith noninsulin-dependent diabetes mellitus, rangingin age from 44 to 75 years (mean 61.96 6.1 years),who were diagnosed and treated in the Hospital deClınicas ‘‘Jose´ de San Martı´n.’’ All patients wereunder treatment with sulphonylureas. Fifteen healthyand age-matched male donors served as controls(mean age 56.36 4.7 years). Clinical data of patientsand controls are shown in Table 1.

Chemicals

All reagents were of analytical grade. Butyl hydroper-oxide, sodium azide, sodium dodecyl sulphate, ando-phenanthroline were purchased from Sigma, thiobarbitu-ric acid (TBA) was purchased from Fluka, and potassiumiodide was purchased from Merck.

Routine biochemical analyses were performed usinghigh-performance standardized kits from Boehringer–Manheim Argentina. Ultraviolet absorbancy measure-ments were made using a Shimadzu UV 210A doublebeam spectrophotometer.

Samples

Blood samples were drawn after an overnight fast.Samples were centrifuged at 3000 rps for 5 min usingheparin as anticoagulant. Plasma and buffy coat wereremoved and erythrocytes were then washed three timeswith buffer phosphate pH: 7.4 NaCl 0.85%. Packed cellsand plasma samples of each patient were subjected tospecific tests (E-1; P-1; S-1, S-2, and S-3) which aredescribed as follows.

E-1 Oxidative treatment of erythrocytes

Erythrocytes were incubated 1 h at 37°C with tert-butyl hydroperoxide (t-BOOH) 3 mM final concentrationin the presence of 2.5 mM sodium azide. Aliquots of theincubate were taken every 15 min, deproteinized byadding HC1 1 N, TBA 20% and centrifuged at 2000 rpmfor 10 min. Supernatants were removed and their thio-barbituric acid reaction product content was tested (E-2).

E-2 Quantification of lipid peroxidation

Thiobarbituric acid reactive substances (TBARS)were measured as an index of the oxidative insult of theRBC by thet-BOOH incubation. Sodium dodecyl sul-phate 10%, butylated hydroxyl toluene 13.4 mM in eth-anol and TBA 0.7% were added to the aliquots of thedeproteinized supernatants. Mixtures were incubated 15min in a boiling water bath, and after cooling the absor-bance at 532 nm was recorded. Results are expressed asnm TBARS/ml packed RBCs. As standard of the coloredcomplex we used the complex TBA-1, 1, 2, 2-tethrae-thoxypropane.

P-1 Total reactive antioxidant potential

The reaction was carried out in a manner parallel tothe study of the oxidation kinetics of red blood cells andconsisted in inducing peroxidation of the erythrocytes byt-BOOH and the subsequent evaluation of TBARS in thepresence of the plasma aliquot under study, in the vol-ume ratio 3:1 (packed RBCs/plasma). The antioxidantpotential was calculated using the expression:

TRAP ~%! 5 100 F1 2Ap~t! 2 Ap~t0!

A~t! 2 A~t0!G

Table 1. Plasmatic Controls of Lipemia and Routine for the Diabeticand Control Groups: Triglycerides, Creatinin, Total Cholesterol,

HDL Cholesterol, LDL Cholesterol, Total Lipid and Total Protein

Diabetes(NIDDM)(n 5 25)

Control(n 5 15)

Age, years 61.96 6.1 56.36 4.7Hematocrit % 45.66 3.0 43.16 2.3Creatinin, mg/dl 0.936 0.29 0.996 0.3Triglycerides, mg/dl 141.26 76.8 150.16 67.8Cholesterol T, mg/dl 214.26 59.8 194.96 47.7HDL Cholesterol, mg/dl 47.96 18.6 43.56 13.6Lipids T, mg/dl 7.226 2.04 6.746 2.00LDL cholesterol, mg/dl 139.46 54.0 121.06 46.3Proteins T, mg/dl 7.466 0.73 7.386 0.66Uric acid, mg/dl 5.386 0.39 5.326 0.33Glycemia, mg/dl 1426 15* 796 7Fructosamine, mg/dl 6496 47† 3046 40

Results are expressed as mean6 SD *p , .002 against the controlvalues;†p , 0.001 against the control values.

581Oxidative damage, plasma antioxidants

where

A 5 mean absorbance at 532 nm in the absence ofplasma

Ap 5 mean absorbance at 532 nm in the presence ofplasma

t 5 incubation timet0 5 initial time

A calibration curve was made utilizing Trolox C asinhibitor of the oxidation of the red cells. Results areexpressed as equivalent concentration to Trolox (mM).

S-1 Selective precipitation of low-density lipoproteins

The LDL was precipitated from an aliquot of serumby aggregation with sodium citrate buffer 64 mM, hep-arin 50,000 UI/l, final pH 5.11 and subsequent centrifu-gation at 4,000 rpm for 10 mins. The supernatant liquidwas discarded leaving behind the high-density lipid(HDL) and very low-density lipid (VLDL) fractions.19

The lipids of the precipitate were extracted using a mix-ture of chloroform methanol 2:1, butylhydroxy tolueneBHT 0.002% (P-3).

S-2 Peroxide determination

The solvent used was deoxygenated absolute etha-nol bubbled with nitrogen for 5 min. A solution ofaluminum– chloride (2 g%),o-phenanthroline (0.02g%) in nitrogen-bubbled and hydrous ethanol was keptat 4°C until use. Under these conditions the solution isstable for 30 days. Potassium iodide solution (0.2g/10ml nitrogen-bubbled anhydrous ethanol) was pre-pared on the day of use.

The solvent of the lipid extract was evaporated invacuum at a temperature below 25°C. Residues wereredissolved with 50ml of absolute ethanol, and 50ml ofthe solution of aluminum–chloride–o-phenanthrolineand 50ml of the solution of KI were added. The mixtureswere strongly mixed and the tubes were incubated at37°C for 15 min, in darkness. After incubation, 1 mlethanol was added and the absorbance was recorded at357 nm. Simultaneously to each set of determinations astandard curve was obtained using cumin–hydroperox-ide as standard.

S-3 Phospholipid determination

The lipid extract in chloroform–methanol was dried ina bath at 37°C in a current of warm air. It was redissolvedin chloroform and added to a solution of ammoniumferrothiocyanate (27.03 g FeCl3 6H2O 1 30.4 g

NH4SCN in 1 l H2O). It was mixed for 1 min andcentrifuged at low speed to separate the chloroformphase whose absorbance at 472 nm was recorded.20

Statistical analysis

Statistical analysis was performed using Student’st-test and Pearson correlation coefficient. All data arereported as Mean6 SD. A p-value,.05 was consideredto be significant.

RESULTS

The biochemical controls glycemia, fructosamine,uric acid, triglycerides, creatinin, total cholesterol, HDLcholesterol, LDL cholesterol, total lipids, and proteins ofboth the diabetic and the control groups are shown inTable 1. The hematological control routine exhibited novariations so the data are not shown.

Table 1 shows a statistically significant increase of thevalue of glucose (p , .002) and fructosamine (p , .001)in fasting NIDDM patients with respect to their controls.The reported levels of fructosamine are high relative tothose reported in the literature, including those in thecontrol group. This could be explained by an over esti-mation of the values due to the method employed. Theoxidative parameters are evaluated in Figs. 1 to 4.

Basal oxidation products and kinetics of RBCt-BOOHinduced peroxidation are shown in Figs. 1 and 2. Theerythrocytes of diabetic patients show greater basal oxida-tion products (p , .05). The susceptibility of the diabeticerythrocytes to oxidation injury was also shown to increasein the oxidation induced byt-BOOH, which is made clearnot only by reaching higher levels of lipoperoxidation prod-

Fig. 1. Basal levels of RBCt-BOOH peroxidation products. Endoge-nous oxidative products were measured in diabetic and control RBC bythe thiobarbituric acid assay. Results are expressed as nM TBARS/mlpacked RBC. Bars are mean6 SD.

582 F. AGUIRRE et al.

ucts shown in Fig. 2, but also with kinetic oxidation reach-ing a plateau more rapidly. Total Reactive AntioxidantPotential of the plasma of diabetics is much reduced relativeto the control, as shown in Fig. 3. In contrast to this,uricemia (the uric acid is considered as one of the mostimportant plasmatic antioxidants) exhibited no changes(Table 1). The ratio lipohydroperoxide/phospholipid in theLDL fraction of the plasma lipoproteins was found todecrease in DM patients (Fig. 4).

To determine possible relations between the evaluatedoxidative parameters and the biochemical control param-eters of the patients, linear regression studies were car-

ried out between the NIDDM and control groups. Table2 presents a summary of these linear regression studies.TRAP (Fig. 3) correlated inversely with the basal per-oxidation damage of the erythrocytes (r 5 2.487,p ,.05) in the diabetic patients, but not with those in thecontrol group. This suggest a relation between the circu-lating erythrocyte oxidation damage together with a de-crease in the activity of the antioxidant systems.

Among the biochemical control parameters, the plas-matic reaction antioxidant potential was associated directlywith the uric acid (r 5 .433,p , .05) and inversely with thefructosamine and with the glucose (r 5 2.655,p , .03 andr 5 2.495,p , .05, respectively) in patients with NIDDM,but not in the controls. The levels of fructosamine werefound to be related to the basal damage of the red bloodcells (direct correlationr 5 .856,p , .001).

DISCUSSION

Erythrocytes are especially susceptible to oxidativedamage due to the high degree of polyunsaturated fattyacids in their membranes, and the high concentration ofintracellular oxygen and hemoglobin whose redox chem-

Fig. 2. t-Butyl hydroperoxide induced diabetic and control RBC per-oxidation. The kinetics of erythrocytet-BOOH induced peroxidationwas measured in diabetic and control RBC by the thiobarbituric acidassay. Results are expressed as nM TBARS/ml packed RBC. Bars aremean6 SD.

Fig. 3. Total reactive antioxidant potential from diabetic and controlplasma. Plasma antioxidant properties were measured in diabetic andcontrol samples by the inhibition of thet-BOOH induced peroxidativedamage in RBC. Results are expressed as the equivalent Trolox con-centration of the percentage inhibition oft-BOOH induced oxidativedamage to RBC. Bars are mean6 SD.

Fig. 4. Lipohydroperoxide content of low-density lipoproteins wasmeasured in diabetic and control plasma samples. Results are expressedas nm hydroperoxides/mg phospholipids. Bars are means6 SD.

Table 2. Statistically Significant Correlation Studies of LinearRegression Between the Biochemical and Oxidative

Parameters of NIDDM Patients

Dependent Variable Independent Variable r p

Antioxidant Potential Uric Acid 0.433 , 0.05Antioxidant Potential Fructosamine 20.655 , 0.03Antioxidant Potential Glucose 20.495 , 0.05Antioxidant Potential Peroxidation of RBC 20.487 , 0.05Fructosamine Peroxidation of RBC 0.856 , 0.001Fructosamine Glucose 0.703 , 0.01

583Oxidative damage, plasma antioxidants

istry is known to produce oxyradicals. On the other handthe relative differences in oxyradical concentration be-tween the tissues and the blood generates a gradient ofthe reactive form of oxygen which, when uncontrolled bythe many plasmatic antioxidant mechanisms, may initiateoxidative chain reactions capable of damaging blood andvascular elements. For this reason we undertook thestudy of the complex and interrelated antioxidant sys-tems under different pathological conditions. In manycases it is extremely difficult to evaluate the action ofeach participating element contributing to plasma anti-oxidant activity, and it is clinically more useful to eval-uate the overall system.

NIDDM was selected as the pathological state for thisreport. Advances in the physiopathology research haveidentified several examples of free radical reactions in thediabetic state. Oberley (1988) and other authors reportedraised levels of lipohydroperoxides in the plasma of diabeticpatients.1 Peripheral vascular disease is a major factor indiabetic complications.21 Other health problems seen indiabetics are renal failure, coronary heart disease, blindness,infections, and arterosclerosis.16,18,22There is little doubtabout the primary causal factor of the development of mostdiabetic complications; it is thought to be the prolongedexposure to hyperglycemia and the consequent nonenzy-matic, posttransitional modification of proteins resultingfrom chemical reaction between glucose and primary aminogroups of proteins: glycation.9–12,21,22The evidence, eventhough largely indirect, suggests that the oxygen free radi-cals and the peroxy-free radicals are produced during theslow oxidation of the glucose with formation of Amadoriproducts and during the oxidation of other molecules suchas ascorbic acid, arachidonic acid, all of which are pro-moted in the hyperglucemic state. Alteration in energymetabolism or changes in the levels of mediators of theinflammation may also take place in the diabetic state.2 Inaddition to the origin of the oxyradicals, and excess oftransition metals that characterizes diabetes serves to cata-lyze and enhance oxidative damage through Fenton andHaber–Weiss type of reaction.23,24

Our results are in accordance with those of previousfindings. Figures 1 and 2 show the clear increase in thesusceptibility of the erythrocytes of diabetic patients tooxidative stress.11,21–25

The measure of TRAP, achieved by reaching theplateau of the oxidation curve gives and idea of the totalpotential of the plasma antioxidation system under studyrather than the relative efficiency of the measured totalantioxidant reactivity.26

In Fig. 3 we see a pronounced decrease in the anti-oxidant potential in the plasma of diabetic patients. Inspite of the reports that point to uric acid as one of themain components of the various antioxidant defense sys-tems, these parameters are statistically associated

through linear regression in the diabetic group (r 5 .433,p , .05; Table 2), the statistically significant decrease ofthe TRAP is not accompanied by a decrease in theplasmatic levels of uric acid of diabetic patients vs.controls (Table 1). These results suggest that in spite ofbeing an important element of the plasmatic antioxidantdefense systems, uric acid is not significantly altered indiabetes.

Diabetes is a disease accompanied by a marked in-crease in the risk of pathological cardiovascular arterio-sclerosis. The LDL lipoprotein factor has been recentlyclaimed to be responsible in large measure for the in-creased arterogenesis of diabetes.23 The mechanism thatincreases the risk of arteriosclerosis and its complica-tions in diabetes has not been established. There aremany aspects of cardiovascular risk that coexist in pa-tients with noninsulin-dependent diabetes mellitus, andamong them are obesity, insulin resistance, hypertension,family history of premature arteriosclerotic disease, anddislipemia. This last is characterized by hyperglyceri-demia (through the accumulation of VLDL), low levelsof HDL and the presence of modified LDL, either gly-cated or oxidized. in recent years there has been increas-ing interest in the potential role of oxidatively modifiedlipoproteins in the mechanisms leading to arteriosclero-sis, and many studies show that the nonspecific oxidationof LDL is a factor involved in the development ofartherogenic plaque.24

The oxidized LDL possesses various biological prop-erties that can promote artherogenesis, including theadhesion of monocytes to the endothelial cells, chemo-taxis, cytotoxicity, modulation of the expression ofgrowth factors, and the capability for the sequestration ofmacrophages.

It has also been reported that the nonspecific oxidation oflow-density lipoproteins increases the rigidity of the eryth-rocyte plasma membrane and of the endothelial cells thatalter their role in cholesterol transport.13 In this manner theincrease in lipohydroperoxide in the LDL fraction of dia-betic patients (Fig. 4) can be related with the precociousatherogenesis associated with the progress of the disease.

It can be supposed that the decrease of the antioxidantplasmatic capacity of the diabetic patients (Fig. 3) is inpart responsible for the major oxidation damage sufferedby the LDL plasmatic fraction, namely those who havehigher levels of hydroperoxides relative to the controlsubjects (Fig. 4), but these two parameters were notshown to be directly related in the linear regressionstudies (Table 2). Further studies are necessary to deter-mine if the oxidative damage suffered by the LDL pa-tients is based on alterations of this antioxidant mecha-nism, which were not evaluated in the present study (e.g.,ceruloplasmin, transferrin, etc.). On the contrary, a de-crease of the value of the glycosilation of the plasmatic

584 F. AGUIRRE et al.

antioxidant defense is directly associated with a largeoxidative insult to erythrocytes (p , .001, Table 2).

The greater the plasmatic fructosomine or glucosecontent the less the blood total reactive antioxidant po-tential (Table 2). These results suggest that diabetes is analtered metabolic state of oxidation-reduction and that itis convenient to give therapeutic interventions with an-tioxidants. The loss of antioxidant capacity is statisticallyassociated with accelerated aging processes in diabeticpatients, due to an increase in basal oxidation products oferythrocytes associated with monosaccharide autoxida-tive glycation.1,2,4,5,7,9,21,27,28,29Our present work takesinto account the hyphothesis involving the relation be-tween individual components in the intact clinical modeland the complex peroxidant–antioxidant plasmatic sys-temic balance. It remains to investigate the possiblecauses of the decrease in plasmatic defense, its effects onoxidative imbalance, and its effects on the production ofthe oxyradicals.

With respect to mechanisms that promote oxidativechanges or accelerate the secondary complications ofdiabetes, there is a lack of studies that attempt to clarifythe relations between the individual components of thecomplex peroxidant–antioxidant plasmatic system in theintact organism.

REFERENCES

1. Oberley, L. W. Free radicals adn diabetes.Free Radic. Biol. Med.5:113–124; 1988.

2. Jain, S. K.; Palmer, M. The effect of oxygen radicals metabolitesand vitamin E on glycosylation of proteins.Free Radic. Biol. Med.22:593–596; 1997.

3. Brownlee, M. Glycation and diabetic complications.Diabetes43:836–841; 1994.

4. Baynes, J. W. Role of oxidative stress in development of compli-cations in diabetes.Diabetes40:405–412; 1991.

5. Brownlee, M. Glycation products and the pathogenesis of diabeticcomplications.Diabetes Care15:1835–1843; 1992.

6. Ghiselli, A.; Laurenti, O.; De Mattia, G.; Maiani, G.; Ferro Luzzi,A. Salicylate hydroxylation as an early marker of in vivo oxidativestress in diabetic patients.Free Radic. Biol. Med.13:621–626;1992.

7. Wolff, S. P. diabetes mellitus and free radicals. Free radicals,transition metals and oxidative stress in the etiology of diabetesmellitus and complications.Br. Med. Bull.49:642–652; 1993.

8. Mooradian, A. D. Increased serum conjugated dienes in elderlydiabetic patients.J. Am. Geriatr. Soc.39:571–574; 1991.

9. Vlassara, H.; Bucala, R.; Striker, L. Pathogenic effects of advancedglycosylation: Biochemical, biologic and clinical implications fordiabetes and aging.Lab Investigat.70:138–151; 1994.

10. Tsai, E. C.; Hirsch, I. B.; Brunzell, J. D.; Chait, A. Reduced plasmaperoxyl radical trapping capacity and increased susceptibility ofLDL to oxidation in poorly controlled IDDM.Diabetes43:1010–1014; 1994.

11. Asayama, K.; Uchida, N.; Nakane, T.; Hayashibe, H.; Dobashi,K.;. Ameniya, S.; Kato, K.; Nakazawa, S. Antioxidants in theserum of children with insulin-dependent diabetes mellitus.FreeRadic. Biol. Med.15:597–602; 1993.

12. Ookawara, T.; Kawamura, N.; Kitagawa, Y.; Taniguchi, N. Site-specific and random fragmentation of Cu, Zn-superoxide dis-mutase by glycation reaction. Implication of reactive oxygen spe-cies.J. Biol. Chem.267:18505–18510; 1992.

13. Urano, S.; Hashi-Hashizuma, M.; Tochigi, N.; Matsuo, M.;Shiraki, M.; Ito, H. Vitamin E and the susceptibility of erythrocytesand reconstitued liposomes to oxidative stress in aged diabetics.Lipids 26:58–61; 1991.

14. Lissi, E. A.; Salim-Hanna, M.; Pascual, C.; del Castillo, M. D.Evaluation of total antioxidant potential (TRAP) and total antiox-idant reactivity from luminol enhanced chemiluminiscence mea-surements.Free Radic. Biol. Med.18:153–156; 1995.

15. Panasenko, O. M.; Vol’Nova, T. V.; Azizova, O. A.; Vladimirov,Y. Free radical modification of lipoproteins and cholesterol accu-mulation in cells upon atherosclerosis.Free Radic. Biol. Med.10:137–148; 1991.

16. Lopez Virella, M.; Virella, G. Immune mechanisms of atheroscle-rosis in diabetes mellitus.Diabetes41(Suppl. 2):86–91; 1992.

17. Sevanian, A.; Hochstein, P. Mechanism adn consequences of lipidperoxidation in biological systems.Annu. Rev. Nutr.5:365–390;1985.

18. Hunt, C. C.; Bottoms, M. A.; Clare K., Skamarauskas, J. T.;Mitchinson, M. J. Glucose oxidation and low-density lipopro-tein-induced macrophage Ceriod accumulation: Possible impli-cations for diabetic atherosclerosis.Biochem J.300:243–249;1994.

19. Weiland, H.; Seidel, D. A simple specific method for precipitationof low density lipoproteins.J. Lipid Res.24:904; 1983.

20. Stewart, J. Ch. Colorimetric determination of phospholipids withammonium ferrothiocyanate.Analy. Biochem.104:1014; 1980.

21. Wolff, S.; Jiang, Z.; Hunt, J. Protein glycation and oxidative stressin Diabetes Mellitus and aging.Free Radic. Biol. Med.10:339–352; 1991.

22. Dyer, D.; Dunn, J.; Thorpe, S.; Lyons, T. Accumulation of Mail-lard Reaction Products in skin collagen in diabetes and aging.Ann.NY Acad. Sci.663:421–422; 1992.

23. Cadenas, E. Oxidative stress and formation of excited species. InSies, H., ed.Oxidative stress.London: Academic Press; 1985.

24. Miller, D. M.; Buettner, G. R.; Aust, S. D. Transition metals ascatalysts of ‘‘autoxidation’’ reactions.Free Radic. Biol. Med.8:95–108; 1990.

25. Schwartz, R. S.; Madsen, J. W.; Rybicki, A. C.; Nagel, R. L.Oxidation of spectrin and deformability defects in diabetic eryth-rocytes.Diabetes40:701–708; 1991.

26. Aguirre, F.; Hager, A.; Ilho, J.; Martı´n, Y.; Facorro, G.; de Paoli,T., Propuesta para el estudio de las defenas antioxidantes sanguı´n-eas, en Reunio´n Anual de la Sociedad Argentina de Biofisica(SAB); 1992.

27. Sevanian, A.; Hwang, J.; Hodis, H. Antioxidant intervention ofatherosclerosis: Action on oxidatively modified low density li-poproteins (LDL). In: Abstract Book of VII Biennial MeetingInternational Society for Free Radical Research; 1996; 22.

28. Giugliano, D.; Ceriello, A.; Paolisso, G. Oxidative stress anddiabetic vascular complications.Diabetes Care19:257–267; 1996.

29. Jain, S. K.; Levine, S. N.; Duett, J.; Hollier, B. Reduced vitamin Eand increased lipofucsin products in erythrocytes of diabetic rats.Diabetes40:1241–1244; 1991.

585Oxidative damage, plasma antioxidants