Biochimica et Biophysica Acta xxx (2013) xxx–xxx

BBAGEN-27523; No. of pages: 7; 4C: 4

Contents lists available at SciVerse ScienceDirect

Biochimica et Biophysica Acta

j ourna l homepage: www.e lsev ie r .com/ locate /bbagen

Oxidized albumin. The long way of a protein of uncertain function☆

Maurizio Bruschi b, Giovanni Candiano a, Laura Santucci b, Gian Marco Ghiggeri a,⁎a Division of Nephrology, Dialysis, and Transplantation, Istituto Giannina Gaslini, Genoa, Italyb Laboratory on Pathophysiology of Uremia, Istituto Giannina Gaslini, Genoa, Italy

Abbreviations:HOCI, hypochlorous acid; AOPPs, advanelectrospray ionization mass spectrometry; LC–MS–MS, lidem mass spectrometry; DSC, differential scanning calori☆ This article is part of a Special Issue entitled Serum⁎ Corresponding author at: Nephrology, Dialysis, Tr

Pathophysiology of Uremia, G. Gaslini Children HospGenova, Italy. Tel.: + 39 010 380742; fax: +39 010 3

E-mail address: labnefro@ospedale-gaslini.ge.it (G.M

0304-4165/$ – see front matter. Crown Copyright © 20http://dx.doi.org/10.1016/j.bbagen.2013.04.017

Please cite this article as: M. Bruschi, et al., Ohttp://dx.doi.org/10.1016/j.bbagen.2013.04.0

a b s t r a c t

a r t i c l e i n f o

Article history:

Received 11 March 2013Received in revised form 11 April 2013Accepted 15 April 2013Available online xxxx

Keywords:Reactive oxygen speciesOxidationSerum albuminRenal diseasesNephritic syndromeFocal segmental glomerulosclerosis

Background: Proteins are extremely reactive to oxidants and should represent a potential target of instable reac-tive oxygen. This may represent a problem for plasma proteins since they may be directly modified in vivo in acompartment where antioxidant enzymatic systems are scarcely represented. On the other hand, it is possiblethat some plasma components have evolved over time to guarantee protection, in which case they can be con-sidered as anti-oxidants.Scope of review: To present and discuss main studies which addressed the role of albumin in plasma antioxidantactivitymainly utilizing in vitromodels of oxidation. To present some advances on structural features of oxidizedalbumin deriving from studies carried out on in vitro models as well as albumin purified in vivo from patientsaffected by clinical conditions characterized by oxidative stress.Major conclusions: There are different interaction with HOCl and chloramines. In the former case, HOCl producesan extensive alteration of 238Trp and 162Tyr, 425Tyr, 47Tyr, while thiol groups are only partially involved. Chlora-mines are extremely reactive with the unique free SH group of albumin (34Cys) with the formation of sulfenic

and sulfinic acid as intermediates and sulfonic acid as end-product. Oxidized albumin has a modified electricalcharge for the addition of an acidic residue and presents α-helix and random coil reorganization with subtlechanges in domain orientation.General significance: Albumin, is the major antioxidants in plasma with a concentration (0.8 mM) higher thanother antioxidants by an exponential factor. Functional and protective roles in the presence of oxidative stressmust be defined. This article is part of a Special Issue entitled Serum Albumin.

Crown Copyright © 2013 Published by Elsevier B.V. All rights reserved.

1. Introduction

Albumin is the most abundant serum protein in mammalians andrepresents the oldest marker of phylogeny from primates to humans.While evolutionary stability suggests a critical role of the protein, itsfunctions are generally confined to the transport of ions and drugs.The role of albumin as ion transporter is allowed by its plasticitythat is the conformational change this protein may achieve in relationto pH. The regulation of plasma levels of Ca++ is the most remarkableexample, in which case albumin-Ca++ binding varies from 10% in thecase of acidosis to 80% for alkalosis and represents a reservoir of dis-posable Ca++ that can satisfy rapid requests [1].

Several in vitro models of oxidation indicate that albumin alsoplays key anti-oxidant functions [2–6] that strengthens its biologicalrole and makes it necessary to review anti-oxidant functions in

ced oxidation products; ESI-MS,quid chromatography with tan-metryAlbumin.ansplantation, Laboratory onital, Largo G. Gaslini 5, 1614895214.. Ghiggeri).

13 Published by Elsevier B.V. All rig

xidized albumin. The long w17

human livings. In the following sections we present different aspectsof albumin as an anti-oxidant protein including the definition of majoranti-oxidant groups inside the protein, biochemistry and structural fea-tures of the oxidized derivate. The possibility that oxidized metabolitesmay serve as surrogate biomarker of oxidative damage will be finallydiscuss. This review would represent a starting point on the way todeepen the knowledge of the role of oxidants as mediators of organdamage in selected diseases.

2. Considerations on extracellular anti-oxidant systems

Reactive oxygen species may play tremendous impact in humanlivings [7,8]. They are physiologically generated at low levels insidecells during oxidative phosphorylation where are easily integratedby specialized anti-oxidant defenses. The physiological response tothe oxidative burst is less definite outside cells where the levels ofboth glutathione and antioxidant enzymes are low and probably in-adequate to blunt a severe stress [9]. Generally speaking, all kinds ofoxygen radicals may induce translational modifications onmost pro-teins [10]. The list includes several oxygen (O−, OH−, RO−, RO2−,H2O2, HOCL, HOBr) and nitrogen (NO−, HNO2, ONOO−, etc.) spe-cies [11] that modify almost all amino acid residues [10,12]. Inblood, the generation of free radicals, mainly hypochlorous acid

hts reserved.

ay of a protein of uncertain function, Biochim. Biophys. Acta (2013),

2 M. Bruschi et al. / Biochimica et Biophysica Acta xxx (2013) xxx–xxx

(HOCl) and chloramines, by polymorphonuclear leukocytes (neutro-phils and monocytes) is crucial for defense against exogenous factors(infectious agents, etc.)[9,13]. HOCl derives from chloride and hydrogenperoxide in a reaction that is catalyzed bymyeloperoxidase, an enzymegenerated by leukocytes during the oxidative burst. Chloramines areproduced by the oxidation of circulating aminoacids by HOCl and areto be considered a secondorder oxidation product that retains an oxida-tive activity for biological molecules. Actually, it seems paradoxical thatlevels of anti-oxidant enzymes are much lower in blood than in the in-tracellular compartment since blood is comparably, if not more, ex-posed to the oxidative stress by leukocytes. Proteins are target ofoxygen radicals in this setting and may function as scavengers of themajority of reactive compounds [6,14]. Owing to their stability, oxida-tion products of plasma proteins can retain the fingerprint of the initialmodification and be utilized as markers of oxidative stress [12,15]. Inspite that several biomarkers of oxidative stress are available for analyt-ical strategies only few data are available on protein oxidation in vivothat report structural analysis and most of our knowledge are derivedfrom the characterization of proteins in artificial oxidation models[16,17]. It is likely that only the definition of chemical structure of oxi-dized proteins as they occur in humansmay serve as a trace for any evo-lution in pathology or be utilized as surrogates for clinical outcome. Thestarting point is to consider which protein and which reactive oxygenspecies interact.

The unique descriptions of oxidized proteins in vivo are related to theso called AOPPs (advanced oxidation protein products) that have beendetected in several clinical conditions such as diabetes mellitus, IgA glo-merulonephritis and more in general in diseases affecting the kidneyand leading to uremia [18–21]. In these cases AOPP levels seem to pre-dict the progression of atherosclerotic cardiovascular events, renal le-sions and death [18,22,23]. More in general, AOPPs play critical roles inbasic physiologic functions as potent inducers of oxidative burst invitro and in vivo suggesting that oxidized proteins may contribute bythemselves to the inflammatory process [24]. In spite these seeminglyimportant pathologic implications, we still lack a clear structural charac-terization of AOPPs that is mandatory for any evolution. Based on indi-rect techniques, Capeillere-Blandin et al. [25] have identified albuminas themain AOPP product in plasma, a finding that confirmed the previ-ous data indicating albumin as the major target of oxidant stress inserum [20,21,26].

3. Albumin as an antioxidant protein

In early 2000, preliminary studies based on the separation of ox-idized albumin by HPLC had shown its presence in the serum of pa-tients with renal dysfunction mainly affected by uremia [20,21,26].The indication of albumin as themost important plasma/serum compo-nent undergoing oxidation seemed to be the logical consequence of thefact that it retains amyeloperoxidase activity and is, for this reason, vul-nerable to oxidation by HOCl. Data deriving from studies on albuminoxidized ‘in vitro’ [6] showed different mechanisms of reaction withHOCl and chloramine [16,17]. In fact, HOCl reactivity with albumin pro-duces an extensive alteration of 238Trp and 162Tyr, 425Tyr, 47Tyr, whilethiol groups are only partially involved [14,16]. On the contrary, the

Fig. 1. Proposed scheme for albumin oxidation. Oxidation of a free SH leads to the formationand sulfinic acids.

Please cite this article as: M. Bruschi, et al., Oxidized albumin. The long whttp://dx.doi.org/10.1016/j.bbagen.2013.04.017

unique free SH group of albumin (34Cys) is extremely reactive withchloramines which induce the formation of two instable intermediates(i.e. sulfenic and sulfinic acid) and terminate with the end-productsulfonic acid [27] (see scheme in Fig. 1). It seems of note that thesame 34Cys scavenges also other electrophilic compounds and/or sub-stances such as heterocyclic aromatic amines [28] that may generateconfusion in the characterization of oxidized albumin. Therefore, stud-ies on albumin oxidized in vivo that define its precise structure werecrucial to address the role of this molecule as an anti-oxidant. Due tothe very low-level of oxidized albumin in normal plasma structuralstudies addressing these aspects could not be done considering normalclinical conditions. This possibility turned up when Musante et al.[29–31] showed an extensive sulfonation of albumin 34Cys SH in pa-tients affected by primary nephrotic syndrome, a clinical conditioncharacterized by renal lesions leading to urinary waste of proteins. Inthe following sections we give an outline of the main features of oxi-dized albumin including the description of main biochemistry featuresand laboratory techniques for its determination.

4. Biochemistry of oxidized albumin

4.1. Modification of Cys residues

The chemistry of oxidized residues of serum albumin has been eval-uated in vivo in clinical conditions that are now known to be associatedwith an oxidative stress (see the above description) and wereconfronted with normal serum albumin. To our knowledge, studiesperformed by Musante et al.. [29,30] were the first and still unique toaddress this aspect from a structural point of view. The analysis of theexact mass of albumin was carried out by ESI-MS showing a modest(+48 Da) albeit constant increment from 66.555 to 66.507 kDa thatcorresponds to the addition of 3 oxygen radicals (Fig. 2a). LC–MS–MSof tryptic digest of albumin was utilized to identify potential sites of ox-idation. The major finding was that the oxidation of the unique free SHof albumin sequence at 34Cys (peptide LVLIAFAQQCPFEDHV) that istransformed into a sulfonic group (m/z 511.71 in triply charge and1610.5 in double charge); two intermediate products of oxidation (i.e.sulfenic -SO− and/or sulfinic derivates SO2

−) at 169Cys and 265Cyswere, in parallel, observed. The same peptide above in normal albuminshowed instead two precursor ions (i.e. m/z 831.21 and 1245.67) at34Cys that are compatible with alkylation by iodoacetamide utilizedfor LC analysis. Normal albuminwas presented in two C5H9NO2S groupsinvolving 329Met and 446Met that probably represent a signal of physio-logic aging of the protein. Comparison of results deriving from the anal-ysis of in vivo and in vitro oxidized albumin [16,17,27] demonstratedthat modifications in the later situation are far more extensive thanwhat we can observe in vivo. In fact, in the later condition (i.e. albuminoxidized in vitro) an extensive sulfonic transformation (SO3

−) of severalfree SH residueswas observed togetherwith severalmethionine sulfox-ide (C5H9NO2S) derivatives. The finding of sulfonation 13 Cys residuessuggests conformational derangement with the rupture of disulfidebonds that in a normal conformation are liked in disulfide bonds. Thisaspect was confirmed by differential calorimetry (see below).

of a stable sulfonic acid as end product; two intermediates of the reaction are sulfenic

ay of a protein of uncertain function, Biochim. Biophys. Acta (2013),

4 9IEF non linear pH

EP

H

+

_

not oxidized

oxidized

oxidized

notoxidized

calculated

a b

Fig. 2. Characterization of ‘in vivo’ oxidized albumin. a — ESI-MS shows a +48 Da increment of exact mass of oxidized albumin (66.555 vs. 66.507 kDa) that suggested the additionof 3 oxygen radicals. This change is consistent with the formation of a sulfonic acid as reported in the scheme of Fig. 1. b — electrophoretic titration curves of oxidized andnon-oxidized albumin obtained from normal and pathological sera (top right panel) and their mathematical analysis using the Linderström–Lang equation corrected byelectrophoretic frictional coefficient [32] (bottom right panel). The oxidized protein presents a cathodic shift in the pH range between 4.6 and 7 that is consistent with thepresence of more anionic charges on the surface of the protein.

3M. Bruschi et al. / Biochimica et Biophysica Acta xxx (2013) xxx–xxx

4.2. Electrical charge and quantitative analysis of oxidized derivates

There are two peculiarities of oxidized albumin that derive fromthe changes described above. One is related to the charge of the pro-tein, the second is relate to the possibility to titrate the amount of ox-idation as an inverse function of the amount of free 34Cys SH. As forthe former point, we can predict that the electrical charge of oxidizedalbumin should be reduced by the presence of a sulfonic acid in placeof the 34Cys SH since this change introduces an acidic charge. Thismodification can be detected by electrophoretic titration curve [32]that is a two-dimensional electrophoretic technique able to deter-mine the charge of any given protein along a pH gradient between4 and 9. In accord with the prediction of an extra acidic charge dueto sulfonic acid, the isoelectric point of oxidized albumin showed,in fact, a global shift in the pH 5―7 interval with an acidic changein the molecule (Fig. 2b).

With reference to the possibility to determine the amount of ox-idized albumin methods have been utilized for labeling the sulfenicderivative of the protein (dimedone) [33] that is however an instableproduct. We proposed an essay in which the amount of oxidizedalbumin is calculated as the inverse of the content of 34Cys SH.According to the structural analysis above, the oxidation of 34CysSHwith the formation of an irreversible sulfonic acid residue reducesthe amount of free SH of albumin that in normal condition is limitedto the unique free SH residue relative to 34Cys. Overall, serum albu-min contains 35 cysteines that are linked in 17 disulfide bonds withthe exception of this residue (34Cys) that is free. A simple methodto titrate free SH in any protein sequence is the direct ‘in gel’ deter-mination after alkylation by maleimide-PEO2 and detection withbiotin-streptavidin. This method can be utilized as an alternative toEllman's reagent that was historically utilized for this purpose.Results from ‘in gel’ titration (Fig. 3a) confirmed the absence of free

Please cite this article as: M. Bruschi, et al., Oxidized albumin. The long whttp://dx.doi.org/10.1016/j.bbagen.2013.04.017

reactive SH groups in albumin from nephrotic patients. A second ap-proach to oxidized albumin that follows the same chemical back-ground above has been developed by Bruschi et al. [34] utilizingdifferential display electrophoresis (DIGE) based on two newiodoacetamide-substituted cyanines, C3NIASO3 and C5NIASO3 (syn-thesized starting from hemicyanine). Specificity for SH of these cya-nines was shown to be superior compared with other fluorescentlabels such as maleimide. The results of this single shot techniqueconfirmed what is already known on the reduction of free SH in oxi-dized albumin (Fig. 3b).

4.3. Spectroscopy and thermodynamic parameters

Prediction of structural stability of albumin suggests a role of theregion I where 34Cys is endowed. It is worth noting that in accord tostructural models, this region does not participate in the transportcapacity of the protein and it is likely that for structural reasons it es-caped the selection pressure to specialize this protein as a majorplasma transporter. In vitro studies that utilizes other proteins asmodels predict that the presence of a sulfonic residue in place of afree SH triggers the formation of new hydrogen bonds with adjacentaminoacids [35,36] with protein stabilization due to the formation ofa thermodynamically favored structure (Fig. 4). These aspects wereanalyzed by utilizing DSC (Fig. 5a), intrinsic Trp fluorescence (Fig.5b) and second derivative absorption spectra by Bruschi et al. [31]and are in part here reported. The results confirmed a thermal stabi-lization of the structure upon oxidation. In fact, the melting temper-ature (Tm transition) of the oxidized form is shifted at higher valueof a temperature of 16.57 °C; the enthalpy change contribution cal-culated from van't Hoff formula applying an optical procedure aswell by direct DSC experiment, corresponds to ΔH 510.8 kJ/mol forhealthy and ΔH 607.1 kJ/mol for oxidized albumin. These evidences

ay of a protein of uncertain function, Biochim. Biophys. Acta (2013),

HSA

notoxidized

C5NIASO3oxidized

C3NIASO3

coom

assi

est

aini

ngw

este

rn b

lot

PE

O-m

alei

mid

e

HSA

HSA

notoxidizedoxidized

Bluesilver

colloidalstaining

DIG

EIodoacetam

idecyanines

HSA

a b

Fig. 3. Determination of oxidized albumin in serum. The rationale for the determination of the amount of oxidized albumin in serum is based on the concept that the oxidation of theunique free SH (34Cys) by chloramines is associated with the lack of reactivity with stains specific for SH. In this case, reactivity of serum albumin with PEO-maleimide (a — toppanel) and with iodoacetamide cyanines (b — top panel) is markedly reduced.

4 M. Bruschi et al. / Biochimica et Biophysica Acta xxx (2013) xxx–xxx

are corroborated by circular dichroism data and by small angle X-rayscattering [31] suggesting slight changes in protein secondary struc-ture possibly involving α-helix and random coil reorganization andsubtle changes in domain orientation.

5. Fate and excretion of oxidized albumin

Human serum albumin has a half life of 20 days. Degradationfollows mechanisms not entirely defined: in plasma it is degradedinto peptides whereas in the kidney is filtered at the glomerularlevel, reabsorbed at the proximal tubule and degraded to peptides.Charge and conformation play a role in urinary excretion [37,38].We have no reason to suggest that oxidized albumin follows a

34cysthiolic groupsulfonic group

regions modified by oxidation of 34cys

region I

Fig. 4. Structural model for ‘in vivo’ oxidized albumin. Molecular dynamic simulation of strugray) in place of a free SH (in red) at 34Cys (in yellow) triggers the formation of new hydrblack). We adapted our results and utilized Accelrys DS Viewer Pro 5 software to generate

Please cite this article as: M. Bruschi, et al., Oxidized albumin. The long whttp://dx.doi.org/10.1016/j.bbagen.2013.04.017

different process or has an accelerated degradation. Several albuminfragments can be detected in urine; overall, they represent the 20–30% of the urinary pool in normal conditions. Levels of urinary frag-ments increase during nephrotic syndrome and seem in part corre-lated with albumin oxidation [39]. Therefore, this apparentcorrelation indirectly supports the idea that fragmentation is in-creased for oxidized albumin and that fragments are freely excretedinto urine.We do not knowwhether tubular handling with increasedre-absorption of oxidized albumin or its fragments occur and howthis potential mechanism modifies renal homeostasis. What we canunderstand from human pathology is that the kidney in conditionsof heavy proteinuria is massively inflamed in tubulo-interstitialareas and the possibility exists that infiltration by inflammatory

oxidized

not oxidized

thiolic

sulfonic

34cys

34cys

ctural model of oxidized and normal albumin. The presence of a sulfonic acid (in darkogen bond and promotes the formation of conformational changes in five regions (inthis figure.

ay of a protein of uncertain function, Biochim. Biophys. Acta (2013),

20 40 60 80 100

0,0

0,2

0,4

0,6

0,8

1,0

f

Temperature [°C]

20 40 60 80 100

0,0

4,0

8,0

12,0

ΔCp

x 10

4 [

joul

es/m

ole

K]

Temperature [°C]

oxidized

not oxidized

DS

C a

naly

sis

Flu

ores

cenc

ean

d U

V a

naly

sis

not oxidized

oxidized Fluorescence

UV

b

a

Fig. 5. Spectroscopy and thermodynamic parameters of ‘in vivo’ oxidized albumin. Analysis of oxidized and normal albumin by (a) differential scanning calorimetry (DSC) and (b)unfolding fraction calculated from the second derivate of fluorescence and UV absorption spectra. Results indicate that the difference of Tm between the oxidized and normal al-bumin can be justified by the presence of new hydrogen bonds formed in oxidized molecule.

5M. Bruschi et al. / Biochimica et Biophysica Acta xxx (2013) xxx–xxx

cells is due to the presence of unusual peptides or aminoacids deriv-ing from altered reabsorbed proteins.

6. Significance in normality and pathology

The observation that massive oxidation of serum albumin doesoccur in vivo is new and suggests a few considerations. The first isabout physiologic functions of albumin that can now be extended toan anti-oxidation activity. In the past, the anti-oxidant role of albuminin serum has been only hypothesized on the basis of ‘in vitro’ experi-ments. In normal conditions, the oxidation of albumin should involveonly a minor part of the protein due to limited stress. It is likely thatthe quota of non-mercaptoalbumin represents the end product of thereaction that involves the formation of a reactive sulfenic intermediateof the free 34Cys that reacts with free glutathione and/or with homocys-teine. Extension of structural and conformational analysis of albumin inpathologic conditions was critical to obtain a clear demonstration of astable thermodynamic end product of oxidation. On the basis of theabove data, albumin may be considered the major anti-oxidant sub-stance in serum where, due to the high concentration (0.6–0.8 mM),the amount of reactive 34Cys SH overwhelms by an exponential factorthe amount of other antioxidant SH such as free glutathione whoseplasma levels are only 0.006 mM.

The second point implies that in nephrotic syndrome oxidation ismarkedly activated and probably plays a key role as pathogenic trig-ger. This concept is supported by preliminary and indirect evidenceon the peroxidation of plasma membranes and consumption ofintra-erythrocyte GSH in humans with FSGS [40–42] and on animalmodels of the disease that include puromycin (PAN) and adriamycin(ADR) nephrosis in rats and Mvp 17−/− mice [41,43–46]. Overall,

Please cite this article as: M. Bruschi, et al., Oxidized albumin. The long whttp://dx.doi.org/10.1016/j.bbagen.2013.04.017

experimental data and results from oxidized albumin in human be-ings support a general concept on the participation of oxidativemechanisms in nephrotic syndrome. We should start to considerthe use of anti-oxidant drugs with the hope that they modify theclinical outcome in these patients and blunt progression to endstage organ failure.

Acknowledgements

The Giannina Gaslini Institute provided financial and logistic sup-port to the study. This work was also supported by the Italian Ministryof Health ‘Ricerca Corrente’ and from contributions derived from ‘Cinqueper mille dell ‘IRPEF’. We also acknowledge contributions from theRenal Child Foundation, Fondazione Mara Wilma e Bianca Querci (pro-ject “Ruolo dello stress reticolare nella progressione del danno renale etumorale”), and Fondazione La Nuova Speranza (‘Progetto integrato perla definizione deimeccanismi implicati nella glomerulo sclerosi focale’).

Appendix A. Fluorescence spectra

Fluorescence spectra were recorded and evaluated in a LS50BPerkin Elmer Luminescence Spectrometer (Wellesley, MA, USA). Ex-citation at 295 nmwas utilized to minimize the contribution of tyro-sine in intrinsic fluorescence. Excitation and emission slit widthswere maintained at 4.5 nm and scan speed was 300 nm/min. Thetemperature of sample cell was maintained constant by circulatingwater bath (Haake, Karsruhe-Berlin, Germany) and monitored witha teflon coated microthermistor probe. Spectra were analyzed afterthe subtraction of baselines. The position of the middle of a chord

ay of a protein of uncertain function, Biochim. Biophys. Acta (2013),

6 M. Bruschi et al. / Biochimica et Biophysica Acta xxx (2013) xxx–xxx

drawn at the 80% level of themaximum intensity (λmax) was taken asthe position of the spectrum [47].

Ultraviolet spectra

Both normal and second derivative spectra were obtained with aUV–VIS Cary BIO400 (Varian, Palo Alto, CA, USA). The temperature inthe sample cell was maintained constant by circulating water bathand monitored with a teflon coated microthermistor probe as above.

Dielectricmicroenvironment surrounding Tyr residueswas calculat-ed from the ratio (r) between differences in second derivative absor-bance peak (a/b) relative at 284 nm and 279 nm for Tyr and 293 nm286 nm for Trp according to Ragone et al. [48]

r ¼ ab¼ A} 284nmð Þ−A} 279nmð Þ

A} 293nmð Þ−A} 286nmð Þ

where A″ is the second derivative adsorbance at any given wavelengthλ. The degree of Tyr exposure, α, is obtained from the equation

α ¼ rn−raru−ra

where rn and ru are the numerical values of the ratio a/b determined fornative and unfolded conformers and ra corresponds to the ratio be-tween Tyr and Trp in a model compound in the presence of ethyleneglycol and represents a complete burial of all aromatic residues of pro-tein. Sincehuman serumalbumin contains 18 Try and 1 Trp residues thesecond derivative peak and trough ratios ra and ruwere obtained for themolar ratio (18/1) with the second derivative molar extinction coeffi-cients of Ragone et al. [48]. Thus for alb, the calculated values werera = −2.74 and ru = 17.82 for the complete burial and exposure ofTry residues, respectively.

Circular dichroism

The CD measurements of Healthy and FSGS albumin performedwith J-810 spectropolarimeter (Jasco, Great Dunmow, Essex, UK) in-struments in the range of 190–260 nm at 20 °C, using a 1 mm cell at1 nm intervals. The concentration of both macromolecule was0.1 mg/ml in 50 mM phosphate buffer pH 7.2. All spectra were back-ground corrected and analyzed using CDNN software.

Differential scanning calorimetry. Degassed samples with a concen-tration ranging from 0.5 to 1.0 mg/mlwere performedwith aMicroCal-orimeter VP–DSC and curves have been analyzed with Origin software(MicroCal, Northampton, MA, USA). The scan heating rate was 30, 60and 90 °C/h. Independent instrumental baselines were carried out byscanning 50 mM phosphate buffer at pH 7.4, at the correspondentheating rate and subtracted from experimental runs.

References

[1] O.J. Bos, J.F. Labro, M.J. Fischer, J. Wilting, L.H. Janssen, The molecular mechanismof the neutral-to-base transition of human serum albumin. Acid/base titrationand proton nuclear magnetic resonance studies on a large peptic and a large tryp-tic fragment of albumin, J Biol Chem 264 (1989) 953–959.

[2] M.K. Cha, I.H. Kim, Glutathione-linked thiol peroxidase activity of human serumalbumin: a possible antioxidant role of serum albumin in blood plasma, Biochem.Biophys. Res. Commun. 222 (1996) 619–625.

[3] H. Lee, M.K. Cha, I.H. Kim, Activation of thiol-dependent antioxidant activity ofhuman serum albumin by alkaline pH is due to the B-like conformational change,Arch. Biochem. Biophys. 380 (2000) 309–318.

[4] M. Anraku, K. Yamasaki, T. Maruyama, U. Kragh-Hansen, M. Otagiri, Effect of ox-idative stress on the structure and function of human serum albumin, Pharm.Res. 18 (2001) 632–639.

[5] S. Carballal, R. Radi, M.C. Kirk, S. Barnes, B.A. Freeman, B. Alvarez, Sulfenic acid for-mation in human serum albumin by hydrogen peroxide and peroxynitrite, Bio-chemistry 42 (2003) 9906–9914.

[6] P. Salavej, H. Spalteholz, J. Arnhold, Modification of amino acid residues in humanserum albumin by myeloperoxidase, Free Radic. Biol. Med. 40 (2006) 516–525.

Please cite this article as: M. Bruschi, et al., Oxidized albumin. The long whttp://dx.doi.org/10.1016/j.bbagen.2013.04.017

[7] I. Fridovich, The biology of oxygen radicals, Science 201 (1978) 875–880.[8] I. Fridovich, Oxygen free radicals and tissue damage: chairman's introduction,

CIBA Found. Symp. (1978) 1–4.[9] M.B. Hampton, A.J. Kettle, C.C. Winterbourn, Inside the neutrophil phagosome:

oxidants, myeloperoxidase, and bacterial killing, Blood 92 (1998) 3007–3017.[10] E.R. Stadtman, R.L. Levine, Protein oxidation, Ann. N. Y. Acad. Sci. 899 (2000)

191–208.[11] I. Dalle-Donne, A. Scaloni, D. Giustarini, E. Cavarra, G. Tell, G. Lungarella, R. Colombo,

R. Rossi, A. Milzan, Proteins as biomarkers of oxidative/nitrosative stress in diseases:the contribution of redox proteomics, Mass Spectrom Rev 24 (2005) 55–99.

[12] E.R. Stadtman, Protein oxidation in aging and age-related diseases, Ann. N. Y.Acad. Sci. 928 (2001) 22–38.

[13] B. Halliwell, J.M. Gutteridge, Role of free radicals and catalytic metal ions inhuman disease: an overview, Methods Enzymol. 186 (1990) 1–85.

[14] A.C. Carr, C.L. Hawkins, S.R. Thomas, R. Stocker, B. Frei, Relative reactivities ofN-chloramines and hypochlorous acid with human plasma constituents, FreeRadic. Biol. Med. 30 (2001) 526–536.

[15] A. Telci, U. Cakatay, R. Kayali, C. Erdogan, Y. Orhan, A. Sivas, T. Akcay, Oxidative proteindamage in plasma of type 2 diabetic patients, Horm. Metab. Res. 32 (2000) 40–43.

[16] C.C. Winterbourn, A.J. Kettle, Biomarkers of myeloperoxidase-derived hypochlorousacid, Free Radic. Biol. Med. 29 (2000) 403–409.

[17] A.V. Peskin, C.C. Winterbourn, Kinetics of the reactions of hypochlorous acid andamino acid chloramines with thiols, methionine, and ascorbate, Free Radic. Biol.Med. 30 (2001) 572–579.

[18] B. Descamps-Latscha, V. Witko-Sarsat, T. Nguyen-Khoa, A.T. Nguyen, V. Gausson,N. Mothu, G.M. London, P. Jungers, Advanced oxidation protein products as riskfactors for atherosclerotic cardiovascular events in nondiabetic predialysis pa-tients, Am J Kidney Dis 45 (2005) 39–47.

[19] B. Descamps-Latscha, V. Witko-Sarsat, T. Nguyen-Khoa, A.T. Nguyen, V. Gausson,N. Mothu, C. Cardoso, L.H. Noel, A.P. Guerin, G.M. London, P. Jungers, Early predic-tion of IgA nephropathy progression: proteinuria and AOPP are strong prognosticmarkers, Kidney Int. 66 (2004) 1606–1612.

[20] H. Terawaki, K. Yoshimura, T. Hasegawa, Y. Matsuyama, T. Negawa, K. Yamada, M.Matsushima, M. Nakayama, T. Hosoya, S. Era, Oxidative stress is enhanced in cor-relation with renal dysfunction: examination with the redox state of albumin,Kidney Int. 66 (2004) 1988–1993.

[21] K. Mera, M. Anraku, K. Kitamura, K. Nakajou, T. Maruyama, M. Otagiri, The struc-ture and function of oxidized albumin in hemodialysis patients: Its role in elevat-ed oxidative stress via neutrophil burst, Biochem. Biophys. Res. Commun. 334(2005) 1322–1328.

[22] T. Drueke, V. Witko-Sarsat, Z. Massy, B. Descamps-Latscha, A.P. Guerin, S.J.Marchais, V. Gausson, G.M. London, Iron therapy, advanced oxidation proteinproducts, and carotid artery intima-media thickness in end-stage renal disease,Circulation 106 (2002) 2212–2217.

[23] H.Y. Li, F.F. Hou, X. Zhang, P.Y. Chen, S.X. Liu, J.X. Feng, Z.Q. Liu, Y.X. Shan, G.B. Wang,Z.M. Zhou, J.W. Tian, D. Xie, Advanced oxidation protein products accelerate renalfibrosis in a remnant kidney model, J. Am. Soc. Nephrol. 18 (2007) 528–538.

[24] V. Witko-Sarsat, M. Friedlander, T. Nguyen Khoa, C. Capeillere-Blandin, A.T.Nguyen, S. Canteloup, J.M. Dayer, P. Jungers, T. Drueke, B. Descamps-Latscha, Ad-vanced oxidation protein products as novel mediators of inflammation andmonocyte activation in chronic renal failure, J. Immunol. 161 (1998) 2524–2532.

[25] C. Capeillere-Blandin, V. Gausson, B. Descamps-Latscha, V. Witko-Sarsat, Bio-chemical and spectrophotometric significance of advanced oxidized proteinproducts, Biochim. Biophys. Acta 1689 (2004) 91–102.

[26] J. Himmelfarb, E. McMonagle, Albumin is the major plasma protein target of oxi-dant stress in uremia, Kidney Int. 60 (2001) 358–363.

[27] M. Bruschi, A. Petretto, G. Candiano, L. Musante, E. Movilli, L. Santucci, A. Urbani,R. Gusmano, E. Verrina, G. Cancarini, F. Scolari, G.M. Ghiggeri, Determination ofthe oxido-redox status of plasma albumin in hemodialysis patients, J ChromatogrB Analyt Technol Biomed Life Sc 864 (2008) 29–37.

[28] L. Peng, R.J. Turesky, Mass spectrometric characterization of 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine N-oxidized metabolites bound at Cys34 of humanserum albumin, Chem. Res. Toxicol. 24 (2011) 2004–2017.

[29] L. Musante, M. Bruschi, G. Candiano, A. Petretto, N. Dimasi, P. Del Boccio, A.Urbani, G. Rialdi, G.M. Ghiggeri, Characterization of oxidation end product of plas-ma albumin ‘in vivo’, Biochem. Biophys. Res. Commun. 349 (2006) 668–673.

[30] L. Musante, G. Candiano, A. Petretto, M. Bruschi, N. Dimasi, G. Caridi, B. Pavone, P.Del Boccio, M. Galliano, A. Urbani, F. Scolari, F. Vincenti, G.M. Ghiggeri, Activefocal segmental glomerulosclerosis is associated with massive oxidation of plas-ma albumin, J. Am. Soc. Nephrol. 18 (2007) 799–810.

[31] M. Bruschi, L. Musante, G. Ghiggeri, G. Rialdi, G. Candiano, Comparative study ofthermal stability of healthy and focal segmental glomerulosclerosis plasma albu-min, J. Therm. Anal. Calorim. 87 (2007) 27–31.

[32] M. Bruschi, L. Musante, G. Candiano, L. Santucci, C. Zennaro, M. Carraro, P. DelBoccio, R. Gusmano, F. Perfumo, A. Urbani, G.M. Ghiggeri, Transitions of serum al-bumin in patients with glomerulosclerosis ‘in vivo’ characterization by electro-phoretic titration curves, Electrophoresis 27 (2006) 2960–2969.

[33] B. Alvarez, S. Carballal, L. Turell, R. Radi, Formation and reactions of sulfenic acidin human serum albumin, Methods Enzymol. 473 (2010) 117–136.

[34] M. Bruschi, S. Grilli, G. Candiano, S. Fabbroni, L. Della Ciana, A. Petretto, L. Santucci,A. Urbani, R. Gusmano, F. Scolari, G.M. Ghiggeri, New iodo-acetamido cyanines forlabeling cysteine thiol residues. A strategy for evaluating plasma proteins andtheir oxido-redox status, Proteomics 9 (2009) 460–469.

[35] G. Buhrman, B. Parker, J. Sohn, J. Rudolph, C. Mattos, Structural mechanism of ox-idative regulation of the phosphatase Cdc25B via an intramolecular disulfidebond, Biochemistry 44 (2005) 5307–5316.

ay of a protein of uncertain function, Biochim. Biophys. Acta (2013),

7M. Bruschi et al. / Biochimica et Biophysica Acta xxx (2013) xxx–xxx

[36] A. Kawakami, K. Kubota, N. Yamada, U. Tagami, K. Takehana, I. Sonaka, E. Suzuki,K. Hirayama, Identification and characterization of oxidized human serum albu-min. A slight structural change impairs its ligand-binding and antioxidant func-tions, Febs J 273 (2006) 3346–3357.

[37] G.M. Ghiggeri, G. Candiano, F. Ginevri, R. Gusmano, M.R. Ciardi, F. Perfumo, G.Delfino, C. Cuniberti, C. Queirolo, Renal selectivity properties towards endoge-nous albumin in minimal change nephropathy, Kidney Int. 32 (1987) 69–77.

[38] G.M. Ghiggeri, F. Ginevri, G. Candiano, R. Oleggini, F. Perfumo, C. Queirolo, R.Gusmano, Characterization of cationic albumin in minimal change nephropathy,Kidney Int. 32 (1987) 547–553.

[39] G. Candiano, L. Musante, M. Bruschi, A. Petretto, L. Santucci, P. Del Boccio, B.Pavone, F. Perfumo, A. Urbani, F. Scolari, G.M. Ghiggeri, Repetitive fragmenta-tion products of albumin and alpha1-antitrypsin in glomerular diseases associ-ated with nephrotic syndrome, J. Am. Soc. Nephrol. 17 (2006) 3139–3148.

[40] F. Ginevri, G.M. Ghiggeri, G. Candiano, R. Oleggini, R. Bertelli, M.T. Piccardo, F.Perfumo, R. Gusmano, Peroxidative damage of the erythrocyte membrane in chil-dren with nephrotic syndrome, Pediatr. Nephrol. 3 (1989) 25–32.

[41] F. Ginevri, R. Gusmano, R. Oleggini, S. Acerbo, R. Bertelli, F. Perfumo, G. Cercignani, S.Allegrini, F. D'Allegri, G. Ghiggeri, Renal purine efflux and xanthine oxidase activityduring experimental nephrosis in rats: difference between puromycin amino-nucleoside and adriamycin nephrosis, Clin Sci (Lond) 78 (1990) 283–293.

Please cite this article as: M. Bruschi, et al., Oxidized albumin. The long whttp://dx.doi.org/10.1016/j.bbagen.2013.04.017

[42] S. Kinra, B. Rath, B.C. Kabi, Indirect quantification of lipid peroxidation in steroidresponsive nephrotic syndrome, Arch. Dis. Child. 82 (2000) 76–78.

[43] G.M. Ghiggeri, G. Cercignani, F. Ginevri, R. Bertelli, L. Zetta, F. Greco, G. Candiano,A. Trivelli, R. Gusmano, Puromycin aminonucleoside metabolism by glomeruliand glomerular epithelial cells in vitro, Kidney Int. 40 (1991) 35–42.

[44] V. Thakur, P.D. Walker, S.V. Shah, Evidence suggesting a role for hydroxylradical in puromycin aminonucleoside-induced proteinuria, Kidney Int. 34(1988) 494–499.

[45] T. O'Bryan, H. Weiher, H.G. Rennke, S. Kren, T.H. Hostetter, Course of renal injury inthe Mpv17-deficient transgenic mouse, J. Am. Soc. Nephrol. 11 (2000) 1067–1074.

[46] C.J. Binder, H. Weiher, M. Exner, D. Kerjaschki, Glomerular overproduction of ox-ygen radicals in Mpv17 gene-inactivated mice causes podocyte foot process flat-tening and proteinuria: a model of steroid-resistant nephrosis sensitive to radicalscavenger therapy, Am. J. Pathol. 154 (1999) 1067–1075.

[47] A. Ruiz-Arribas, G.G. Zhadan, V.P. Kutyshenko, R.I. Santamaria, M. Cortijo, E. Villar,J.M. Fernandez-Abalos, J.J. Calvete, V.L. Shnyrov, Thermodynamic stability of twovariants of xylanase (Xys1) from Streptomyces halstedii JM8, Eur. J. Biochem. 253(1998) 462–468.

[48] R. Ragone, G. Colonna, C. Balestrieri, L. Servillo, G. Irace, Determination of tyrosineexposure in proteins by second-derivative spectroscopy, Biochemistry 23 (1984)1871–1875.

ay of a protein of uncertain function, Biochim. Biophys. Acta (2013),