Prvi Rad - ABTS

07019901

Clinical &lence (1993) N, 407-412 (Printed in Great Britain)

A novel method for measuring antioxidant capacity andits application to monitoring the antioxidant status inpremature neonatesNicholas 1 . MILLER', Catherine RICE-EVANS', Michael J . DAVIES2, Vimala GOPINATHAN' andAnthony MILNER''Free Radical Research Group, Division o( Biochemistry, UMDS-Guy's Campus, London, U .K.,'Department of Chemistry, University of York, York, U .K ., and aDeportment of Paediatrics,St Thomas' Hospitul, London, U .K.

(Reteived 15 June 19948 January 1993 ; accepted 20 January 1993)

1. A new method has been developed for measuringthe total antioxidant capacity of body fluids and dmfsolutions, based on the absorbance of the ABTS'radical cation .2. An automated method for use on a centrifugalanalyser, as well as a manual method, Is described.3. The procedure has been applied to physiologicalantioxidant compounds and radical-scavenging drugs,and an antioxidant ranking was established based ontheir reactivity relative to a 1 .0mmolfl Troloxstandard.4. The Trolox equivalent antioxidant capacity ofplasma from an adult reference population has beenmeasured, and the method optimized and validated .5. The method has been applied to investigate thetotal plasma antioxidant capacity of neonates andhow this may be compromised in prematurity.

INTRODUCTIONOn exposure to hydrogen peroxide, metmyo-

globin and methaemoglobin are activated to ferrylstates [1, 2] in which the iron is one oxidizingequivalent above the original level and one oxidiz-ing equivalent is on the surface of the protein [1] .With reducing agents (electron or hydrogen donors)these species are reduced back to metmyoglobin ormethaemoglobin [2-4].

In this study we have exploited the peroxidaseactivity of metmyoglobin combined with its inter-action with a phenothiazine compound to form aradical cation intermediate [5], to establish amethod for measuring antioxidant status [6] . Themethod derives from the observation that when 2,2'-azinobis-(3-etbylbenzothiazoline-6-sulphonie acid)(ABTS) is incubated with a peroxidase (such asmetmyoglobin) and hydrogen peroxide, the rela-tively long-lived radical cation, ABTS'* [7], is

407

formed A large number of free radicals, such ashydroxyl, peroxyl, alkoxyl and inorganic radicals,also react rapidly with ABTS to form this species[8]. When the peroxidase is metmyoglobin, theformation of the ABTS'~ radical cation on inter-action with fetryl myoglobin has absorption max-ima at 650, 734 and 820nm, beyond the region ofabsorption of the haem proteins. In the presence ofantioxidant reductants and hydrogen donors, inpure solution or in plasma, the absorbance of thisradical cation is quenched to an extent related tothe antioxidant capacity of the added fluid (in theabsence of interfering compounds) .

The major antioxidant defences in plasma includeascorbate, protein thiols, bilirubin, urate and a-tocopherol. The 'chain-breaking' or 'radical-scavenging' agents against oxidative stress are con-sidered to act in the above sequence of increasing todecreasing effectiveness against free radicals gener-ated in the plasma aqueous phase [9] . Plasma alsocontains the 'preventive antioxidants, caerulo-plasmin and transferrin, the iron-scavenging pro-teins whose contribution to the total antioxidantcapacity is to prevent iron availability [10] .

The limited ability of pre-term infants to copewith oxidative stress may contribute to the develop-ment of broncho-pulmonary dysplasia, necrotizingenterocolitis and retrolental fibroplasia . The possiblerole of antioxidants in protecting against oxygentoxicity has been inconclusively investigated for anumber of y'eaxs . Pre-tarm infants have incompletelydeveloped antioxidant defences and suffer from adeficiency of vitamin E, which is normally derivedat the ead of the third trimester from the maternalcirculation .

Applying this method, the total plasma anti-oxidant status of pre-term babies has been moni-tored and compared with that of term babies andmaternal levels . The results show that infants bornat 27 f 2 weeks have a significantly depressed total

Key vror-0e antloxldaM, 7.t'~bnh0{3-ethylhenaathiamlinedaulphenir add), ferryl myotbbin, free radinis, premature neonata, Trobr .Abhredations: ABTS, 7,1'ealnohis{3ethylbenmohluollnedaulphonic uid) ; TEAC, Trolox equivalent antlaxldant Upacity ; TRAP, totel perevyl redinkrapplnt entlavidantparameter.Correrpondence : Dr Gtherlne Rlcefvan,, Free Radical Research Group, Divelon of Biahemlrtry, UMDS-Guy's Campm, St Thoma Strest, Lendon SEI 9RT, U .K .

PM3006722631http://legacy.library.ucsf.edu/tid/qqc07a00/pdf

07019901

4118 N. ). Miller a al.

plasma antioxidant activity at birth . The differentialvalues between pre-term and term babies are re-flected in the maternal levels, supporting not onlythe need for antioxidant supplementation of pre-term babies from birth, but also the monitoring ofmaternal blood levels during pregnancy .

MATERIALS AND METHODS

MaterialsMyoglobin type III, iron in ferric state, irascorbic

acid, uric acid, albumin (human fraction V), a-tocopherol, glutathione, bilirubin, EDTA (disodiumsalt), heparin (sodium salt) and Sephadex G-15-120(40-120µm), were purchased from Sigma; ABTS(diammonium salt) and Trolox (registered trade-mark of Hotfinan-LaRoche) were purchased fromAldrich; hydrogen peroxide, disodium hydrogenorthophosphate 12-hydrate, sodium dihydrogenorthophosphate dihydrate, sodium chloride, urea,creatinine, glucose, mannitol and potassium ferri-cyanide were obtained from BDH . Desferrioxamine(Desferal) was obtained from Ciba-Geigy . Freeze-dried serum quality control pools were obtainedfrom Biostat.

Metmyoglobin was purified after adding the stockmyoglobin solution (400µmol/l) in 5aunol/l isotonicPBS (pH 7.4) to an equal volume of freshly prepared740µmo1/1 potassium ferricyanide ; after mixing, thesolution was passed through a G15-120 Sephadexcolumn equilibrated in the buffer, and the metmyo-globin fraction was collected . The final concen-tration of purified metmyoglobin was estimated byapplying the Whitburn equations:

[Met Mb] = 146A4se-108Asse+ 2.lAseo[Ferryl Mb]=-62Aaso+242A,so-123A.eo[Mb02] = 2.8A4so-127Asao+153Aseo

where Mb is myoglobin. These equations [11] arederived by solving simultaneous equations based onBeer's law, measuring the absorbance at 490, 560and 580nm, and subtracting the reading at 700nmto correct for background absorbance . The purity ofthe metmyoglobin prepared was estimated by apply-ing all three equations . Normally the metmyoglobinfraction is >95% of the total haem protein .A 2.5mmo1/I Trolox solution was prepared by

dissolving 0.15643g of Trolox in 250m1 of buffer .Dissolution was facilitated by gentle sonication. a-Tocopherol was solubilized for introduction into theassay system by preparing an emulsion in 60%ethanol/40% water containing 2% Nonidet P-40.ABTS was prepared as a 5mmo1/1 solution by

dissolving 0.02743g in 10.0ml of buffer; 500(tmol/1ABTS was prepared from this stock solution for useas a working reagent. Hydrogen peroxide wasdiluted to a 0.45mmol/1 working solution .

Spsetroscopic assay

ARTS (300µ1, 500yanol/1), 36yd of metmyoglobin(70µmol/I) and 497µ1 of buffer (of which 8 .4µ1 wasreplaced when a sample was being investigated)were mixed, and the reaction was initiated by theaddition of 167µ1 of hydrogen peroxide (450µmol/I),giving final concentrations of 2 .5 µmol/I metmyo-globin, 150µmo1/1 ABTS amd 75pmol/1 hydrogenperoxide. Spectra were recorded (see Fig . 1) at 90sintervals for 12 min (450-900 nm), and the absor-bance at 734nm was measured as a function oftime.

t.entrifugal analyser assay

For the assay to be carried out on the RocheCobas Bio centrifugal analyser, the final incubationvolume in the ouvette and the method of makingadditions required that reagents be made up inconcentrations that suited the configuration of thisinstrument. In the protocol devised, 300µ.1 of ABTS/myoglobin reagent was mixed with 3pl of sampleand a further 30p1 of diluent that flushed thesample probe. The initiator of the reaction, hydro-gen peroxide (1 .078mmo1/k), was added last (25µ1)to give a final concentration in the cuvettc of75Fmto1/1 and a final incubation volume of 358p1.

A quantitative relationship exists between theabsorbance at 734 nm at 6 min and the antioxidantconcentrations of the added sample or standard.After 6min the absorbance of the solution was readat 734nm, along with a buffer blank that did notcontain the antioxidant solution or fluid to betested . The absorbance value of the blank at 734nmat 6min minus that for the test solution, divided bythe blank value, was the fractional inhibition of thereaction.

Patients

Babies were recruited at the time of delivery whencord venous blood samples were collected. Furthervenous blood samples (0.5ml) were obtained on day5 in term babies at the time of the Guthrie test andin pre-term babies at the time of routine bloodsampling.Term babies (3"1 weeks gestation) weighed

2.20-3.92kg; pre-term babies (25-29 weeks ges-tation) weighed 0.73-1.46kg. Three of the prematurebabies subsequently died of their complications(intraventicular haemorrhage, group B streptococcalsepsis, necrotizing enterocolitis) and three developedbronchopulmonary dysplasia . The remaining tenpremature infants did well after their initial periodof ventilation. Each premature infant was treatedwith surfactant (Exosurf), which showed no activitywhen introduced into the antioxidant assay as asample, and penicillin and gentamicin, which bothshowed moderate antioxidant activity. None of


07019901

Automated mkaazurement of aetioxidam eapadcy 409

1.0 1.5 2.0 2.5Trolox sonm . (mmol/I)

Fig.1 Oegree of inhlbition of the ebsorbanee at 73/nm as a functlen of the eontentration of Trolox . Values aze means±SD (n ss 16) .

Flg . I . Visible epectra remrding the appearance of the ABTa'+ndlnl cation with time . Scos were run at 90s Imernb rommeneingat 15s. Condlelonc' 2.Spmol/I metmyoglobhln, 150µnoI/I AOTg and

75µmoI/I H,O, .

these substances could therefore act as negativeinterferents in the antioxidant assay .

RESU LTS

Incorporation of a range of individual anti-oxidants (initial concentration from 0.25 tol0.Ommol/I) before the addition of hydrogen perox-ide induced a delay in the appearance of theABTS'* radical cation and hence increased thepercentage inhibition of the absorbance at 734nm(Fig. 1) . All the following results described werederived from analyses carried out on the Cobas Biocentrifugal analyser.

Standardization of the assay

The system was standardized using Trolox, ana-tocopherol analogue with enhanced water solubi-hty. A 2.5mmo1/1 solution of Trolox was preparedin buffer . This solution is stable for 1 week whenstored at 4°C or for 6 months when stored at-20'C. Standards of lower concentration wereprepared fresh daily.

The results for dose-response curves derived from16 sequential and separately prepared stockstandards are summarized in Fig. 2. The SD of the0.5 and 2.5mmo1/I points is less than 2 .0, and sofalls within the area of the plotted point. The dose-response curve is thus highly reproducible .

Units of mmsurement: Trolox equivalent antioxidantcapacity (TEAC)

The unit of activity is the TEAC, which is definedas the concentration (mmol/1) of Trolox having theequivalent antioxidant capacity to a 1 .0mmo1/1 solu-tion of the substance under investigation. The deri-

495 540 585 630 675 720 ~ 165 810. Wawlengeh (nm)

855 900

Table I . TEAC .alues

TEAC n 8D

DeAerrioxemine 7.96 4 0 .09

gllirubin 1 .50 3 0I3

Urae 1 .02 5 0,06Ascorbate 0.99 S 0.04

s70copharal 0.97 3 0.01

Glutathione 0 .90 3 0.03

Albumin 0 .63 3 0.02

EDTA 0 .05 3 0.01

Mannitol 0.00 3

Gluaose 0.00 2Edanol 0.00 3Heparin 0.W 5Ura 0A0 3Creatinine OA0 3

vation of a TEAC value provides a method forcomparison of antioxidant activity among groups ofdrugs and chemicals, provided that they are water-soluble or can be solubilized. The results are shownin Table 1 . Whereas urate, a-tocopherol and ascor-bate are as effective as Trolox in their antioxidantactivity, desferrioxantine is three times more effectiveas a scavenger. Bilirubin is more efficacious thanthese compounds, whereas albumin is considerablyless active .

Total antioxidant activity of normal human plasma

Incorporation of plasma from an adult referencepopulation as samples into the assay gave a 0 .95interfractile interval of 1.46±0.14mmol/1 (n=312) .

PrecisionTwo control pools were prepared from eommer-

cial freeze-dried quality control sera. The results forintra-assay precision when analysed repeatedly in asingle assay, or inter-assay precision, obtained by re-analysing the pools on 20 sequential days, areshown in Table 2 The data indicate the ability to


410 N. J . Miller et al.

Tahle L Estittuts d anatyaal perfarmaaee Table 3. Antbxidsnt hler.rchy in the plesnv water. A ranking of

Migh panl Low pad endo6anous amloaldants, baaad an TFAC x mid-polnt of the plasma refer-Bnte Imerval

I i intrrauay prec s onNa 25 25 Cunen. (µmol/I) Antioaidant

Mean ' 1 .51 0.83 attivhy

SD 0.008 0.013 TEAC Range Mld-pnint TEAC x (X of tmil

Coefficient of variation (%) 0,54 1 .59 mid-point plumaaancn. attlvlty)

Interassay precislon Albumin 0.63 535-760 640 403 43No. 30 20 Urtte 1.02 Ip0-n0 300 306 33Mean 1 .52 0.84 Afourbata 0.99 34-III 73 73 9SD 0.055 0.050 wTampharol 0.97 14-44 39 28 3CaaMclenr of verlatlan (X) 3 .6 6.1 Bilirubin 1.50 <10 10 IS 1

Unmeawred . 10andoddents

carry out the assay on a day-to-day basis with anacceptable degree of precision .

Interferences

The only true interfering substances in this assayare peroxidases, such as haem proteins, which, inthe presence of hydrogen peroxide, promote theformation of the ABTS" radical cation. Visiblehaemolysis thus makes plasma samples unsuitable,but a study of delay in centrifugation and of g forceapplied during centrifugation showed that moreminor levels of haemolysis do not affect the results .

Drugs and other exogenous materials with signifi-cant absorption at 734nrn might also be potentialsources of positive interference in the assay.

Specimens from neonates and their mothers

Applying the method to monitoring the totalplasma antioxidant activity of neonates, the findingsreveal that the total plasma antioxidant status ofpre-term infants at birth is significantly depressed(1.21t0.08mmol/l, n=16) compared with termbabies (1 .46±0.07mmo1/I, n=18) . After 5 days theantioxidant activity in the plasma of the prematureinfants had risen (1.41±0.09mmo1/1, n=14), where-as that of the term infants showed no significantchange. The differential values between pre-termand term babies were reflected in the related mater-nal blood levels of the small number of pair-matched samples studied so far [pre-term1.17t0.06mmo1/l (n=6) to maternal 1.25tO.08mmo1/1 (n=6); and term 1.46t0.07mmo1/I(n=18) to maternal 1 .41t0.07mmo1/1 (n=17)],demonstrating the influence of the maternal totalantioxidant levels .

DISCUSSION

The derivation of a TEAC value, as shown above,provides a method for the comparison of anti-oxidant activity among groups of drugs, providedthey are water-soluble (or can be solubilized), orbody fluids, such as plasma . The method thus has

important pharmacological and nutritionalapplications.

Applying this method, utate, a-tocopherol, ascor-bate and glutathione have the same molar equiva-lent antioxidant capacity as Trolox. Of the otherantioxidants found in human plasma, bilirubin is amore efficient antioxidant and albumin is less active,on a mol for mol basis. Desferrioxamine, the thera-peutic iron chelator, has been identified as ahydrogen-donating antioxidant to the superoxideradical [12], activated peroxidases [13] and ferrylmyoglobin radical species [3,4] . This compound hasa three-fold greater activity in the assay as com-pared with urate, ascorbate, ete, which is as wouldbe predicted on the basis of its trihydroxamatestructure. The value derived for bilirubin was of theorder of 1.5 Trolox equivalents . EDTA (10mmol/1),heparin (10000i .u.), urea (10mmo1/1), creatinine(10mmol/l), glucose (10mmo1/1) and ethanol (60%)were all unreactive in the assay at the concen-trations applied . (The results for EDTA and heparinindicate their lack of interference as anticoagulantsfor plasma studies.) Cysteine has a low TEAC valueand, under most circumstances, a very low concen-tration in the plasma . Glutathione is an activeantioxidant with a TEAC value equivalent to thatof urate and ascorbate. With one exchangeablehydrogen per mol of glutathione, this result is aspredicted. Glutathione is unlikely to play a signif3-c.ant role in the extracellular fluid . The extremelyhydrophobic nature of #-carotene, dihydrolipoateand ubiquinone makes it difficult to evaluate themin an aqueous system such as the one we havedescribed.

Although the plasma is not a simple chemicalsystem as regards antioxidant activity (the differentantioxidants may be active at specific sites andhence perform special functions) one can extrapolatefrom the above to a ranking of antioxidants in theplasma water . This is shown in Table 3 . The relativeefficacy of the contribution of each antioxidant doesnot define the actual importance of the antioxidantbecause of variations in the plasma concentrations .Albumin is first in this ranking, due to its relatively


07019901

Automated mmsuraman ot aaiazidan npaity 411

high concentration in the plasma water, eventhough it is a comparatively less efficient anti-oxidant. For purposes of comparison, albuminconcentrations in Table 3 are expressed in molarunits (applying an M, of 65000 [14]), although thisis not the common practice . Uric acid ranks secondto albumin, again due to its high concentration .More than 70% of the antioxidant activity of theplasma water is accounted for by these twosubstances.

The previously published [15] total peroxylradical-trapping antioxidant parameter (TRAP) isbased on the application of the thermal decom-position of the water-soluble azocompound 2,2'-azo-bis{2-amidinopropane hydrochloride) to peroxylradicals. The improved method from these workers[16] applied to plasma samples shows contributionsfrom urate (58±18%), plasma proteins (21t10'/),ascorbate (14t8%) and vitamin E(7±2%) toTRAP. The residual contribution after accountingfor urate and 'plasma proteins', in these authors'terminology, is equivalent to the residual anti-oxidant activity presented here after accounting foralbumin and urate.

Lindeman et. [17], applying the TRAP assay ofWayner et al. [16], have determined the percentagecontributions to the measured TRAP as : uric acid,47.4%; thiol (plasma proteins), 12.2%; vitamin C,13.4%; vitamin E, 6 .4%q unidentified substances,21.5%. Whereas uric acid, vitamin C and vitamin Eare reasonably dose to our estimate, some difer-ences are apparent. We deduce that the thiol mea-sure used by these workers is that in the plasmaproteins, i.e. mainly albumin, for which they quote aplasma concentration of 496{tmolJl. We have calcu-lated from a reference interval for albumin of 35-50g/l [18] (which converts to 535-760µmol/1), Theresulting underestimate could account for the highervalue quoted for unidentified compounds, as com-pared with our value of 10% .

Albumin is well recognized as a plasma expander,through its maintenance of the plasma oncoticpressure. It is also regarded as an important non-specific transporter of endogenous substances, drugsand toxins. The importance of albumin as a free-radical scavenger apart from these two other func-tions has been reported [19] . There is also someepidemiological evidence for its importance as anantioxidant: a consistent negative relationshipbetween albumin and coronary heart disease isknown [20]. The importance of uric acid is alsowell recognized [21] . These two substances may actas a general pool for the removal of radicals passedin the plasma water. Ascorbate itself accounts forjust under 10% of the total plasma antioxidantactivity, and its high renal and hepatic clearancemakes it unlikely that this value could be greatlyraised. Bilirubin, like uric acid, may prove to be animportant antioxidant in certain pathological situa-tions where levels in the plasma rise well above thereference interval . The remaining 10% of the total

plasma antioxidant activity is accounted for bysubstances such as cysteine and glutathione, whichare water-soluble, and ft-carotene, dihydrolipoateand ubiquinone, which are non-polar hydrophobicsubstances.

Pre-term infants have incompletely developedantioxidant defences and sulfer from a deficiency ofvitamin E, which is normally derived from thematernal circulation at the end of the third tri-mester. Karmazsin et a] . [22] have demonstrated areduction in the plasma antioxidant activity ofpremature babies, using the bovine brain homoge-nate methods of Stocks et al . [10] . Oxygen therapyreduces the levels of free-radical scavengers in theextraoJlular Fluid . It has been proposed that thisstate of affairs may predispose to free-radicaldamage. Vitamin E therapy has been shown to bebeneficial in this event [23] . However, two double-blind controlled studies failed to show that vitaminE administration reduced the incidence of broncho-pulmonary dysplasia in pre-term infants and mayindeed increase the babies' susceptibility to infectionby reducing the oxygen-dependent ability of neutro-phils to kill pathogenic bacteria [24] .

Other workers [17] have measured the TRAP ofcord blood from babies at 32 weeks gestationcompared with that of term babies and peripheralvenous plasma from adults ; their study revealed nosignificant differences, although this probablyrelated to the degree of prematurity of the neonates .Our results, using serial plasma samples from

babies of less than 32 weeks gestation and theirmothers, support the contention that such infantsare profoundly antioxidant-deficient at birth . Moni-toring maternal antioxidant levels during pregnancywhere indicated may also be ultimately beneficial tothe neonate.

Studies are in progress to identify patient groupsamong whom the measurement of total antioxidantactivity in body fluids may prove to be an import-ant prognostic or diagnostic guide . These includepatients with atherosclerosis, reperfusion injury,septic shock and diabetes.Thus we have presented an assay for measuring

total plasma antioxidant status, which can be auto-mated and which uses only 3,u1 of plasma . The newmethod may be marginally less rigorous than theTRAP assay as it involves a number of reactivespecies rather than just the chain-carrying peroxylspecies. However, the TRAP assay is manual andtime-consuming. The new method presented here isextremely simple and rapid to perform, includesmulti-point calibration of the dose-response curve,and is suitable for the analysis of large numbers ofsamples .

ACKNOWLEDGMENTS

We thank the British Technology Group forsupporting this study. Grateful thanks are also dueto Dr Brenda Slavin of the Department of Chemical


07019901

412 N. J. Miller et al .

Pathology, St Thomas' Hospital, for advice and formaking equipment available.

REFERENCES

I . George P, Irvine PH. The interaction between mNmyoglobin and hydrogenperoxide. BiachemJ 1951; B2.• 511-I5.

2. Howell BR. Wyngaarden JB. On the mechanism of peronidatlon of uric acldby haemeprotains . J Blol Chem 1960 ; 139: 3544-SD.

3 . Rice-Evans C, Okunade G, Khan R. The suppressioe of iron releate fromactivated myoglobin by physiological ele<tron donon and desferriosemine .Free Radlal Res Commun 1989; 7: 307-13.

4 . Turner JJO, Rlce-Evans C, Davies MJ, Newman ESR. The formation of treeradicals by urdiac myocytes under exldetive atress and the effects ofelectron-0onming drugs. Blothem J 1991 : 277: 767a1 .

5 . Fornst IS, Carr CJ, Urdin E . Phenothiavna and soructuraly relrted drup.Raven Prms: New York. 1974.

6 . Rice-Evans C, Davies MJ . U.K. Patent no. 91765025, 1991,7. Childs RE, Bxdeky WG . The steadystate klnetin of pero:;dare with

2,2'-aslno-0 :(3.thyl-bentthluulinedaulphonic ackl) as chromugen . BiochemJ1975 ; 16 : 93.

& Wugenden RS, Willson RL Radical catbns u reference chromagens In kineticstudies of oneelectron transfer reactions: pulse radlolysis sGSdies of2,2'-aalnoNs{3ethylbenzchiuqllneltulphanata) . J Chem Soc Perkin Trana1982; II : 805-12

9. Frel B, Stocker R, Ames gN. Anaoddent defences and lipid peroxidation Inhuman blood plasms. Proc Nad Acid Sd USA 1988 ; gg :974852.

10. Stocks J, Gutterldge MC, Sharp RJ, Dormandy I .. Atsay using brainhemogenzce for measuring the antloxldant activity ol biological fluids, Gin SdMol Med 1974; 47; 215-22.

11 . WRltburn KD, Shleh, Sellers RM, Hoffman MZ . Redox tnnsfarmatlore InPorrlmyglnbin induced by radleGen-generated free radicab In equeout solutlne .J Blol Chem 19g7; 257s 1860-9 .

12 . Davies MJ, Donkor R, Dunmr CA, Gee CA, Jonas 5, Wlllson RL .

Dnferrlexemine end suparexlde sree radioJs : formation of anentymedamaging nftroxide . BlochemJ 1987; 146 : 775-9.

13. Mareheuee KM, Fllter WD, Meson RP . The entymaic oxldatlan of desferal toa nitroxlde radical . FEBg Lett 1987; 222: 246-50.

14. He XM, Carter DC. Atomic nrucmre and chemistry of humen serumalbumin . Nmure (London) 1992 ; Mg: ID9-15.

IS. Barday LRC, Locke SJ, MacNell JM, Van Kettel J, Burton GW, Ingold KU .Aucoxidatlon of micelles and model membrme. Quanthative kineticmwsuremenn aen be made by ueing water-solubfe or lipid<eluble Initiatarswlth waterioluble or Ilpid .oluble chain-0reaking antsoxldants. J Am Ch .mSoc 1984 : 106 :2479d1 .

16 . Wayner DDM, Burton GW, Ingold KU, Barclay LRC, Locke $J, The relativewnsri6utlom of vitamin E, urate, umr6ete and protelns tu the toal permrylrtdlcalirappIng antioxidant aalvlty of human blood plasma Biethim BtephysAae 1987; 9gL 40B-19,

17. Lindemen JHN, Van Zaaren4robbenm D, SchrijverJ, Speak AJ,Poorthuls BHJM, Bcrger HM. The tord free radical trapping ability of cordNood plasma In preterm and term babies. Pediatr Res 1989;16; 20-4 .

I& Gowenlotk AH . Varley's practical clinical biechemistry . 6th ¢d. Oxford :Helnemann, 1990: 419 .

19. Greenough A, Greenall F, Gannu HR . Immediate eReca of albumin iMmionin III premature nwnagee. Arch Dls Chiklh 1908; 6g; 307-17,

7& Kuller LH, Ekhner JE. Orchard TJ, Grandks GA, M<Callum L, Tracy RP. Tgerelulon between aerum albumin levalt and rak of eeronery heart dReate inthe multiple risk factor Intervention trlal . Am) Epldemlol 1991 ; 134r 1266-77,

21 . Hoaluteln P, Hatch L, Sevenlan A. Uric add : funnlon and decerminatlon .1984 ; IOL- 162-6 .

22 . Kermmin L, Olah VA, Balla G, Mackay A . Serom antioxidant activity Inpremsture babla. Acts Pediatr Hung 1990 ; 30: 217-24 .

23. Ehrenkrant B, Ablow R, Warshaw W. Prevention of bradchnpulmomry,dysplaia with vhamin E administration during the acute stages of respiratorydi¢ress eyndrome, J Pedimr 1979 ; 95: g73-B.

24. Johmon L, Bowen F, Abbaai 5, at al . Relationship of prolonged pharmawloglcserum lerela of vi[amin E to the Incidence of se{uis and necrotleingencerecolitir in infante of binhweight 15Wg or ieu . Pediatrics 1985; 73:619-30 .


Documents

Prvi Rad - ABTS