Albumin Is the Main Nucleophilic Target of Human Plasma: A

The aim of this work was to study the metabolic fate of 4-hydroxy-trans-2-nonenal (HNE) in human plasma, which represents the main vascular site of re...
2 downloads 0 Views 2MB Size
Download PDF
824

Chem. Res. Toxicol. 2008, 21, 824–835

Albumin Is the Main Nucleophilic Target of Human Plasma: A Protective Role Against Pro-atherogenic Electrophilic Reactive Carbonyl Species? Giancarlo Aldini,*,† Giulio Vistoli,† Luca Regazzoni,† Luca Gamberoni,† Roberto Maffei Facino,† Satoru Yamaguchi,‡ Koji Uchida,‡ and Marina Carini† Istituto di Chimica Farmaceutica e Tossicologica “Pietro Pratesi”, Faculty of Pharmacy, UniVersity of Milan, I-20131, Milan, Italy, and Graduate School of Bioagricultural Sciences, Nagoya UniVersity, Nagoya 464-8601, Japan ReceiVed September 26, 2007

The aim of this work was to study the metabolic fate of 4-hydroxy-trans-2-nonenal (HNE) in human plasma, which represents the main vascular site of reactive carbonyl species (RCS) formation and where the main pro-atherogenic target proteins are formed. When HNE was spiked in human plasma, it rapidly disappeared (within 40 s) and no phase I metabolites were detected, suggesting that the main fate of HNE is due to an adduction mechanism. HNE consumption was then monitored in two plasma fractions: low molecular weight plasma protein fractions (10 kDa; HMWF). HNE was almost stable in LMWF, while in HMWF it was consumed by almost 70% within 5 min. Proteomics identified albumin (HSA) as the main protein target, as further confirmed by a significantly reduced HNE quenching of dealbuminated plasma. LC-ESI-MS/MS analysis identified Cys34 and Lys199 as the most reactive adduction sites of HSA, through the formation of a Michael and Schiff base adducts, respectively. The rate constant of HNE trapping by albumin was 50.61 ( 1.89 M-1 s-1 and that of Cys34 (29.37 M-1 s-1) was 1 order of magnitude higher with respect to that of GSH (3.81 ( 0.17 M-1 s-1), as explained by molecular modeling studies. In conclusion, we suggest that albumin, through nucleophilic residues, and in particular Cys34, can act as an endogenous detoxifying agent of circulating RCS. Introduction 1

The involvement of reactive carbonyl species (RCS) in atherosclerosis was first documented by immunohistochemical studies, showing that oxidized low density lipoproteins (LDLs) and atherosclerotic lesions of varying severity from human aorta contained material recognized by antibodies raised against specific aldehyde adducts such as HNE-His, MDA-Lys, and ACR-Lys (FDP-Lys) (1). The cause or effect role of RCS is still an open issue, but it seems reasonable to consider RCS, and 4-hydroxy-trans-2-nonenal (HNE) in particular, as molecules consistently involved in the atherosclerotic process, both in the formation of the atheroma and in the fibrotic transformation of the arterial wall (2). Accordingly, HNE has been shown to activate both macrophage and smooth muscle cells, that is, the two key cell types in chronic inflammatory processes characterized by excessive fibrinogenesis (3). Furthermore, RCS are electrophilic compounds and covalent adduct-susceptible proteins, leading to pro-atherogenic processes (4). For example, * To whom correspondences should be addressed. Tel: +39-02- 50319296. Fax: +39-02-50319359. E-mail: [email protected]. † University of Milan. ‡ Nagoya University. 1 Abbreviations: ACR, acrolein; DHN, 1,4-dihydroxy-trans-nonene; DTDP, 4,4′-dithiodipyridine; DTT, 1,4-dithio-DL-threitol; HMWF, high molecular weight plasma protein fractions (>10 kDa); HNA, 4-hydroxytrans-2-nonenoic acid; HNE, 4-hydroxy-trans-2-nonenal; HSA, human serum albumin; HSA-Cys34-SH, mercaptoalbumin; HSA-Cys34-S-IAA, iodoacetamide sulfhydryl-blocked HSA; IAA, iodoacetamide; LDL, low density lipoproteins; LMWF, low molecular weight plasma protein fractions ( 0.5, one sample t test, 95% confidence intervals) to the content determined in the serum sample. This indicates that no significant loss of thiols took place during the sample preparation. HNE Consumption in Plasma and Plasma Fractions. When HNE (20 µM) was incubated in plasma at 37 °C, it was totally consumed in less than 40 s, while no significant consumption of HNE was observed in PBS solution (data not shown). HPLC was used to check the formation of HNA and DHN, the main phase I metabolites of HNE arising from the oxidation and reduction of the aldehydic moiety, respectively. The two metabolites were undetectable in plasma incubated with 20 µM HNE, and their levels were always below the limit of detection (LOD) even by increasing the concentration of the aldehyde up to 200 µM and by prolonging the incubation time to 3 h (at this time, only 5% of residual HNE was found) (data not shown). By considering the LODs of HNA and DHN (0.125, and 1.00 µM, respectively), these results indicate that the conversion of HNE to HNA and DHN, if any, is less than 0.5% of the initial incubated dose. These data seem to indicate that the main fate of HNE in plasma involves an adduction mechanism rather than an enzymatic or nonenzymatic reduction/oxidation of the aldehydic function. Proteins and peptides represent a reactive nucleophilic target of electrophilic aldehydes, mainly through the covalent adduction to Lys, Cys, and His residues. To understand whether HNE mainly reacts with the peptide or protein plasma fraction, the kinetics of HNE consumption was performed using the two plasma fractions containing proteins 10 kDa (HMWF). As shown in Figure 1, HNE was slightly reduced in the LMWF fraction (1.51 ( 0.06 nmol mL-1 loss in 60 min). This value is almost similar to the calculated consumption (1.67 nmol mL-1), as determined by considering (i) the thiol pool (6.1 µM, since diluted 1:1), (ii) the HNE final concentration (20 µM), and (iii) the rate constant of HNE trapping by low molecular weight thiols (for GSH ) 3.8 M-1 s-1; see below). By contrast, HNE spiked in the HMWF was consumed by almost 70% within 5 min, to gradually disappear within 60 min of incubation.

828

Chem. Res. Toxicol., Vol. 21, No. 4, 2008

Aldini et al.

Figure 2. Identification of plasma targets of HNE. HNE (0–10 mM) was incubated in plasma at 37 °C for 24 h. (A) SDS-PAGE. (B) Immunoblot analysis with polyclonal antibodies against HNE-modified proteins. (C) Two-dimensional gel electrophoresis/immunoblot analysis. The proteins were separated by isoelectrofocusing (pH range 3–10) and then by SDS-PAGE. In panels B and C, the arrows represent the major target of HNE.

Identification of HSA as the Major Nucleophilic Plasma Target of HNE. To identify the protein targets of HNE in plasma, HNE was incubated in plasma at 37 °C for 24 h, and plasma proteins were analyzed by SDS-PAGE followed by immunoblot analysis using polyclonal antibodies against HNEmodified proteins. As shown in Figure 2A,B, a protein band with the molecular mass corresponding to HSA was detected as the major HNE target. The plasma samples were further analyzed by two-dimensional gel electrophoresis followed by immunoblot analysis, also showing that the spot corresponding to the 70 kDa HNE-protein conjugate was the major target of HNE in this analysis (panel C). The spot was then excised from two-dimensional gels, subjected to trypsin digestion, and then successfully analyzed by MALDI-TOF mass fingerprint analysis (Figure S2, Supporting Information). Using MASCOT, the probability-based MOWSE score was 197 for HSA (p < 0.05), with five peptide matches (error ( 0.01%), which represents 9% sequence coverage. The broad gel spots for a single identification may be due to post-translational modification, such as oxidation and phosphorylation. To confirm the main role of HSA in plasma HNE consumption, the consumption of HNE in plasma and dealbuminated plasma was monitored by HPLC. The results shown in Figure 3 well indicate that when plasma was depleted of HSA by affinity chromatography, the stability of HNE significantly increased. In particular, the rate of HNE consumption (calculated in the first 5 min) was of 1.54 ( 0.08 nmol mL-1 min-1 in plasma to be reduced to 0.52 ( 0.02 nmol mL-1 min-1 in dealbuminated plasma. The specific removal of HSA from plasma samples was checked by SDS-PAGE as shown in Figure 3. Identification of Plasma Albumin HNE-Adducted Peptides by LC-ESI-MS/MS. HSA is characterized by several nucleophilic and accessible Cys, Lys, and His residues, which can be adducted by HNE thus to explain the high reactivity of the protein toward RCS. We previously found that when HNE was mixed with HSA in PBS and at a 5:1 molar ratio, nine nucleophilic sites were covalently modified, namely, His67

Figure 3. Kinetics of HNE consumption in diluted plasma (9) and dealbuminated (0) plasma. At each time point, the HNE content was determined by HPLC analysis as described in the text. On the right, the SDS-PAGE electrophoresis (blue comassie staining) relative to diluted plasma (a) and dealbuminated plasma (b).

(MA), His146 (MA), His242 (MA), His288 (MA), His510 (MA), Lys 195 (SB), Lys 199 (MA, SB), Lys525 (MA, SB), and Cys34 (MA), forming 11 different adducts through both a Michael and a SB mechanism. As summarized in Table 1, when human plasma was incubated with HNE at a final concentration of 500 µM, all of the HSA peptide adducts were easily identified, with the exception of the MA of His67. When the HNE concentration was reduced to 100 µM, seven adducts were found, due to the HNE covalent adduction to the following sites: Lys199 (MA), Lys199 (SB), His242 (MA), Lys525 (SB), His288 (MA), His146 (MA), and Cys34 (MA). At 10 µM HNE, only two adducted peptides were identified, arising from the formation of a SB and MA adducts of HNE with Lys199 and Cys34, respectively. The identity of the different adducts was unequivocally confirmed by the MS/MS spectra (data not shown). Figure 4 shows the LC-ESI-MS/MS chromatograms in MRM mode obtained by colliding the product ion at m/z 544.5, relative to the HNE-adducted peptide LKCASLQ through a SB. The peak relative to the adduct (RT ) 42.1 min) was prominent in HSA incubated with an excess of HNE (ratio 1:5),

HSA as Plasma Target of RCS

Chem. Res. Toxicol., Vol. 21, No. 4, 2008 829

Table 1. LC-ESI-MS/MS Analysis in MRM Mode of Albumin HNE-Adducted Peptides in Plasma Spiked with 500, 100, and 10 µM HNEa HNE (µM) adducted peptide sequence

adducted AA

type of adduct

parent ion (m/z)

ASSAK*QR LK*CASLQK

Lys195 Lys199

SB MA

trypsin 435.4 [M + 2H]2+ 553.3 [M + 2H]2+

LK*CASLQK VH*TEC°C°HG DLLEC°ADDR

Lys199 His242

SB MA

544.5 [M + 2H]2+ 749.0 [M + 3H]3+

K*QTALVELVK

Lys525

MA

644.2 [M + 2H]2+

K*QTALVELVK SLH*TLFGDK

Lys525 His67

SB MA

635.0 [M + 2H]2+ 588.3 [M + 2H]2+

EFNAETFTFH*ADIC°TLSEK

His510

MA

1209.9 [M + 2H]2+

SH*C°IAEVE NDEMPADL PSLAADFVESK

His288

MA

1566.3 [M + 2H]2+

H*PY

His146

MA

trypsin + chymotrypsin 574.3 [M + H]+

LQQC°PF

Cys34

MA

893.5 [M + H]+

a

product ions (m/z)

500

100

10

RT (min)

435.4 [M + 2H - H2O]2+ 544.4 [M + 2H - H2O]2+ 831.4 (b6*) 959.3 (b7*) 535.5 [M + 2H - H2O]2+ 627.9 (b9*2+) 684.3 (b10*2+) 978.4 (b15*2+) 928.4 (b7*) 1041.5 (b8*) 1140.6 (b9*) 626.1 [M + 2H - H2O]2+ 488.2 (y7*2+) 579.4 (y5) 1029.4 (b8*) 1726.3 (y13*) 1579.7 (y12*) 1200.8 [M + 2H - H2O]2+ 1276.6 (y12) 1856.6 (b15*) 1969.6 (b16*)

+ +

+

-

29.5 39.2

+ +

+ +

+ -

41.7 45.2

+

-

-

55.5

+ -

+ -

-

57.9 59.9

+

-

-

68.3

+

+

-

72.9

268.2 (His-HNE imm. ion) 393.2 (b2*) 556.24 [M + H - H2O]+ 613.2 (b4*-H2O) 631.1 (b4*) 875.2 [M + H - H2O]+

+

+

-

42.5

+

+

+

66.9 67.9

*, HNE adducted site; °, carboxamidomethyl cysteine.

Figure 4. LC-ESI-MS/MS analysis in MRM mode of enzyme-digested HSA reacted with an excess of HNE (1:5 molar ratio) (A) and of HSA isolated from plasma incubated in the absence (B) and presence of 500 µM (C), 100 µM (D), and 10 µM HNE (E). Left panel: LC-ESI-MS/ MS of trypsin-digested HSA setting the parent ion at m/z 544.5; panel F shows the MS/MS spectrum relative to the peak at RT 42.1, attributed to the SB-adducted peptide LK*CASLQK. Right panel: LC-ESI-MS/ MS of trypsin + chymotrypsin-digested HSA setting the parent ion at m/z 893.5; panel F shows the MS/MS spectrum relative to the peaks at RT 66.9 and 67.9, attributed to the Michael-adducted peptide LQQC*PF. The asterisk indicated the adducted aa.

as well as in human plasma spiked with HNE at a final concentration of 500 and 100 µM, while it was absent in the control. The adduct was still well detectable (S/N > 3) when the HNE concentration was reduced to 10 µM. Similar results were obtained by analyzing the LQQCPF peptide adducted by HNE at the Cys34 residue (parent ion at m/z 893.5). In this case, a set of two peaks was identified, attributed to the four diastereoisomers of the NaBH4-reduced HNE adduct, and they were still easily identified even when the HNE concentration was reduced to 10 µM. Kinetics of HNE Reaction with HSA: Reaction Order and Rate Constant. The rate law of GSH with HNE was first determined at pH 7.4 and at 37 and 25 °C (pseudo first-order conditions), by using both the UV approach previously described by Doorn et al. (31) and the HPLC method here reported. The results obtained by the two methods were not significantly different, and the rate constant at 25 °C was 1.92 ( 0.12 M-1 s-1, similar to that found in the literature (1.33 M-1 s-1) (31), thus indicating the goodness of the HPLC approach; the rate constant determined at 37 °C was 3.81 ( 0.17 M-1 s-1. The kinetics of HNE reaction with HSA was then studied by using only the HPLC method, because HSA interferes in the UV method (HNE absorption at 224 nm). The reaction order with respect to HSA and HNE was first determined. Figure 5 shows the time-dependent HNE consumption for solutions containing a fixed starting concentration of 20 µM HSA and different HNE concentrations (5, 10, 20, 40, and 60 µM) (Figure 5A) and a fixed HNE concentration (20 µM) and varying the HSA concentrations (5, 10, 20, 40, and 60 µM) (Figure 5B). The calculated slopes relative to the Log - Log plots of the initial rates of HNE consumption vs the variable reactant concentration were 1.08 and 1.02 for varying HNE and HSA concentrations, respectively. None of the slopes was significantly different from

830

Chem. Res. Toxicol., Vol. 21, No. 4, 2008

Aldini et al.

Figure 6. Direct infusion ESI-MS of HSA isolated from human plasma: (A) native HSA, (B) HSA reduced by DTT treatment (HSA-Cys34SH), and (C) HSA reduced by DTT and then treated with IAA (HSACys34-S-IAA).

Figure 5. Kinetic traces of HNE loss in the presence of HSA. (A) Time-dependent HNE consumption for solutions containing a fixed starting concentration of 20 µM HSA and varying the HNE concentrations: 20 (2), 40 (b), and 60 (0) µM and (B) a fixed HNE concentration (20 µM) and varying HSA concentration: 20 (2), 40 (b), and 60 (0) µM.

1.0, as determined by the overlaps of their 95% confidence intervals with 1.0. This indicates a first-order reaction in terms of each reactant, and a second-order overall with rate ) k[HNE][HSA]. The approach to determine the rate constant is usually simplified by the isolation method in which the concentrations of all of the reactants except one are in large excess. However, for HNE reaction with HSA, it was not possible to achieve a pseudofirst-order condition, because a 10fold excess of HSA with respect to HNE greatly affected the precision and accuracy of the HPLC method used to measure the HNE content. Hence, the reactants were used in the following molar ratios: 1:0.2, 1:0.5, 1:1, 1:2, and 1:3, and according to the second-order kinetic, the 1/concentration vs time plots were drawn to calculate the rate constant. This simplified calculation approach was chosen since the rate constants were determined in a time range characterized by a HNE consumption lower than 10% of the initial amount. For the following reaction schemes:

HNE + HSA-Cys34-S-IAA (k1) f products HNE + HSA-Cys34-SH (k2) f products k1 and k2 are the rate constants of HNE adduction with IAA sulfhydryl-blocked HSA (HSA-Cys34-S-IAA) and DTT-reduced HSA (HSA-Cys34-SH), respectively. The identity and purity of adducted and reduced HSA were determined by mass spectrometry and by determining the sulfhydryl to albumin molar ratio (R). In Figure 6A is shown the positive-ion ESI mass spectrum of native HSA (scan range m/z 1410–1500) characterized by three abundant ions at m/z 1415.10, 1445.90, and 1478.00 and attributed to the +47, +46, and +45 multicharged ions of native HSA (compound c1, MW determined by deconvolution analysis ) 66446 Da). The MS spectrum is also characterized by two other series of ions attributed to the cysteinylated (compound c2) and glycated (compound c3) forms of HSA, as previously reported (19). The MS spectrum of DTTreduced HSA (Figure 6B) is characterized by only two ion

Figure 7. (A) Representative kinetic plot of 1/[HNE] vs time of HNE (20 µM) incubated with a 1:3 molar ratio of HSA-Cys34-SH (9) and HSA-Cys34-S-IAA (0). The second-order rate constant was calculated by the slope of the kinetic plot K′ vs [HSA]/[HNE].

series, attributed to the reduced (compound c1) and glycated (compound c3) forms of HSA, while the peaks relative to the cysteinylated form completely disappeared, to indicate the efficacy of the reduction. Panel C shows the ESI-MS spectrum of IAA-treated HSA. The deconvoluted spectrum indicates that the MW of both c1 and c3 HSA species was increased by 59 Da (MW) 66524 Da, compound c4; MW ) 66667 Da, compound c5). This indicates that in our experimental conditions, only Cys34 was reduced by DTT and converted to the carboxamidomethylated derivative (theoretical mass increase 57 Da). The amount of Cys34 in reduced state was then determined spectrophotometrically. The R values obtained for HSA-SR and HSA-Cys34-SH were 0.01 and 0.96, respectively, to indicate that the yield of the reactions was almost quantitative. The rate constant of the reaction of HNE with HSA was then determined. As an example, Figure 7 shows the 1/[HNE] vs time of a reaction mixture containing HSA-Cys34-SH and HNE in a 3:1 molar ratio. The plot of 1/[HNE] gives a straight line (R2 ) 0.9980), thus to confirm the second-order reaction. The rate constant of HNE reaction with HSA-Cys34-SH was 50.61 ( 1.89 M-1 s-1, as determined from the slope of the plot of k vs

HSA as Plasma Target of RCS

Chem. Res. Toxicol., Vol. 21, No. 4, 2008 831

Figure 9. Dynamic profile of accessible surface (expressed in Å2) for Cys34 residue as obtained by MD simulations of albumin with Cys34 neutral thiol (normal line) and Cys34 thiolate anion (dashed line).

Figure 8. (A) Deprotonation equilibrium for Cys34 thiol function. On the left, it can be noted that the polarity of the thiol group is markedly increased by closeness of His39 and Tyr84, while in the right the thiolate anion appears stabilized by interactions with Lys41 and His39. It is worth observing that Tyr84, which acts as a proton acceptor, moves away from Cys34 in its thiolate form, explaining the increased accessibility that Cys34 shows in its anionic form. (B) The quite superimposable arrangement of residues in GST1a-1a enzyme surrounding the bound glutathione. Also here, the acidity of GSH thiol group is enhanced by the closeness with one histidine and one tyrosine, which can act as a proton acceptor due to the closeness of a positive residue, which stabilizes the phenate. Moreover, a second tyrosine residue (Tyr115) can further increase the reactivity of GSH.

[HSA]/[HNE]. The slope of the plot significantly decreased to 21.24 ( 1.02 M-1 s-1 for sulfhydryl-blocked HSA (Figure 7B). The rate constant of HNE reaction with the single sulfhydryl of HSA, calculated as the difference between k2 and k1, was 29.37 M-1 s-1. By considering a second-order reaction and a mean plasma concentration of 600 nmol mL-1, the half-life of 10 nmol mL-1 of HNE in human plasma would be less than 17 s. This is in accordance with the experimental data obtained in human plasma, showing a half-life of less than 40 s, and in diluted human plasma (HSA, 75 µM final concentration), where the half-life of 10 nmol min-1 was 2.5 min (calculated, 2.2 min). Computational Studies: Understanding the Cys34 Reactivity. Because the high reactivity of Lys199 was already investigated by in silico studies (32), which unveiled the influence of the protonation state on its binding capacities, the computational studies were here focused on Cys34 with the aim to rationalize its remarkable reactivity toward RCS at an atomic level. Hence, the first analysis of the optimized albumin structure involved the prediction and rationalization of ionization constant for Cys34 thiol group using the empirical approach PropPka, which suggested that such thiol function is significantly more acid than in a free cysteine (i.e., pKa equal to 6.55 instead of 8.22). Even empirically predicted, this result is in line with experimental data (33) and well understandable for the surrounding residues (as detailed in Figure 8), which are able to both increase the polarity of thiol group, attracting the proton (e.g., Asp38, His39, and Tyr84) and stabilizing the thiolate anion

(e.g., His39, Lys41, and Tyr84). Figure 8 reports the plausible mechanism with which the Cys34 acidity is increased. It involves a first step in which the thiol proton is grabbed by His39 or, more probably, by Tyr84, whose phenate anion is stabilized by close Lys41 as proven by mutagenesis (34), which unveiled that Tyr84 influences the Cys34 acidity more than His39 and also modulates the accessibility of thiol group (as later suggested by MD simulations). In a second step, the Cys34 thiolate is stabilized by ion pair with Lys41 and π-sulfur interaction with His39. Intriguingly, there is a clear similarity between the architecture of key residues around Cys34 in HSA and the active site of glutathione-S-transferase M1a-1a (as reported in Figure 8; 35). Indeed, in both cases, the acidity of the reactive thiol group is enhanced by the closeness of one tyrosine and one histidine residue able to accept the thiol proton and stabilize the thiolate anion. Again, in both binding sites, a positively charged residue near to the tyrosine (a lysine in albumin and an arginine in GT, as seen in Figure 8) increases the polarity of phenol group and stabilizes the thiolate anion. Overall, such analogy suggests that the albumin can act with an autoenzymatic activity to maximize the reactivity of Cys34 thiol group toward RCS and implies that at physiological pH there is an equilibrium between neutral thiol and anionic thiolate in which both forms are markedly populated. Although this first finding can explain the significant reactivity of Cys34 residue toward HNE, the effect on Cys34 thiolate anion on albumin folding was explored. Two 5 ns MD simulations of albumin with Cys34 thiol group neutral and in its anionic state were carried out to clarify whether conformational shifts occurring in the thiolate form can further affect the Cys34 reactivity, which in the starting albumin structure is limited because it is nearly inaccessible as one can argue from Figure 9. A preliminary analysis of performed simulations concerned the folding stability of albumin as monitored considering the percentage of residues, which fall in the allowed regions of the Ramachandran plot (as computed by VEGA). In both simulations, such percentages rise at the beginning and then remain quite constant during all simulation period, confirming an overall folding stability of both ionization states (data not shown). This suggests that the conformational shifts here described are mainly due to the dynamic response of these structures at the equilibrium and not to unphysical structural distortions. A more specific analysis concerns the differences in Cys34 accessibility as monitored in the two MD simulations. Figure 9 clearly shows that the accessible surface of Cys34 is significantly greater in the thiolate form than in the neutral cysteine. Such different profiles suggest that the shift from buried to accessible

832

Chem. Res. Toxicol., Vol. 21, No. 4, 2008

Aldini et al.

Table 2. Dynamic Profiles (Namely, Minimum, Maximum, Average Values, and Standard Deviations) of the Main Interactions Between Cys34 Sulfur Atom and Principal Surrounding Residues in Albumin Structures as Obtained by MD Simulations (All Values Are Expressed in Å) interaction

min

max

mean

SD

Cys34-SH-Lys41 Cys34-S-Lys41 Cys34-SH-His39 Cys34-S-His39 Cys34-SH-Tyr84 Cys34-S-Tyr84

3.25 1.61 2.18 1.92 3.75 4.25

11.66 3.71 15.97 13.85 16.34 21.62

7.81 2.73 8.86 6.82 7.40 13.68

1.46 0.66 3.25 3.35 2.16 5.51

cysteine is promoted by the anionic form. A detailed analysis of the structural differences between MD simulations (as compiled in Table 2) sheds light on the key role of Tyr84 mobility that, while in the neutral Cys34 state, remains near to the thiol during a large part of the simulation period, in the anionic state moves away from sulfur atom, which realizes stable ion -pair with Lys41 as confirmed by very low standard deviation of monitored distance values, which can be roughly considered as an index for interaction stability. Conversely, Table 2 unveils that His39 shows a similar behavior in the two MD runs and, hence, cannot vastly influence the Cys34 exposition in the thiolate state. The accessibility of Cys34 residue in thiolate state is also induced by a set of apolar residues (e.g., Leu31 and Val 46), which during MD simulation moves away from the sulfur atom and lines the hydrophobic cavity, which accommodates the HNE skeleton (as observed in the following docking results). Taken together, the MD simulations confirm the key role of some residues in determining the significant reactivity of the Cys34 thiol group, increasing its acidity and/or stabilizing the thiolate form. In particular, there are some residues (e.g., His39) that are actively involved in both phases and other residues, which primarily increase the acidity of thiol group (e.g., Tyr84) or stabilize the corresponding thiolate anion (e.g., Lys41). Noticeably, the mobility of the residues, which only increases the acidity of thiol group, also favors the accessibility of Cys34, thus explaining why this residue is markedly more accessible in its anionic state. Figure 10 shows the best complex as obtained by docking (R)-HNE in the albumin structure as detailed under the Experimental Procedures. Docking results reveal the key closeness between the sulfur atom and the C3 atom in (R)-HNE and suggest how the unsaturated aldehyde is here prone to Michael adduction. The complex also emphasizes the marked role of Thr76, which realizes a strong H-bond with the carbonyl group. Conversely, the hydroxyl group is not involved in polar interactions, suggesting that albumin can equally react with both HNE enantiomers and indeed docking analysis realized with (S)-HNE gave quite identical results (data not shown). Finally, the carbon skeleton of (R)-HNE contacts many apolar residues (e.g., Phe27, Ala28, Leu31, Val 46, Phe70, Leu74, and Tyr140), which line the cavity where it is accommodated.

Discussion LDL is susceptible to oxidative damage, and oxidatively modified LDL plays a key role in the development of atherosclerosis lesions. LDL oxidation is associated with the formation of a number of highly reactive molecules and, among them, the reactive and electrophilic carbonyl species (RCS), which are involved in the vascular inflammation and fibrosis (1, 2). We believe that to better understand the pro-atherogenic mechanisms of RCS, it is important to have a clear picture of

Figure 10. Best complex obtained by docking (R)-HNE in the albumin structure. It is worth noting the remarkable closeness between the sulfur atom and the C3 of (R)-HNE, which assumes a pose clearly conducive to Michael adduction.

the RCS detoxification pathway in plasma, which represents the main vascular site of RCS formation and where dietary and tissue-derived RCS are delivered (36) and the main proatherogenic target proteins are formed. HNE, chosen as a model of RCS, was found to be rapidly consumed in plasma due to the covalent adduction with HSA. The high reactivity of HSA toward HNE has already been demonstrated by Szapacs et al. (37), using a kinetic analysis of competing alkylation reactions, as well as by us using a mass spectrometric approach (19). Moreover, Tallman et al. (38), by using a headgroup biotinylated phosphatidylcholine, has recently demonstrated that albumin is one of the major plasma protein targets of electrophilic aldehydes generated by the oxidative decomposition of the biotinylated lipid. The high reactivity of HSA to HNE is due to the different accessible nucleophilic amino acids and in particular to Lys199 and Cys34, this last one accounting for the most part of the trapping effect, as demonstrated by kinetic analyses. The high nucleophilic reactivity of Lys199 has been already studied, and this is due to its unusual low pkA (of ≈8) (32), which implies the existence of a significant amount of the neutral form at physiological pH, which is responsible for the nucleophilic attack on the electrophilic substrates. Here, we fully characterized the high reactivity of Cys34 by means of molecular modeling and kinetic analysis. The rate constant of Cys34 was found almost 1 order of magnitude higher than that of GSH, and the differences in reactivity are well-explained by molecular modeling results, which confirmed the significant acidity of Cys34 thiol group and unveiled a remarkable similarity between the residues surrounding Cys34 and the catalytic site of certain glutathione transferases. This indicates that albumin can maximize the reactivity of Cys34 thiol group through a kind of autoenzymatic mechanism. Overall, the computational results confirm that

HSA as Plasma Target of RCS

ionization properties and accessibility are key factors to improve the reactivity of nucleophilic residues toward RCS and emphasize that such factors can act in a concerted way, since Cys34 seems to be markedly more accessible in anionic form, as observed in MD simulations. The detoxification pattern of reactive and electrophilic aldehydes has been fully characterized in cells and tissues, and it is mediated by both phase I and II enzymes (39, 8). More recently, some abundant and high reactive proteins toward reactive aldehyde(s)/carbonyl(s) have been suggested to act as endogenous carbonyl scavengers, protecting more susceptible proteins thus to play an important role in RCS detoxification. For instance, epithelial fatty acid-binding (E-FABP) protein has been supposed to function as an antioxidant protein, protecting the integrity of the cellular environment through inactivation of reactive lipids through Cys-120 (40). Wataya et al. (41) found that HNE adduction to neurofilament heavy subunit (NFH) is physiological and may be a role for neurofilaments in augmenting the scavenging of the aldehydes by neurons, thus protecting other proteins from damage. Pedersen et al. (42) reported that apolipoprotein E plays a major role in detoxifying HNE and in protecting against the toxic effects of HNE, particularly in neurodegenerative processes. A protein, in order to act as a scavenger of RCS, should be present in a high relative concentration in comparison with other potential nucleophilic protein targets, which should be highly reactive toward reactive aldehyde(s)/carbonyl(s); more importantly, the carbonylation reaction should not affect the protein function. We have recently suggested that actin partially fulfills these requirements, being a relatively abundant protein in most mammalian cells and, through its highly reactive Cys374 residue, scavenges reactive electrophilic aldehydes, up to C9, such as acrolein and HNE (43, 44), without its resulting function (in particular, the polymerization process) being impaired. In the present work, we found that albumin, through its reactive nucleophilic sites, acts as a protein carbonyl scavenger in the extracellular fluid, which contain only small amounts of GSH (1.5–4 µM in human plasma with respect to 1–10 mM in cells). HSA seems to act as a protein carbonyl scavenger, protecting more vulnerable circulating proteins such as apoprotein B and PDGFRβ, whose covalent adduction would result in a proatherogenic mechanism. Hence, we suggest that blood electrophilic aldehydes, beside being rapidly hydrolyzed by an enzymatic pathway in the erythrocytes, undergo a nonenzymatic detoxification pathway in plasma, mainly mediated by HSA. The importance of albumin in the RCS vascular detoxification is further sustained by considering the existence of an erythrocytefree zone of streaming plasma along the endothelium (45), which creates a diffusional barrier between RCS and the erythrocytic detoxification system. The kinetic studies carried out in plasma and plasma fractions suggest that the low molecular weight thiols quench HNE only in a very limited extent (10 kDa), and HSA in particular, represents the main nonenzymatic pathway for HNE removal. Interestingly, when HSA was depleted, the HNE quenching was reduced by 3-fold, to suggest not only the pivotal role of HSA but also that other plasma proteins, in the absence of HSA, react with HNE. Further studies are in progress to identify these target proteins and to clarify whether their function results were impaired upon carbonylation. The plasma content of free HNE in normal healthy individuals has been determined in several previous studies and falls within the nanomolar concentration range. A significant increase with

Chem. Res. Toxicol., Vol. 21, No. 4, 2008 833

respect to controls was found in different pathological conditions, such as autoimmune disease (46), congestive heart failure (47), and kidney disease (48), but also, in these conditions, the concentration of free HNE, although greater than that of controls, was in the nanomolar range. The low level of free HNE in physiopathological conditions may be the consequence of a rapid and efficient clearance of free HNE due to the HSA adduction. In this case, once the reaction equilibrium between HNE and HSA is reached, only trace amounts of the originally present HNE may be detected. This mechanism has been confirmed by Salomon et al. (49), who reported higher levels of HNE protein adducts with respect to the free form. In particular, by using an enzyme-linked immunosorbent assay and antibodies raised against 2-pentylpyrrole epitopes (generated by reaction of HNE with protein lysyl residues), it was found that the level of adducted HNE in normal healthy individuals was 8 ( 1.6 µM, which increased to 15.2 ( 6.2 and 24.3 ( 7.0 in patients with atherosclerosis and end-stage renal disease, respectively, a concentration at least 1 order of magnitude higher than that of free HNE. Hence, we can reasonably suppose that the initial plasma concentration of HNE in physiopathological conditions may be at least micromolar. For this reason, the quenching effect of HSA and of the different plasma fractions was assayed by using micromolar concentrations of HNE (10–500 µM for LCESI-MS, 20 µM for the quenching studies). In view of these considerations, we believe that to obtain a real estimate of circulating HNE, the HNE protein adducts should be determined instead of free HNE. However, because most of the studies measuring plasma-adducted HNE proteins only give a semiquantitative analysis, the overall HNE absolute plasma concentration is substantially lacking. The role of Cys34 albumin in the vascular RCS detoxification pathway could partially explain the high CVD risk for uremic patients on chronic hemodialysis (HD) therapy. For these subjects, the annual mortality caused by cardiovascular cause is approximately 9%, which is 10-20-fold higher than the general population, even when adjusted for age, sex, race, and the presence or absence of diabetes. Several studies found that plasma thiol levels of chronic HD patients were lower than the normal subjects (50). Terawaki et al. (51) found that the oxidation status of Cys34 in HSA was enhanced in correlation with the level of renal dysfunction among HD patients. In particular, in uremic patients, Cys34 can undergo different modifications such as an oxidation to sulfenic, sulfonic, and sulfinic acid, homocysteinylation (52), or cysteinylation, as we recently observed (manuscript in preparation). Diabetes is another well-established condition that profoundly accelerates the development of atherosclerosis and increases the morbidity and mortality of cardiovascular events (53). The carbonyl stress induced by RCS, including glyoxal and methylglyoxal, is considered a mechanism involved in micro- and macrovascular complications in both uremic and diabetic patients (54, 55). It is interesting to note that also in diabetic patients the plasma thiol content has been found to be significantly reduced (56, 57), and this is probably due to the covalent adduction of R-dicarbonyls to albumin Cys34 (58). In the case of glyoxal, a S-(carboxymethyl)cysteine derivative has been identified (59). Oxidation and carbonylation of Lys residues (carboxy methyl lysine modification) of HSA have been shown to substantially contribute to the risk of cardiovascular events in HD patients, by triggering the neutrophil oxidative burst and sustaining inflammation, which plays a key role in the pathogenesis of atherosclerosis (54). Anyway, we strongly believe that in those subjects undergoing a massive carbonylation and oxidative

834

Chem. Res. Toxicol., Vol. 21, No. 4, 2008

stress, such as nephropatic and diabetic patients (60, 54), a substantial oxidation/covalent adduction of Cys34 takes place, thus making circulating proteins such apoprotein B and PDGFR more vulnerable to electrophilic aldehydes, whose covalent adduction would result in a pathogenetic mechanism. This represents the aim of our current studies, focused on identifying the modifications occurring on HSA-Cys 34 isolated from uremic subjects undergoing hemodialysis (preliminary results indicate massive S-cysteinylation) and correlating them with the cardiovascular risk (determination of surrogate and hard end points of organ damage). Additional evidence regarding the protective effect of HSA is given by several studies suggesting that a low serum albumin level is a strong independent predictor of cardiovascular mortality (61, 62). In other terms, a low amount of albumin would represent a reduced ability of the vasculature to counteract the RCS-mediated carbonylation. In conclusion, in the present study, we found that albumin, through nucleophilic residues and in particular Cys34 and Lys199, can act as detoxifying agent of RCS. The lack of this residue (due to oxidation, cysteinilation, or massive covalent adduction) would result in a significant impairment of the vascular protection against atherogenic aldehydes, leading to the carbonylation of key functional proteins and tissue damage. This study also suggests that the development of therapeutic approaches to control modification of albumin Cys34 may be beneficial in preventing cardiovascular complications in nephropatic and diabetic patients and, in general, in those pathological conditions characterized by a significant oxidative/carbonyl stress. Acknowledgment. We thank Professors A. Gavezzotti, I. Dalle-Donne, and A. Milzani for the critical review of this manuscript. This study was partially supported by FIRST 2006 (Fondo Interno Ricerca Scientifica e Tecnologica, University of Milan). Supporting Information Available: Deconvoluted ESI-MS spectrum of HSA isolated from human plasma by affinity chromatography and using 1 M NaCl as an eluent (Figure S1) and MALDI-TOF mass fingerprint analysis of plasma HSA extracted from the spot excised from two-dimensional gels and subjected to trypsin digestion (Figure S2). This material is available free of charge via the Internet at http://pubs.acs.org.

References (1) Uchida, K. (2000) Role of reactive aldehyde in cardiovascular diseases. Free Radical Biol. Med. 28, 1685–1696. (2) Leonarduzzi, G., Chiarpotto, E., Biasi, F., and Poli, G. (2005) 4-Hydroxynonenal and cholesterol oxidation products in atherosclerosis. Mol. Nutr. Food Res. 49, 1044–1049. (3) Poli, G., and Schaur, R. J. (2000) 4-Hydroxynonenal in the pathomechanisms of oxidative stress. IUBMB Life 50, 315–321. (4) Uchida, K. (2000) Role of reactive aldehyde in cardiovascular diseases. Free Radical Biol. Med. 28, 1685–1696. (5) Berliner, J., and Heinecke, J. W. (1996) The role of oxidized lipoproteins in atherogenesis. Free Radical Biol. Med. 20, 707–727. (6) Vindis, C., Escargueil-Blanc, I., Elbaz, M., Marcheix, B., Grazide, M. H., Uchida, K., Salvayre, R., and Negre-Salvayre, A. (2006) Desensitization of platelet-derived growth factor receptor-beta by oxidized lipids in vascular cells and atherosclerotic lesions: Prevention by aldehyde scavengers. Circ. Res. 98, 785–792. (7) Tamamizu-Kato, S., Wong, J. Y., Jairam, V., Uchida, K., Raussens, V., Kato, H., Ruysschaert, J. M., and Narayanaswami, V. (2007) Modification by acrolein, a component of tobacco smoke and agerelated oxidative stress, mediates functional impairment of human apolipoprotein E. Biochemistry 46, 8392–8400. (8) Ellis, E. M. (2007) Reactive carbonyls and oxidative stress: Potential for therapeutic intervention. Pharmacol. Ther. 115, 13–24.

Aldini et al. (9) Yang, Y., Yang, Y., Trent, M. B., He, N., Lick, S. D., Zimniak, P., Awasthi, Y. C., and Boor, P. J. (2004) Glutathione-S-transferase A4-4 modulates oxidative stress in endothelium: Possible role in human atherosclerosis. Atherosclerosis 173, 211–221. (10) Takagi, S., Iwai, N., Yamauchi, R., Kojima, S., Yasuno, S., Baba, T., Terashima, M., Tsutsumi, Y., Suzuki, S., Morii, I., Hanai, S., Ono, K., Baba, S., Tomoike, H., Kawamura, A., Miyazaki, S., Nonogi, H., and Goto, Y. (2002) Aldehyde dehydrogenase 2 gene is a risk factor for myocardial infarction in Japanese men. Hypertens. Res. 25, 677– 681. (11) Srivastava, S., Dixit, B. L., Cai, J., Sharma, S., Hurst, H. E., Bhatnagar, A., and Srivastava, S. K. (2000) Metabolism of lipid peroxidation product, 4-hydroxynonenal (HNE) in rat erythrocytes: Role of aldose reductase. Free Radical Biol. Med. 29, 642–651. (12) Michelet, F., Gueguen, R., Leroy, P., Wellman, M., Nicolas, A., and Siest, G. (1995) Blood and plasma glutathione measured in healthy subjects by HPLC: Relation to sex, aging, biological variables, and life habits. Clin. Chem. 41, 1509–1517. (13) Dockham, P. A., Sreerama, L., and Sladek, N. E. (1997) Relative contribution of human erythrocyte aldehyde dehydrogenase to the systemic detoxification of the oxazaphosphorines. Drug Metab. Dispos. 25, 1436–1441. (14) Uchida, K., Osawa, T., Hiai, H., and Toyokuni, S. (1995) 4-Hydroxy2-nonenal-trapping ELISA: Direct evidence for the release of a cytotoxic aldehyde from oxidized low density lipoproteins. Biochem. Biophys. Res. Commun. 212, 1068–1073. (15) Alary, J., Bravais, F., Cravedi, J. P., Debrauwer, L., Rao, D., and Bories, G. (1995) Mercapturic acid conjugates as urinary end metabolites of the lipid peroxidation product 4-hydroxy-2-nonenal in the rat. Chem. Res. Toxicol. 8, 34–39. (16) Rees, M. S., van Kuijk, F. J., Siakotos, A. N., and Mundy, B. P. (1995) Improved synthesis of various isotope-labelled 4-hydroxyalkenals and peroxidation intermediates. Synth. Commun. 25, 3225–2236. (17) Bar-Or, D., Heyborne, K. D., Bar-Or, R., Rael, L. T., Winkler, J. V., and Navot, D. (2005) Cysteinylation of maternal plasma albumin and its association with intrauterine growth restriction. Prenatal Diagn. 25, 245–249. (18) Glowacki, R., and Jakubowski, H. (2004) Cross-talk between Cys34 and lysine residues in human serum albumin revealed by Nhomocysteinylation. J. Biol. Chem. 279, 10864–10871. (19) Aldini, G., Gamberoni, L., Orioli, M., Beretta, G., Regazzoni, L., Maffei Facino, R., and Carini, M. (2006) Mass spectrometric characterization of covalent modification of human serum albumin by 4-hydroxy-trans-2-nonenal. J. Mass Spectrom. 41, 1149–1161. (20) Riener, C. K., Kada, G., and Gruber, H. J. (2002) Quick measurement of protein sulfhydryls with Ellman’s reagent and with 4,4′-dithiodipyridine. Anal. Bioanal. Chem. 373, 266–276. (21) Laemmli, U. K. (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227, 680–685. (22) Sugio, S., Kashima, A., Mochizuki, S., Noda, M., and Kobayashi, K. (1999) Crystal structure of human serum albumin at 2.5 A resolution. Protein Eng. 12, 439–446. (23) Pedretti, A., Villa, L., and Vistoli, G. (2002) VEGA: A versatile program to convert, handle and visualize molecular structure on windows-based PCs. J. Mol. Graph. 21, 47–49. (24) Porter, M. A., Hall, J. R., Locke, J. C., Jensen, J. H., and Molina, P. A. (2006) Hydrogen bonding is the prime determinant of carboxyl pKa values at the N-termini of alpha-helices. Proteins 63, 621–635. (25) Kalé, L., Skeel, R., Bhandarkar, M., Brunner, R., Gursoy, A., Krawetz, N., Phillips, J., Shinozaki, A., Varadarajan, K., and Schulten, K. (1999) NAMD2: Greater scalability for parallel molecular dynamics. J. Comp. Physiol. 151, 283–312. (26) McGann, M., Almond, H., Nicholls, A., Grant, J. A., and Brown, F. K. (2003) Gaussian docking functions. Biopolymers 68, 76–90. (27) Kinscherf, R., Cafaltzis, K., Röder, F., Hildebrandt, W., Edler, L., Deigner, H. P., Breitkreutz, R., Feussner, G., Kreuzer, J., Werle, E., Michel, G., Metz, J., and Dröge, W. (2003) Cholesterol levels linked to abnormal plasma thiol concentrations and thiol/disulfide redox status in hyperlipidemic subjects. Free Radical Biol. Med. 35, 1286–1292. (28) Andersson, A., Hutlberg, B., and Lindgren, A. (2000) Redox status of plasma homocysteine and other plasma thiols in stroke patients. Atherosclerosis 151, 535–539. (29) Carru, C., Deiana, L., Sotgia, S., Pes, G. M., and Zinellu, A. (2004) Plasma thiols redox status by laser-induced fluorescence capillary electrophoresis. Electrophoresis 25, 882–889. (30) Williams, R. H., Maggiore, J. A., Reynolds, R. D., and Helgason, C. M. (2001) Novel approach for the determination of the redox status of homocysteine and other aminothiols in plasma from healthy subjects and patients with ischemic stroke. Clin. Chem. 47, 1031–1039. (31) Doorn, J. A., and Petersen, D. R. (2002) Covalent modification of amino acid nucleophiles by the lipid peroxidation products 4-hydroxy2-nonenal and 4-oxo-2-nonenal. Chem. Res. Toxicol. 15, 1445–1450.

HSA as Plasma Target of RCS (32) Diaz, N., Suarez, D., and Sordo, T. L., Jr. (2001) A theoretical study of the aminolysis reaction of lysine 199 of human serum albumin with benzylpenicillin: Consequences for immunochemistry of penicillins. J. Am. Chem. Soc. 123, 7574–783. (33) Narazaki, R., Hamada, M., Harada, K., and Otagiri, M. (1996) Covalent binding between bucillamine derivatives and human serum albumin. Pharm. Res. 13, 1317–1321. (34) Stewart, A. J., Blindauer, C. A., Berezenko, S., Sleep, D., Tooth, D., and Sadler, P. J. (2005) Role of Tyr84 in controlling the reactivity of Cys34 of human albumin. FEBS J. 272, 353–362. (35) Patskovsky, Y., Patskovska, L., Almo, S. C., and Listowsky, I. (2006) Transition state model and mechanism of nucleophilic aromatic substitution reactions catalyzed by human glutathione S-transferase M1a-1a. Biochemistry 45, 3852–3862. (36) Links Surh, J., and Kwon, H. (2005) Estimation of daily exposure to 4-hydroxy-2-alkenals in Korean foods containing n-3 and n-6 polyunsaturated fatty acids. Food Addit. Contam. 22, 701–708. (37) Szapacs, M. E., Riggins, J. N., Zimmerman, L. J., and Liebler, D. C. (2006) Covalent adduction of human serum albumin by 4-hydroxy2-nonenal: kinetic analysis of competing alkylation reactions. Biochemistry 45, 10521–10528. (38) Tallman, K. A., Kim, H. Y., Ji, J. X., Szapacs, M. E., Yin, H., McIntosh, T. J., Liebler, D. C., and Porter, N. A. (2007) Phospholipidprotein adducts of lipid peroxidation: Synthesis and study of new biotinylated phosphatidylcholines. Chem. Res. Toxicol. 20, 227–234. (39) Alary, J., Gueraud, F., and Cravedi, J. P. (2003) Fate of 4-hydroxynonenal in vivo: Disposition and metabolic pathways. Mol. Aspects Med. 24, 177–187. (40) Bennaars-Eiden, A., Higgins, L., Hertzel, A. V., Kapphahn, R. J., Ferrington, D. A., and Bernlohr, D. A. (2002) Covalent modification of epithelial fatty acid-binding protein by 4-hydroxynonenal in vitro and in vivo. Evidence for a role in antioxidant biology. J. Biol. Chem. 277, 50693–50702. (41) Wataya, T., Nunomura, A., Smith, M. A., Siedlak, S. L., Harris, P. L., Shimohama, S., Szweda, L. I., Kaminski, M. A., Avila, J., Price, D. L., Cleveland, D. W., Sayre, L. M., and Perry, G. (2002) High molecular weight neurofilament proteins are physiological substrates of adduction by the lipid peroxidation product hydroxynonenal. J. Biol. Chem. 277, 4644–4648. (42) Pedersen, W. A., Chan, S. L., and Mattson, M. P. (2000) A mechanism for the neuroprotective effect of apolipoprotein E: Isoform-specific modification by the lipid peroxidation product 4-hydroxynonenal. J. Neurochem. 74, 1426–1433. (43) Aldini, G., Dalle-Donne, I., Vistoli, G., Maffei Facino, R., and Carini, M. (2005) Covalent modification of Actin by 4-hydroxy-trans-2nonenal (HNE): LC-ESI-MS/MS evidence for Cys374 Michael adduction. J. Mass Spectrom. 40, 946–954. (44) Dalle-Donne, I., Carini, M., Vistoli, G., Gamberoni, L., Giustarini, D., Colombo, R., Maffei Facino, R., Rossi, R., Milzani, A., and Aldini, G. (2007) Actin Cys374 as a nucleophilic target of alpha,betaunsaturated aldehydes. Free Radical Biol. Med. 42, 583–598. (45) Vaughn, M. W., Kuo, L., and Liao, J. C. (1998) Effective diffusion distance of nitric oxide in the microcirculation. Am. J. Physiol. 274, H1705–H1714. (46) Grune, T., Michel, P., Sitte, N., Eggert, W., Albrecht-Nebe, H., Esterbauer, H., and Siems, W. G. (1997) Increased levels of 4-hydroxynonenal modified proteins in plasma of children with autoimmune diseases. Free Radical Biol. Med. 23, 357–360. (47) Mak, S., Lehotay, D. C., Yazdanpanah, M., Azevedo, E. R., Liu, P. P., and Newton, G. E. (2000) Unsaturated aldehydes including 4-OH-

Chem. Res. Toxicol., Vol. 21, No. 4, 2008 835

(48)

(49)

(50) (51)

(52) (53) (54)

(55) (56) (57) (58) (59)

(60)

(61) (62)

nonenal are elevated in patients with congestive heart failure. J. Card. Failure 6, 108–114. Sommerburg, O., Grune, T., Hampl, H., Riedel, E., van Kuijk, F. J., Ehrich, J. H., and Siems, W. G. (1998) Does long-term treatment of renal anaemia with recombinant erythropoietin influence oxidative stress in haemodialysed patients? Nephrol., Dial., Transplant. 13, 2583–2587. Salomon, R. G., Kaur, K., Podrez, E., Hoff, H. F., Krushinsky, A. V., and Sayre, L. M. (2000) HNE-derived 2-pentylpyrroles are generated during oxidation of LDL, are more prevalent in blood plasma from patients with renal disease or atherosclerosis, and are present in atherosclerotic plaques. Chem. Res. Toxicol. 13, 557–564. Lim, P. S., Cheng, Y. M., and Yang, S. M. (2007) Impairments of the biological properties of serum albumin in patients on haemodialysis. Nephrology (Carlton) 12, 18–24. Terawaki, H., Yoshimura, K., Hasegawa, T., Matsuyama, Y., Negawa, T., Yamada, K., Matsushima, M., Nakayama, M., Hosoya, T., and Era, S. (1988) (2004) Oxidative stress is enhanced in correlation with renal dysfunction: examination with the redox state of albumin. Kidney Int. 66, 1993. Perna, A. F., Satta, E., Acanfora, F., Lombardi, C., Ingrosso, D., and De Santo, N. G. (2006) Increased plasma protein homocysteinylation in hemodialysis patients. Kidney Int. 69, 869–876. Reusch, J. E., and Draznin, B. B. (2007) Atherosclerosis in diabetes and insulin resistance. Diabetes Obes. Metab. 9, 455–463. Mera, K., Anraku, M., Kitamura, K., Nakajou, K., Maruyama, T., Tomita, K., and Otagiri, M. (2005) Oxidation and carboxy methyl lysine-modification of albumin: Possible involvement in the progression of oxidative stress in hemodialysis patients. Hypertens. Res. 28, 973–980. Pennathur, S., and Heinecke, J. W. (2007) Mechanisms for oxidative stress in diabetic cardiovascular disease. Antioxid. Redox Signaling 9, 955–969. Piwowar, A., Knapik-Kordecka, M., and Warwas, M. (2007) AOPP and its relations with selected markers of oxidative/antioxidative system in type 2 diabetes mellitus. Diabetes. Res. Clin. Pract. 77, 188–192. Matteucci, E., Malvaldi, G., Fagnani, F., Evangelista, I., and Giampietro, O. (2004) Redox status and immune function in type I diabetes families. Clin. Exp. Immunol. 136, 549–554. Faure, P., Troncy, L., Lecomte, M., Wiernsperger, N., Lagarde, M., Ruggiero, D., and Halimi, S. (2005) Albumin antioxidant capacity is modified by methylglyoxal. Diabetes Metab. 31, 169–77. Zeng, J., and Davies, M. J. (2005) Evidence for the formation of adducts and S-(carboxymethyl)cysteine on reaction of R-dicarbonyl compounds with thiol groups on amino acids, peptides, and proteins. Chem. Res. Toxicol. 18, 1232–1241. Toyokuni, S., Yamada, S., Kashima, M., Ihara, Y., Yamada, Y., Tanaka, T., Hiai, H., Seino, Y., and Uchida, K. (2000) Serum 4-hydroxy-2-nonenal-modified albumin is elevated in patients with type 2 diabetes mellitus. Antioxid. Redox Signaling 2, 681–685. Phillips, A., Shaper, A. G., and Whincup, P. H. (1989) Association between serum albumin and mortality from cardiovascular disease, cancer, and other causes. Lancet 2, 1434–1436. Kuller, L. H., Eichner, J. E., Orchard, T. J., Grandits, G. A., McCallum, L., and Tracy, R. P. (1991) The relation between serum albumin levels and risk of coronary heart disease in the Multiple Risk Factor Intervention Trial. Am. J. Epidemiol. 134, 1266–1277.

TX700349R