- Email: [email protected]
The influence of fatty acids on determination of human serum albumin thiol group
Analytical Biochemistry 448 (2014) 50–57
Contents lists available at ScienceDirect
Analytical Biochemistry journal homepage: www.elsevier.com/locate/yabio
The influence of fatty acids on determination of human serum albumin thiol group Vesna B. Jovanovic´ a, Ivan D. Pavic´evic´ a, Marija M. Takic´ b, Ana Z. Penezic´-Romanjuk a, Jelena M. Ac´imovic´ a, Ljuba M. Mandic´ a,⇑ a b
Department of Biochemistry, Faculty of Chemistry, University of Belgrade, Studentski trg 12-16, Belgrade 11158, Serbia Institute for Medical Research, Laboratory for Nutrition and Metabolism, University of Belgrade, Belgrade, Serbia
a r t i c l e
i n f o
Article history: Received 18 September 2013 Received in revised form 13 November 2013 Accepted 25 November 2013 Available online 4 December 2013 Keywords: HSA-thiol group determination Fatty acid Albumin isolation Affinity chromatography Diabetes
a b s t r a c t During investigation of the changes of the Cys34 thiol group of human serum albumin (HSA) (isolated by affinity chromatography with Cibacron Blue (CB)) in diabetes, we found that the HSA-SH content was higher (11–33%) than the total serum thiol content. The influence of fatty acids (FA) binding to HSA on this discrepancy was investigated in vitro (using fluorescence and CD spectroscopy and GC) and with HSA samples from diabetic (n=20) and control groups (n=17). HSA-bound FA determine the selection of HSA molecules by CB and enhance reactivity and/or accessibility of the SH group. A high content of polyunsaturated FA (35.6%) leads to weaker binding of HSA molecules to CB. Rate constants of DTNB reaction with the SH group of HSA applied to a CB column, bound-HSA and unbound-HSA fractions, were 4.810-3, 21.610-3, and 11.210-3 s-1, respectively. The HSA-SH group of diabetics is more reactive compared with control individuals (rate constants 20.910-3±4.410-3 vs 12.910-3±2.610-3 s-1, P<0.05). Recovery values of the SH group obtained after chromatography of HSA with bound stearic acid ranged from 110 to 140%, while those for defatted HSA were from 98.5 to 101.7%. Thus, HSA-bound FA leads to an increase of HSA-SH content and a contribution to total serum thiols, which make the determination of the thiol group unreliable. Ó 2013 Elsevier Inc. All rights reserved.
Human serum albumin (HSA)1 is the most abundant plasma protein (0.6 mM). It is a 66.5-kDa protein organized into three homologous domains (labeled I–III) and each domain comprises two subdomains (A and B) that share common structural elements [1,2]. A total of 17 disulfide bridges located exclusively within subdomains contribute toward HSA’s stability [3]. HSA transports many endogenous ligands such as long chain (C13–C21) fatty acids (FA), hemin, bilirubin, and thyroxin, all of which bind HSA with high affinity [3]. Although most ligands for HSA are hydrophobic anions, heavy metals are also known to bind to this protein [1,3–5]. Moreover, HSA has the ability to bind a wide variety of drug molecules and alter their pharmacokinetic parameters [6]. This binding occurs via hydrophobic cavities in subdomains IIA and IIIA, known as Sudlow I and Sudlow II, respectively
⇑ Corresponding author. Address: Department of Biochemistry, Faculty of Chemistry, University of Belgrade, P.O. Box 51, Studentski trg 16, 11158 Belgrade, Serbia. Fax: +381 11 2184 330. E-mail address: [email protected] (L.M. Mandic´). 1 Abbreviations used: BCG, bromocresol green; CB, Cibacron Blue; CD, circular dichroism; EPA, eicosapentaenoic acid; DHA, docosahexaenoic acid; DTNB, 5,50 dithiobis-(2-nitrobenzoic acid); FA, fatty acids; GC, gas chromatography; HSA, human serum albumin; HSA-SH, human serum albumin thiol content; PUFA, polyunsaturated fatty acids; RSD, relative standard deviation; TG, triglyceride; UV, ultraviolet. 0003-2697/$ - see front matter Ó 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.ab.2013.11.030
[7,8], with the sole tryptophan residue in HSA located in Sudlow I (Trp-214) [9]. HSA is the primary transporter for delivering FA to the tissues and possesses at least seven binding sites of varying affinities for this ligand (Fig. 1) [10–12]. Although none of the FA-binding sites are completely identical, each comprises a hydrophobic pocket that interacts with the hydrocarbon chain, while five binding sites cap FA at one end with basic or polar residues that interact closely with the carboxyl group of bound FA [13]. Under normal physiological conditions, between 0.1 and 2 mol of FA are bound to HSA, but the molar ratio of FA/HSA can rise above 6:1 in the peripheral vasculature during fasting or extreme exercise [5,14] or under pathological conditions such as diabetes, liver disease, and cardiovascular disease [3,15]. In order to completely understand the role of HSA in vivo, it is crucial to obtain detailed information about the variety of ligands that HSA binds, as well as how these ligands interact and influence each other during binding to HSA. Besides its role in transport, HSA is one of the most important extracellular antioxidants [3]. The one free cysteine-derived thiol (-SH) group (Cys34) (located in subdomain I), which can exist in both reduced and oxidized forms, provides a part of the antioxidant property of HSA. As HSA is the most abundant plasma protein
Influence of HSA-bound FA on the HSA-SH reactivity / V.B. Jovanovic´ et al. / Anal. Biochem. 448 (2014) 50–57
51
Materials and methods Chemicals and instrumentation All chemicals were purchased from Merck (Darmstadt, Germany) and Sigma–Aldrich, Chemie (Steinheim, Germany), unless otherwise noted. The 20% solution of HSA (96% purity, containing 0.42 mol -SH/mol HSA) was purchased from Baxter (Vienna, Austria), and Cibacron Blue F3G-A Sepharose CL-6B from Pharmacia Biotech (Uppsala, Sweden). All chemicals used were analytical grade. Spectrophotometric measurements were performed by using a Beckman DU-50 spectrophotometer (Fullerton, CA). Serum samples
Fig.1. The alignment of crystal structures of HSA with bound fatty acids: myristic (light gray), palmitic (black), and stearic (dark gray). Crystal structures of HSAmyristic (PDB ID, 1bj5) [11], HSA-palmitic (PDB ID, 1e7h) [12], and HSA-stearic (PDB ID, 1e7i) [12] are used to create the figure using Jmol [16]. Domains I, II, and III with corresponding subdomains (A and B) are labeled.
[3], it accounts for 80% of the total redox-reactive thiols in plasma [17]. The HSA-thiol group (HSA-SH) content decreases in some diseases such as liver and renal failure [18–20], diabetes mellitus [21], uremia [22], and obstructive sleep apnea [23], for which it is believed that oxidative stress plays an important role in pathogenesis. In addition, the increasing flux of reactive dicarbonyl compounds (glyoxal, methylglyoxal, and 3-deoxyglucosone), which occurs during carbonyl stress (in diabetes, Alzheimer’s disease, renal failure, liver cirrhosis, anemia, uremia, and atherosclerosis [24]), leads to the Cys34 side chain carbonylation and therefore to the decrease of the HSA-SH content [24]. It could be inferred that a differential content of HSA-SH can be a useful marker of oxidative and carbonyl stress during various pathological states, which lead to several advances in its quantification [25]. Binding of FA is associated with significant structural changes in the HSA molecule [10,22] which determine changes in its physical– chemical properties, such as an increase in the stability toward heat and proteases [3]. Moreover, structural changes could cause Cys34 residues to be more or less exposed to the surrounding environment, leading to differential reactivity and susceptibility to oxidative and carbonyl stress. This is of particular importance during pathological states (diabetes, liver and heart diseases) that are followed by an increase in serum-free FA content [3,15] and in the levels of carbonyl and/or oxidative reactive species [18–23]. A single-step affinity chromatographic method based on the specific interaction between the HSA and the textile dye Cibacron Blue F3G-A (CB) [26] has been largely applied for the isolation of HSA, particularly in recent years for studying the antioxidant properties of HSA [23,27]. When we employed this method while investigating the influence of oxidative and/or carbonyl stress in type 2 diabetes on the HSA Cys34 thiol group, we uncovered that the HSASH content was higher than that of total serum thiols. This discrepancy was not a consequence of the influence of other proteins, small molecules, and ions present in the serum on the total thiol group content determination by DTNB [28]. Taking into account that binding of various saturated FA causes HSA to present different binding constants when interacting with CB [29], the aim of this paper was to investigate the possible influence of FA bound to HSA on the HSA isolation by affinity chromatography, as well as on the reactivity of the HSA-SH group. The HSA-SH reactivity changes were investigated in vitro in the presence of different FA and in vivo in samples of diabetic patients and control subjects.
Blood samples were collected from patients with type 2 diabetes who were hospitalized due to poor metabolic control (HbA1c > 8.0) and healthy volunteers of appropriate age and sex. Blood was allowed to clot at room temperature and the serum was separated by centrifugation (4000g, 10 min) and used immediately for HSA isolation. This study was approved by the institutional ethics committee on human research. Informed consent was sought from all participants. Biochemical analysis HSA concentration was measured by the bromocresol green (BCG) method (Albumin liquicolor kit, Human, Wiesbaden, Germany). Briefly, 10 ll of the sample was added to 1 ml of BCG reagent and incubated for 5 min at room temperature. The absorbance was measured at 578 nm against the reagent blank [30]. The obtained absorbance was corrected for the sample blank. Protein content was determined by the biuret reaction. Briefly, 20 ll of the sample was added to 1 ml of biuret reagent and incubated for 30 min at room temperature. The absorbance was measured at 546 nm against the reagent blank [31]. The obtained absorbance was corrected for the sample blank. Protein and HSA concentrations were determined using an HSA standard curve (concentration range from 1 to 100 g/L). Serum triglycerides were determined using enzymatic methods (Boehringer Mannheim, Mannhein, Germany). Isolation of HSA by affinity chromatography Blue Sepharose 6 Fast Flow was used to isolate HSA from sera of patients with diabetes and healthy controls. The gel was packed into a column (4.3 0.8 cm), washed with starting buffer (0.02 M sodium phosphate, pH 7.2), and equilibrated with 10 bed vol of starting buffer. One milliliter of serum was loaded onto the gel and the column was washed with 56 ml of starting buffer. Gelbound HSA (b-HSA) was eluted with 40 ml of starting buffer containing 1.5 M NaCl. The first 28 ml of HSA fractions (this volume contains 91% of the total amount of bound HSA to column) was collected and concentrated by ultrafiltration using an Ultracel-30K device (Millipore, USA). To eliminate NaCl, HSA solution was further diluted with 0.1 M sodium phosphate (pH 7.4) buffer and concentrated by ultrafiltration. The resulting HSA solution was used for further determination of thiol groups. For in vitro experiments, commercial HSA was first washed from caprylate by ultrafiltration using an Ultracel-30K device (Millipore, USA). The conditions for isolating of HSA by affinity chromatography were as noted above, with the exception that only the first 36 ml of 56 ml starting buffer used for rinsing the unbound fraction
52
Influence of HSA-bound FA on the HSA-SH reactivity / V.B. Jovanovic´ et al. / Anal. Biochem. 448 (2014) 50–57
(nb-HSA) from the column was collected and concentrated by ultrafiltration. Preparation of defatted HSA For experiments recording fluorescent and circular dichroism (CD) spectra of defatted HSA and FA/HSA mixtures, and investigating the influence of defatting of HSA before isolation on determination of the HSA-SH content, the HSA was previously defatted with the active charcoal treatment, by the method of Chen [32]. Preparation of reduced HSA Commercial HSA contains 0.4 mol -SH group/mol HSA, so that for experiments in which the HSA-SH content was higher than 0.4 mol -SH/mol HSA, commercial defatted HSA was reduced with dithiothreitol. Prior to the reduction the content of HSA-SH group was determined. An appropriate amount of HSA was mixed with dithiothreitol at molar ratio 1:1 (molar content of oxidized thiol group:dithiothreitol) for 1 h at 37 °C in 0.1 M sodium phosphate, pH 7.4. Subsequently, dithiothreitol was washed away from HSA with 0.1 M sodium phosphate, pH 7.4, using an Ultracel-30K device (Millipore, USA). After this treatment, the HSA-SH content was approximately 0.96 mol -SH group/mol HSA. HSA prepared in this way was used for the preparation of the samples with the content of HSA-SH group from 0.4 to 0.8 mol -SH/mol HSA. Preparation of FA/HSA mixtures The stocks solutions of all FA in methanol are made. An aliquot of FA solution was dried under nitrogen and mixed thoroughly with 2 lM defatted HSA in pH 7.4 sodium phosphate buffer at 2:1 and 4:1 molar ratio FA/HSA. The mixture was incubated at 37 °C for 30 min, and after that the aliquot was subjected to fluorescence measurement. Fluorescence spectroscopy Fluorescence spectra were performed on a Fluoromax-4 Jobin Yvon (Horiba Scientific, Japan) spectrofluorimeter. The HSA concentration used for fluorescence was 2 lM. The spectra were recorded in the wavelength range of 300 to 450 nm upon excitation at 295 nm. Quartz cell (1 cm path length) and slit widths (4 nm) were used for all measurements. Each spectrum was the average of two scans and respective blanks of sodium phosphate buffer were used for the correction of all fluorescence spectra. The experiment was repeated and found to be reproducible within experimental errors. Circular dichroism spectroscopy The CD spectra were recorded on a J-815 spectrometer (Jasco Corporation, Tokyo, Japan) in a solution of defatted HSA (1 mg/ ml) in 0.01 M sodium phosphate pH 7.4 (control) and mixtures of HSA and FA (myristic, stearic and oleic acid) at a ratio 1:4. For the far-UV region (185–250 nm) measurements were made in a 0.01 cm path-length cell at 25 °C. Data were collected at 0.1nm intervals (50 nm/min) with a sensitivity of ±200 mdeg. Spectra represent the average of two accumulations and were baselinecorrected by subtraction of blank buffer. For measurements in the near-UV region (260–320 nm) a 1 cm path-length cell at 25 °C was used. Data were collected at 0.1-nm intervals (50 nm/ min) with a sensitivity of ±200 mdeg.
Thiol quantification Total serum-free thiol groups (protein and nonprotein, i.e., total thiol content) and HSA-SH content were determined spectrophotometrically according to a modified Ellman’s method [33]. 5,50 Dithiobis-(2-nitrobenzoic acid) (DTNB) reagent (100 ll of 2 mM solution) was mixed with equal volumes of sample (serum, isolated HSA, or cysteine standard solutions) and 1 M Tris buffer (pH 8.0) and brought up to 1000 ll with water. All reagents and sample were kept at room temperature 30 min before determination. Absorbance was measured after 30 min at room temperature at 412 nm against the sample and reagent blanks. The standard curve was created with cystein standard solutions concentration range from 0.0001 to 0.1 mM (in probe), which corresponded to the real sample levels of thiol groups of 0.001 to 1 mM; y = 1.36X-0.00935, r = 0.9996). Molar extinction coefficient is 13600 M1 cm1. Determination the pseudo-first-order rate constant For monitoring of the kinetics of the reaction of DTNB with the HSA-SH group commercial HSA and HSA isolated from sera of diabetic patients and healthy individuals (in the concentration of 0.1 mM) were used. DTNB reagent (100 ll of 5 mM solution) was mixed with equal volumes of isolated HSA (0.1 mM) (ratio 50:1) and 1 M Tris buffer (pH 8.0) and brought up to 1000 ll with water. All reagents and sample were previously preincubated for 5 min at 37 °C. The absorbance at 412 nm against the reagent blanks was recorded at intervals of 10 s, for 30 min at 37 °C, and corrected for the sample blank. A plot of ln [HSA-SH group] vs time (t) gives a straight line with a slope of k, which represents the value of pseudo-first-order constant for the reaction of the HSA-SH group with DTNB. Preparation of FA bound to HSA for FA analysis HSA solution (500 ll) was mixed with chloroform:methanol (2:1) at a volume ratio 1:2 and combined with the antioxidant 2,6-di-tert-butyl-4-methylphenol and internal standard (FA 13:0). The mixture was passed through a column with anhydrous ammonium sulfate. The remaining organic phase with FA was evaporated to dryness. FA analysis FA methyl esters were prepared by transmethylation, adding 1 ml sulfuric acid (1 M) in methanol to the dry residue of FA and heating at 85 °C for 2 h. FA methyl esters were then extracted three times with 0.5 ml of hexane and separated by gas chromatography (GC) using a Shimadzu (Kyoto, Japan) GC 2014 equipped with a flame ionization detector and DB-23 fused silica gel capillary column. The flame ionization detector was set at 250 °C, the injection port at 220 °C, and the oven temperature programmed from 130 to 190 °C at a heating rate of 3 °C/min. Identification of individual FA methyl esters was done by comparing sample peak retention times with authentic standards (Sigma Chemical Company) and/or the (PUFA)-2 standard mixtures (Restec, Bellefonte, PA). Statistical analysis The normality of the distribution of the data was assessed using the Shapiro–Wilk test. The statistical analysis of continuous variables was performed by Student’s t test. Continuous data are expressed as mean ± standard deviation (SD). P values <0.05 were considered significant.
Influence of HSA-bound FA on the HSA-SH reactivity / V.B. Jovanovic´ et al. / Anal. Biochem. 448 (2014) 50–57
Results and discussion Oxidative and/or carbonyl stress, which is believed to play an important role in the pathogenesis of several diseases such as liver and renal failure [18–20,34], uremia [22], diabetes mellitus [21,27], Alzheimer disease, and atherosclerosis [24]), was associated with a decrease of the HSA-SH content. In recent years, the research has focused on the modification of serum proteins’ thiol groups, especially the HSA-SH group during oxidative and/or carbonyl stress. Although HSA possesses a single exposed thiol group at Cys34 compared to 59 lysine and 24 arginine residues [3], the abundance of HSA [3] implies that this group has a great importance in the defense against oxidative stress. Therefore, monitoring the changes in HSA-SH content allows an evaluation of the extent of oxidative [20,23,27,34,35] and/or carbonyl [34] stress. In order to investigate the changes in the HSA-SH group content in patients with diabetes, HSA was isolated from serum by affinity chromatography (based on the specific binding of HSA to CB Sepharose, [26]) and HSA-SH content was determined spectrophotometrically according to a modified Ellman’s method [33]. The reliability of the method (from the isolation to quantification) was verified using 10 serum samples from diabetic patients. The amounts of isolated HSA were from 8.5 to 14.7 mg (15 to 25%) of the HSA applied to the column (40.1 to 58.7 mg). The total serum thiol content (mmol/L), surface thiol-group content of the isolated HSA (mol -SH/mol isolated HSA), and HSA-SH contribution (%) to the total thiol content were determined (Table 1). It was found that the content of –SH groups originating from the isolated HSA was between 11 and 33% higher than the total thiol content in the serum. The influence of the presence of proteins, small molecules, and ions in serum on the total thiol content determination was investigated using the standard addition method [28]. The values obtained for the recovery (99.1 ± 1.29%) and relative standard deviation (RSD of 1.37 to 1.64%) showed that the determination is accurate and precise, and that the presence of proteins, small molecules, and ions in the serum cannot be the reason for the found discrepancy. A plausible explanation could be the influence of various molecules bound to HSA (for which HSA is a transporter) which can result in an altered conformation of HSA, and therefore on its affinity for CB (applied for HSA isolation). In real samples, HSA binds various FA (with molar ratios FA/HSA from 0.1 to 6:1 [5,14]), which could lead to the alteration in HSA molecular conformation [3,10]. Therefore, the influence of binding of various FA on the HSA conformation changes was investigated by recording fluorescence and CD spectra. The fluorescence spectra of defatted HSA (control) and mixtures of HSA and FA (myristic, palmitic, stearic, oleic, eicosapentaenoic (EPA), and docosahexaenoic (DHA) acid) at ratios 1:2 and 1:4 are
53
presented in Figs. 2A and B. It can be observed that binding of all FA to HSA caused a decrease in the intensity of fluorescence at the peak wavelength (Figs. 2A and B). When FA/HSA ratio was 2:1 (Fig. 2A) the redundancy was about 20% for myristic and oleic acid, and 10% in the case of stearic, palmitic, EPA, and DHA acids. The enhancing of the FA/HSA ratio from 2:1 to 4:1 resulted in further mild changes in the fluorescent spectra (Fig. 2B), especially in the case of oleic acid. In addition, fluorescence spectra show that all FA binding to HSA molecules leads to the shift of the peak to the left and this effect becomes more prominent when FA/HSA ratio rises. In order to investigate the influence of FA binding on the secondary structure of HSA, the far-UV CD spectra of HSA in the absence (defatted HSA) and presence of FA (mixtures of HSA and myristic, stearic or oleic acid at a ratio 1:4) were recorded. As shown in Fig. 2C, the far-UV CD spectrum of HSA exhibits two negative bands in the UV region at 208 and 222 nm, which are typical for a-helical proteins [36]. The CD spectrum of the mixture of HSA and stearic acid at a ratio 1:4, as well as myristic and oleic acid (not shown), was found to be practically identical (within the experimental error ca. 5%) to the spectrum of defatted HSA. In order to quantify the different types of content of the secondary structures of HSA, far-UV CD spectra have been analyzed by the algorithm CONTINLL using reference set protein SP37. The obtained values of the a-helix, b-sheet, and other (turn and unordered) structures content of HSA were in the defatted HSA (62.6, 15.1, and 22.4%, respectively) and HSA-bound stearic acid (60.2, 17.1, and 22.7%, respectively). So, the CD analysis shows that binding FA to HSA does not induce significant changes in the secondary structure of HSA. The CD spectrum, in the near-UV region was used to probe the asymmetry of the proteins aromatic amino acid environment (Fig. 2D). Near-UV CD of defatted HSA showed characteristic minima at 262 and 268 nm with two shoulders near 283 and 290 nm [37]. The near-UV CD spectrum of the HSA-bound stearic acid at ratio 1:4 revealed that the intensity of the peaks at 262 and 268 nm increased and that they slightly changed the position, and that there is almost complete loss of two shoulders at 283 and 290 nm compared to the CD spectrum of defatted HSA. These results indicate a change in the environment of aromatic amino acid. HSA contains at least seven sites for FA binding (Fig. 1) [11,38]. The first one is located in subdomain IB, the second lies at the interface between subdomains IA and IIA. Sites 3 and 4 are both identified within subdomain IIIA, site 5 is located in subdomain IIIB, site 6 at the interface between subdomains IIA and IIB, and site 7 is associated with subdomain IIA [10]. Not all FA binding sites on the HSA molecule possess the same affinity; site 5 has the highest affinity, followed by sites 2 and 4 [39]. Tryptophan residue (Trp214) plays an important role in the formation of the IIA bind-
Table 1 Results of determination of total serum thiols (mmol/L), thiol-group content on the surface of the HSA isolated by affinity chromatography (mol -SH/mol HSA), and HSA-SH contribution (%) to the total serum thiols content, obtained for 10 serum samples from patients with type 2 diabetes. Serum sample
1 2 3 4 5 6 7 8 9 10 a
Total serum -SH content (mmol/L)
0.401 0.391 0.424 0.311 0.398 0.428 0.366 0.342 0.402 0.302
Serum HSA (mmol/L)
0.832 0.714 0.783 0.603 0.808 0.828 0.884 0.721 0.772 0.666
mol -SH/mol HSA
0.641 0.661 0.603 0.653 0.607 0.645 0.469 0.528 0.613 0.569
Contribution of HSA-SH to total serum thiols (mmol/L)a
(%)
0.533 0.472 0.472 0.393 0.490 0.534 0.415 0.381 0.473 0.379
133 121 111 127 123 125 113 111 118 125
Thiol group content (mmol/L) originated from isolated HSA is calculated according to the formula: mol -SH /mol HSA x serum HSA (mmol/L).
Influence of HSA-bound FA on the HSA-SH reactivity / V.B. Jovanovic´ et al. / Anal. Biochem. 448 (2014) 50–57
54
(A)
(B)
(C)
(D)
Fig.2. Fluorescent spectra of defatted HSA (control) and mixtures of HSA and FA (myristic (MA), palmitic (PA), stearic (SA), oleic (OA), EPA, and DHA) at ratios 2:1 (A) and 4:1 (B). Excitation at 295 nm was used. CD spectra of defatted HSA (control) and mixture of HSA and stearic acid (HSA-bound SA) at ratio 4:1 in far-UV region (C) and near-UV region (D).
ing site with limiting solvent accessibility, as well as in additional hydrophobic packing interactions at the IIA–IIIA interface [2,3]. Taking into account the affinity of FA binding sites on the HSA molecule, the position of these binding sites regarding the Trp214 residue (from which fluorescence originated) and the obtained fluorescence and near-UV CD spectra, it can be concluded that binding of FA to HSA leads to changes in the conformation of HSA molecules. These changes are determined by both the type of FA and the number of bound molecules, which is in accordance with the findings of Curry et al. [11]. Does the affinity column selectivity toward HSA molecules depend on bound FA? As noted above, prior to the determination of HSA-SH content, HSA was isolated from the serum by affinity chromatography with CB dye. Based on the structural similarity of CB and billirubin molecules [29] (which binds to the site within the subdomain IIA of HSA molecule [40]), it was postulated that CB binds at the same site, and where the Trp214 residue is also located. Alterations in HSA molecule conformation due to FA binding to HSA, especially at the afore-noted subdomain, could affect the affinity of HSA toward the dye depending on the type (and amount) of bound FA. In other words, the profile of FA bound to HSA could determine which of the HSA molecules from serum would interact with CB.
In order to test this hypothesis, commercially available HSA (0,752 mmol/L and with HSA–SH content of 0.423 mol/mol HSA) was subjected to affinity chromatography. Fluorescence spectra of the fractions which were unbound (nb-HSA) and bound (bHSA) to CB were recorded (Fig. 3) and profiles of FA in these fractions were analyzed by GC (Table 2). An emission band of b-HSA showed a hypsochromic shift with a reduction of the fluorescence intensity of 30% compared to HSA before applying to the column. The maximal fluorescence intensity of nb-HSA and HSA is quite similar. Thus, Trp214 of b-HSA is more exposed to the surrounding solvent compared to this residue in nb-HSA, indicating that there are differences in the conformation of HSA molecules in these two fractions, which may affect their binding affinity to Cibacron Blue. The FA profile of commercial HSA applied to the column reveals that palmitic, stearic, oleic, and linoleic acids comprise more than 90% of bound FA (Table 2). The FA profiles of the b-HSA and nbHSA were significantly different compared to the applied one and between each other. The percentage of total saturated FA (palmitic (16:0) and stearic (18:0) acids) in the b-HSA (58.9) was higher than that in the nb-HSA (47.4%). The percentage of monounsaturated oleic acid was almost the same in both fractions (19.4 and 17.5%, resp.), whereas the total percentage of polyunsaturated FA was significantly lower in the b-HSA than in the nb-HSA (17.6 and 33.7%, resp.). Among the polyunsaturated FA, linoleic acid has the greatest
Influence of HSA-bound FA on the HSA-SH reactivity / V.B. Jovanovic´ et al. / Anal. Biochem. 448 (2014) 50–57
Fig.3. Fluorescent spectra of commercially available HSA applied to Cibacron Blue column, bound-HSA (b-HSA) and unbound-HSA (nb-HSA). HSA concentration in all samples was 2 lM. Excitation at 295 nm was used.
Table 2 Fatty acid (FA) profiles of HSA applied to Cibacron Blue column, bound (b-HSA) and unbound (nb-HSA) HSA fractions (% of total FA), determined by GC. FA
HSA (%)
b-HSA (%)
nb-HSA (%)
16:0 16:1, 18:0 18:1, 18:1, 18:2, 18:3, 20:3, 20:4, 20:5, 22:4,
46.4 0.8 15.1 18.3 1.7 12.8 1.1 1.2 0.8 0.9 0.9
37.1 1.4 21.8 19.4 1.8 12.8 1.3 0.9 1.6 0.4 0.6
29.6 1.1 17.8 17.5 1.8 27.4 1.5 1.6 1.8 1.4 –
n-7 n-9 n-7 n-6 n-3 n-6 n-6 n-7 n-6
The results are expressed as mean value of two experiments.
impact on the obtained differences in the FA/HSA content between the b-HSA and the nb-HSA. Based on these results, it can be concluded that a high content of unsaturated FA (especially polyunsaturated) bound to HSA leads to the changes of conformation of HSA molecules and thereby to a weaker binding to CB. Thus, during the isolation of HSA from serum, there is a selection of HSA molecules that the dye will bind to, according to their conformation, which in turn depends on the FA profile.
Influence of HSA-bound FA on reactivity and/or accessibility of the HSA-SH group HSA conformational changes that occur as a result of bound FA to HSA affect the intensity of fluorescence of the Trp214 residue that is located in subdomain II. This subdomain is in the proximity of the subdomain I, where the free Cys34 is located. So, conformational changes of the HSA molecule could affect the HSA-SH group to become more or less susceptible to different reactions. Serum FA content increases in diabetes, liver, and cardiovascular disease [3,15]. Some of these diseases are accompanied by increased oxidative and/or carbonyl stress, which may render HSA-SH susceptible to modification with oxidative and/or carbonyl species. Therefore, it is very important to investigate whether FA binding to HSA affects the HSA-SH group reactivity and/or accessibility. The kinetics of the reaction of DTNB with the HSA-SH group from commercial HSA applied to a CB column, b-HSA and nb-
55
Fig.4. The kinetics curves of the HSA-SH group reaction with DTNB obtained for commercial HSA applied to Cibacron Blue column, b-HSA and nb-HSA.
HSA fractions (from the previous experiment), were investigated (Fig. 4). The obtained pseudo-first-order rate constants, obtained by using the molar ratio between HSA and DTNB 1:50, were: 4.8 103, 21.6 103, and 11.2 103 s1, resp. (Table 3). The total amounts of fatty acids (determined using internal standard of FA 13:0) bound in b-HSA and nb-HSA were 1.08 and 0.74 mol FA/mol HSA, resp. As the amount and FA profile of b-HSA is different (Table 2) compared with nb-HSA, it can be inferred that the observed significant increase (21.6 103 s1) of the b-HSA-SH reactivity is the consequence of the nature and abundance of FA molecules bound to HSA and subsequent changes in the HSA conformation. On the other hand, the thiol group content increases in b-HSA fraction (18.2% compared to the applied HSA, 11.8% compared to the nb-HSA, Table 3) obtained after 30 min of incubation (time condition applied for thiol group content determination) were statistically significant (P < 0.05). This indicates an increase of the SH group accessibility for the reaction with DTNB. Thus, the selection (binding) of HSA molecules by CB leads to the increase of HSA-SH group content and therefore to its higher contribution to total serum thiol groups (above described discrepancy). These findings were further verified by affinity chromatography of defatted HSA (with active charcoal, as reported under Materials and methods) and HSA with bound stearic acid. Four samples of commercial HSA with HSA-SH content ranging from 0.4 to 0.8 mol -SH/mol HSA were subjected to affinity chromatography on CB column, and the content of thiol groups in b-HSA fraction was determined (Fig. 5). Recovery values of the b-HSA-SH group content obtained for HSA with bound stearic acid ranged from 110 to 140%, while those for the defatted HSA are in the range from 98.5 to 101.7%. This directly validates that the FA bound to HSA (which is isolated by affinity chromatography) affects the determination of thiol groups with DTNB, such that the HSA-SH content is higher than total serum thiols. Therefore, affinity chromatography with CB is not suitable for isolation of HSA in order to determine the HSA-SH group content.
Reactivity of thiol group of HSA isolated from serum of diabetic patients and healthy individuals To check whether there is a difference in the reactivity of the HSA thiol group in real samples, HSA was isolated by affinity chromatography from serum of 20 diabetic patients (type 2) and 17
Influence of HSA-bound FA on the HSA-SH reactivity / V.B. Jovanovic´ et al. / Anal. Biochem. 448 (2014) 50–57
56
Table 3 HSA-SH content and the pseudo-first-order rate constant (k) for the reaction of HSASH group with DTNB obtained for the commercial HSA applied to Cibacron Blue column, bound (b-HSA) and unbound (nb-HSA) HSA fractions.
HSA b-HSA nb-HSA
mol -SH/mol HSA
k 103 (s1)
0.423 ± 0.011 0.500 ± 0.012* 0.447 ± 0.016
4.8 ± 0.2 21.6 ± 0.4* 11.2 ± 0.4**
The experiment was done in triplicate. * P < 0.05 compared to the applied HSA or nb-HSA fraction. ** P < 0.05 compared to the applied HSA.
Since the content of serum triglyceride (2.4 ± 0.7 mmol/L) and FA bound to HSA (1.3 ± 0.6 mol FA/mol HSA) and the reactivity of HSA-SH group (20.9 103 ± 4.4 103 s1) in the group of diabetic patients were statistically significantly higher (P < 0.05) compared to the control group (Table 4), it can be concluded that the impact of FA on the determination HSA-SH group is greater. This is supported by obtaining a stronger positive correlation (r = 0.828, P < 0.05) between the pseudo-first-order rate constants and the content of free FA bound to HSA in the diabetic group compared to the control group (r = 0.465). Acknowledgments This research was supported by the Ministry of Education, Science, and Technological Development of the Republic of Serbia (Project No. 172049) and FP7 RegPot FCUB-ERA, GA No256715. Our special thanks go to Irena Radovanovic (McGill University, Department of Biochemistry and Complex Traits Group) for help in editing. The authors declare that they have no conflict of interest. References
Fig.5. Recovery values of HSA-SH group content in bound HSA fraction, obtained after affinity chromatography (on Cibacron Blue column) of defatted HSA and HSA with bound stearic acid (HSA-bound SA). The content of -SH group in the samples ranged from 0.4 to 0.8 mol -SH/mol HSA.
Table 4 Serum triglyceride, FA/HSA content, and the pseudo-first-order rate constant (k) (for the reaction of HSA-SH group with DTNB) in patients with type 2 diabetes (n = 20) and control group (n = 17). Diabetic patients
Control group
Triglyceride (mmol/L) 2.4 ± 0.7 1.3 ± 0.4 FA/HSA (mol FA/mol HSA) 1.3 ± 0.6 0.7 ± 0.2 k (s1) 20.9 103 ± 4.4 103 12.9 103 ± 2.6 103
control individuals and the pseudo-first-order rate constants for the reaction of HSA-SH group with DTNB (at the molar ratio between HSA and DTNB 1:50) were determined. The mean value of the pseudo-first-order rate constant obtained for the control group (12.9 103 ± 2.6 103 s1) was statistically significantly lower (P <0.05) compared to the diabetic group (20.9 103 ± 4.4 103 s1). The HSA-SH group of the diabetics is more reactive compared to the control, which could be the consequence of conformational changes of the HSA molecule due to the binding of FA. These results suggest that the increased serum content of FA in the hyperglycemic state may have a positive effect on the body’s defense against oxidative and/or carbonyl stress. Due to the abundance of HSA, the Cys34 thiol group represents an important factor in the defense against oxidative stress [27], and an increase in the reactivity of the HSA-SH group in hyperglycemia could potentiate its capacity for scavenging of reactive oxidative species. A similar effect on the reactivity of the HSA-SH group can be expected during FA supplementation, which is used in various pathological conditions [41–43].
[1] D.C. Carter, J.X. Ho, Structure of serum albumin, Adv. Protein Chem. 45 (1994) 153–203. [2] X.M. He, D.C. Carter, Atomic structure and chemistry of human serum albumin, Nature 358 (1992) 209–215. [3] T. Peters, All about Albumin: Biochemistry, Genetic and Medical Applications, Academic Press, NY, 1996. [4] A.J. Stewart, C.A. Blindauer, S. Berezenko, D. Sleep, P.J. Sadler, Interdomain zinc site on human albumin, Proc. Natl. Acad. Sci. USA 100 (2003) 3701–3706. [5] A.A. Spector, Structure and lipid binding properties of serum albumin, Methods Enzymol. 128 (1986) 320–339. [6] N.A. Kratochwil, W. Huber, F. Muller, M. Kansy, P.R. Gerber, Predicting plasma protein binding of drugs: a new approach, Biochem. Pharmacol. 64 (2002) 1355–1374. [7] G. Sudlow, D.J. Birkett, D.N. Wade, The characterization of two specific drug binding sites on human serum albumin, Mol. Pharmacol. 11 (1975) 824–832. [8] G. Sudlow, D.J. Birkett, D.N. Wade, Further characterization of specific drug binding sites on human serum albumin, Mol. Pharmacol. 12 (1976) 1052– 1061. [9] G. Colmenarejo, In silico prediction of drug-binding strengths to human serum albumin, Med. Res. Rev. 23 (2003) 275–301. [10] S. Curry, Plasma albumin as a fatty acid carrier, Adv. Mol. Cell. Biol. 33 (2003) 29–46. [11] S. Curry, P. Brick, N.P. Franks, Fatty acid binding to human serum albumin: new insights from crystallographic studies, Biochim. Biophys. Acta 1441 (1999) 131–140. [12] A.A. Bhattacharya, T. Grune, S. Curry, Crystallographic analysis reveals common modes of binding of medium and long-chain fatty acids to human serum albumin, J. Mol. Biol. 303 (2000) 721–732. [13] S. Curry, H. Mandelkow, P. Brick, N. Franks, Crystal structure of human serum albumin complexed with fatty acid reveals an asymmetric distribution of binding sites, Nat. Struct. Biol. 5 (1998) 827–835. [14] R. Brodersen, S. Andersen, H. Vorum, S.U. Nielsen, A.O. Pedersen, Multiple fatty acid binding to albumin in human blood plasma, Eur. J. Biochem. 189 (1990) 343–349. [15] D.P. Cistola, D.M. Small, Fatty acid distribution in systems modeling the normal and diabetic human circulation. A 13C nuclear magnetic resonance study, J. Clin. Invest. 87 (1991) 1431–1441. [16] Jmol: an open-source Java viewer for chemical structures in 3D. http:// www.jmol.org/. [17] G.J. Quinlan, G.S. Martin, T.W. Evans, Albumin: biochemical properties and therapeutic potential, Hepatology 41 (2005) 1211–1219. [18] K. Oettl, V. Stadlbauer, F. Petter, J. Greilberger, C. Putz-Bankuti, S. Hallstrom, C. Lackner, R.E. Stauber, Oxidative damage of albumin in advanced liver disease, Biochim. Biophys. Acta 1782 (2008) 469–473. [19] 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 correlation with renal dysfunction: examination with the redox state of albumin, Kidney Int. 66 (2004) 1988–1993. [20] J. Himmelfarb, E. McMonagle, Albumin is the major plasma protein target of oxidant stress in uremia, Kidney Int. 60 (2001) 358–363. [21] K. Oettl, G. Reibnegger, O. Schmut, The redox state of human serum albumin in eye diseases with and without complications, Acta Ophthalmol. 89 (2011) 174–179. [22] M.L. Wratten, L. Sereni, C. Tetta, Oxidation of albumin is enhanced in the presence of uremic toxins, Ren. Fail. 23 (2001) 563–571.
Influence of HSA-bound FA on the HSA-SH reactivity / V.B. Jovanovic´ et al. / Anal. Biochem. 448 (2014) 50–57 [23] P. Faure, R. Tamisier, J.P. Baguet, A. Favier, S. Halimi, P. Levy, J.L. Pepin, Impairment of serum albumin antioxidant properties in obstructive sleep apnoea syndrome, Eur. Respir. J. 31 (2008) 1046–1053. [24] P.J. Thornalley, Protein and nucleotide damage by glyoxal and methylglyoxal in physiological systems—role in ageing and disease, Drug Metab. Drug Interact. 23 (2008) 125–150. [25] J. Zeng, M.J. Davies, Protein and low molecular mass thiols as targets and inhibitors of glycation reactions, Chem. Res. Toxicol. 19 (2006) 1668–1676. [26] J. Travis, J. Bowen, D. Tewksbury, D. Johnson, R. Pannell, Isolation of albumin from whole human plasma and fractionation of albumin-depleted plasma, Biochem. J. 157 (1976) 301–306. [27] A. Guerin-Dubourg, A. Catan, E. Bourdon, P. Rondeau, Structural modifications of human albumin in diabetes, Diabetes Metab. 38 (2012) 171–178. [28] V.B. Jovanovic, A.Z. Penezic-Romanjuk, I.D. Pavicevic, J.M. Acimovic, L.M. Mandic, Improving the reliability of human serum albumin-thiol group determination, Anal. Biochem. 439 (2013) 17–22. [29] R.J. Leatherbarrow, P.D. Dean, Studies on the mechanism of binding of serum albumins to immobilized cibacron blue F3G A, Biochem. J. 189 (1980) 27–34. [30] B.T. Dumas, W.A. Watson, H.G. Biggs, Albumin standards and the measurement of serum albumin with bromcresol green, Clin. Chim. Acta 258 (1997) 21–30. [31] G.R. Kingsley, The determination of serum total protein, albumin and globulin by the biuret reaction, J. Biol. Chem. 131 (1939) 197–200. [32] R.F. Chen, Removal of fatty acids from serum albumin by charcoal treatment, J. Biol. Chem. 242 (1967) 173–181. [33] G. Bulaj, T. Kortemme, D.P. Goldenberg, Ionization-reactivity relationships for cysteine thiols in polypeptides, Biochemistry 37 (1998) 8965–8972. [34] Y. Matsuyama, H. Terawaki, T. Terada, S. Era, Albumin thiol oxidation and serum protein carbonyl formation are progressively enhanced with advancing stages of chronic kidney disease, Clin. Exp. Nephrol. 13 (2009) 308–315.
57
[35] A. Hayakawa, K. Kuwata, S. Era, M. Sogami, H. Shimonaka, M. Yamamoto, S. Dohi, H. Hirose, Alteration of redox state of human serum albumin in patients under anesthesia and invasive surgery, J. Chromatogr., B: Biomed. Sci. Appl. 698 (1997) 27–33. [36] P. Manavalan Jr., W.C. Johnson, Sensitivity of circular dichroism to protein tertiary structure class, Nature 305 (1983) 831–832. [37] I. Sjoholm, I. Ljungstedt, Studies on the tryptophan and drug-binding properties of human serum albumin fragments by affinity chromatography and circular dichroism measurements, J. Biol. Chem. 248 (1973) 8434–8441. [38] J. Reichenwallner, D. Hinderberger, Using bound fatty acids to disclose the functional structure of serum albumin, Biochim. Biophys. Acta 1830 (2013) 5382–5393. [39] J.R. Simard, P.A. Zunszain, J.A. Hamilton, S. Curry, Location of high and low affinity fatty acid binding sites on human serum albumin revealed by NMR drug-competition analysis, J. Mol. Biol. 361 (2006) 336–351. [40] C.E. Petersen, C.E. Ha, K. Harohalli, J.B. Feix, N.V. Bhagavan, A dynamic model for bilirubin binding to human serum albumin, J. Biol. Chem. 275 (2000) 20985–20995. [41] M. de Lorgeril, P. Salen, P. Defaye, M. Rabaeus, Recent findings on the health effects of omega-3 fatty acids and statins, and their interactions: do statins inhibit omega-3?, BMC Med 11 (2013) 5. [42] J.H. Garman, S. Mulroney, M. Manigrasso, E. Flynn, C. Maric, Omega-3 fatty acid rich diet prevents diabetic renal disease, Am. J. Physiol. Renal Physiol. 296 (2009) 306–316. [43] B.E. Karlstrom, A.E. Jarvi, L. Byberg, L.G. Berglund, B.O. Vessby, Fatty fish in the diet of patients with type 2 diabetes: comparison of the metabolic effects of foods rich in n-3 and n-6 fatty acids, Am. J. Clin. Nutr. 94 (2011) 26–33.
Contents lists available at ScienceDirect
Analytical Biochemistry journal homepage: www.elsevier.com/locate/yabio
The influence of fatty acids on determination of human serum albumin thiol group Vesna B. Jovanovic´ a, Ivan D. Pavic´evic´ a, Marija M. Takic´ b, Ana Z. Penezic´-Romanjuk a, Jelena M. Ac´imovic´ a, Ljuba M. Mandic´ a,⇑ a b
Department of Biochemistry, Faculty of Chemistry, University of Belgrade, Studentski trg 12-16, Belgrade 11158, Serbia Institute for Medical Research, Laboratory for Nutrition and Metabolism, University of Belgrade, Belgrade, Serbia
a r t i c l e
i n f o
Article history: Received 18 September 2013 Received in revised form 13 November 2013 Accepted 25 November 2013 Available online 4 December 2013 Keywords: HSA-thiol group determination Fatty acid Albumin isolation Affinity chromatography Diabetes
a b s t r a c t During investigation of the changes of the Cys34 thiol group of human serum albumin (HSA) (isolated by affinity chromatography with Cibacron Blue (CB)) in diabetes, we found that the HSA-SH content was higher (11–33%) than the total serum thiol content. The influence of fatty acids (FA) binding to HSA on this discrepancy was investigated in vitro (using fluorescence and CD spectroscopy and GC) and with HSA samples from diabetic (n=20) and control groups (n=17). HSA-bound FA determine the selection of HSA molecules by CB and enhance reactivity and/or accessibility of the SH group. A high content of polyunsaturated FA (35.6%) leads to weaker binding of HSA molecules to CB. Rate constants of DTNB reaction with the SH group of HSA applied to a CB column, bound-HSA and unbound-HSA fractions, were 4.810-3, 21.610-3, and 11.210-3 s-1, respectively. The HSA-SH group of diabetics is more reactive compared with control individuals (rate constants 20.910-3±4.410-3 vs 12.910-3±2.610-3 s-1, P<0.05). Recovery values of the SH group obtained after chromatography of HSA with bound stearic acid ranged from 110 to 140%, while those for defatted HSA were from 98.5 to 101.7%. Thus, HSA-bound FA leads to an increase of HSA-SH content and a contribution to total serum thiols, which make the determination of the thiol group unreliable. Ó 2013 Elsevier Inc. All rights reserved.
Human serum albumin (HSA)1 is the most abundant plasma protein (0.6 mM). It is a 66.5-kDa protein organized into three homologous domains (labeled I–III) and each domain comprises two subdomains (A and B) that share common structural elements [1,2]. A total of 17 disulfide bridges located exclusively within subdomains contribute toward HSA’s stability [3]. HSA transports many endogenous ligands such as long chain (C13–C21) fatty acids (FA), hemin, bilirubin, and thyroxin, all of which bind HSA with high affinity [3]. Although most ligands for HSA are hydrophobic anions, heavy metals are also known to bind to this protein [1,3–5]. Moreover, HSA has the ability to bind a wide variety of drug molecules and alter their pharmacokinetic parameters [6]. This binding occurs via hydrophobic cavities in subdomains IIA and IIIA, known as Sudlow I and Sudlow II, respectively
⇑ Corresponding author. Address: Department of Biochemistry, Faculty of Chemistry, University of Belgrade, P.O. Box 51, Studentski trg 16, 11158 Belgrade, Serbia. Fax: +381 11 2184 330. E-mail address: [email protected] (L.M. Mandic´). 1 Abbreviations used: BCG, bromocresol green; CB, Cibacron Blue; CD, circular dichroism; EPA, eicosapentaenoic acid; DHA, docosahexaenoic acid; DTNB, 5,50 dithiobis-(2-nitrobenzoic acid); FA, fatty acids; GC, gas chromatography; HSA, human serum albumin; HSA-SH, human serum albumin thiol content; PUFA, polyunsaturated fatty acids; RSD, relative standard deviation; TG, triglyceride; UV, ultraviolet. 0003-2697/$ - see front matter Ó 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.ab.2013.11.030
[7,8], with the sole tryptophan residue in HSA located in Sudlow I (Trp-214) [9]. HSA is the primary transporter for delivering FA to the tissues and possesses at least seven binding sites of varying affinities for this ligand (Fig. 1) [10–12]. Although none of the FA-binding sites are completely identical, each comprises a hydrophobic pocket that interacts with the hydrocarbon chain, while five binding sites cap FA at one end with basic or polar residues that interact closely with the carboxyl group of bound FA [13]. Under normal physiological conditions, between 0.1 and 2 mol of FA are bound to HSA, but the molar ratio of FA/HSA can rise above 6:1 in the peripheral vasculature during fasting or extreme exercise [5,14] or under pathological conditions such as diabetes, liver disease, and cardiovascular disease [3,15]. In order to completely understand the role of HSA in vivo, it is crucial to obtain detailed information about the variety of ligands that HSA binds, as well as how these ligands interact and influence each other during binding to HSA. Besides its role in transport, HSA is one of the most important extracellular antioxidants [3]. The one free cysteine-derived thiol (-SH) group (Cys34) (located in subdomain I), which can exist in both reduced and oxidized forms, provides a part of the antioxidant property of HSA. As HSA is the most abundant plasma protein
Influence of HSA-bound FA on the HSA-SH reactivity / V.B. Jovanovic´ et al. / Anal. Biochem. 448 (2014) 50–57
51
Materials and methods Chemicals and instrumentation All chemicals were purchased from Merck (Darmstadt, Germany) and Sigma–Aldrich, Chemie (Steinheim, Germany), unless otherwise noted. The 20% solution of HSA (96% purity, containing 0.42 mol -SH/mol HSA) was purchased from Baxter (Vienna, Austria), and Cibacron Blue F3G-A Sepharose CL-6B from Pharmacia Biotech (Uppsala, Sweden). All chemicals used were analytical grade. Spectrophotometric measurements were performed by using a Beckman DU-50 spectrophotometer (Fullerton, CA). Serum samples
Fig.1. The alignment of crystal structures of HSA with bound fatty acids: myristic (light gray), palmitic (black), and stearic (dark gray). Crystal structures of HSAmyristic (PDB ID, 1bj5) [11], HSA-palmitic (PDB ID, 1e7h) [12], and HSA-stearic (PDB ID, 1e7i) [12] are used to create the figure using Jmol [16]. Domains I, II, and III with corresponding subdomains (A and B) are labeled.
[3], it accounts for 80% of the total redox-reactive thiols in plasma [17]. The HSA-thiol group (HSA-SH) content decreases in some diseases such as liver and renal failure [18–20], diabetes mellitus [21], uremia [22], and obstructive sleep apnea [23], for which it is believed that oxidative stress plays an important role in pathogenesis. In addition, the increasing flux of reactive dicarbonyl compounds (glyoxal, methylglyoxal, and 3-deoxyglucosone), which occurs during carbonyl stress (in diabetes, Alzheimer’s disease, renal failure, liver cirrhosis, anemia, uremia, and atherosclerosis [24]), leads to the Cys34 side chain carbonylation and therefore to the decrease of the HSA-SH content [24]. It could be inferred that a differential content of HSA-SH can be a useful marker of oxidative and carbonyl stress during various pathological states, which lead to several advances in its quantification [25]. Binding of FA is associated with significant structural changes in the HSA molecule [10,22] which determine changes in its physical– chemical properties, such as an increase in the stability toward heat and proteases [3]. Moreover, structural changes could cause Cys34 residues to be more or less exposed to the surrounding environment, leading to differential reactivity and susceptibility to oxidative and carbonyl stress. This is of particular importance during pathological states (diabetes, liver and heart diseases) that are followed by an increase in serum-free FA content [3,15] and in the levels of carbonyl and/or oxidative reactive species [18–23]. A single-step affinity chromatographic method based on the specific interaction between the HSA and the textile dye Cibacron Blue F3G-A (CB) [26] has been largely applied for the isolation of HSA, particularly in recent years for studying the antioxidant properties of HSA [23,27]. When we employed this method while investigating the influence of oxidative and/or carbonyl stress in type 2 diabetes on the HSA Cys34 thiol group, we uncovered that the HSASH content was higher than that of total serum thiols. This discrepancy was not a consequence of the influence of other proteins, small molecules, and ions present in the serum on the total thiol group content determination by DTNB [28]. Taking into account that binding of various saturated FA causes HSA to present different binding constants when interacting with CB [29], the aim of this paper was to investigate the possible influence of FA bound to HSA on the HSA isolation by affinity chromatography, as well as on the reactivity of the HSA-SH group. The HSA-SH reactivity changes were investigated in vitro in the presence of different FA and in vivo in samples of diabetic patients and control subjects.
Blood samples were collected from patients with type 2 diabetes who were hospitalized due to poor metabolic control (HbA1c > 8.0) and healthy volunteers of appropriate age and sex. Blood was allowed to clot at room temperature and the serum was separated by centrifugation (4000g, 10 min) and used immediately for HSA isolation. This study was approved by the institutional ethics committee on human research. Informed consent was sought from all participants. Biochemical analysis HSA concentration was measured by the bromocresol green (BCG) method (Albumin liquicolor kit, Human, Wiesbaden, Germany). Briefly, 10 ll of the sample was added to 1 ml of BCG reagent and incubated for 5 min at room temperature. The absorbance was measured at 578 nm against the reagent blank [30]. The obtained absorbance was corrected for the sample blank. Protein content was determined by the biuret reaction. Briefly, 20 ll of the sample was added to 1 ml of biuret reagent and incubated for 30 min at room temperature. The absorbance was measured at 546 nm against the reagent blank [31]. The obtained absorbance was corrected for the sample blank. Protein and HSA concentrations were determined using an HSA standard curve (concentration range from 1 to 100 g/L). Serum triglycerides were determined using enzymatic methods (Boehringer Mannheim, Mannhein, Germany). Isolation of HSA by affinity chromatography Blue Sepharose 6 Fast Flow was used to isolate HSA from sera of patients with diabetes and healthy controls. The gel was packed into a column (4.3 0.8 cm), washed with starting buffer (0.02 M sodium phosphate, pH 7.2), and equilibrated with 10 bed vol of starting buffer. One milliliter of serum was loaded onto the gel and the column was washed with 56 ml of starting buffer. Gelbound HSA (b-HSA) was eluted with 40 ml of starting buffer containing 1.5 M NaCl. The first 28 ml of HSA fractions (this volume contains 91% of the total amount of bound HSA to column) was collected and concentrated by ultrafiltration using an Ultracel-30K device (Millipore, USA). To eliminate NaCl, HSA solution was further diluted with 0.1 M sodium phosphate (pH 7.4) buffer and concentrated by ultrafiltration. The resulting HSA solution was used for further determination of thiol groups. For in vitro experiments, commercial HSA was first washed from caprylate by ultrafiltration using an Ultracel-30K device (Millipore, USA). The conditions for isolating of HSA by affinity chromatography were as noted above, with the exception that only the first 36 ml of 56 ml starting buffer used for rinsing the unbound fraction
52
Influence of HSA-bound FA on the HSA-SH reactivity / V.B. Jovanovic´ et al. / Anal. Biochem. 448 (2014) 50–57
(nb-HSA) from the column was collected and concentrated by ultrafiltration. Preparation of defatted HSA For experiments recording fluorescent and circular dichroism (CD) spectra of defatted HSA and FA/HSA mixtures, and investigating the influence of defatting of HSA before isolation on determination of the HSA-SH content, the HSA was previously defatted with the active charcoal treatment, by the method of Chen [32]. Preparation of reduced HSA Commercial HSA contains 0.4 mol -SH group/mol HSA, so that for experiments in which the HSA-SH content was higher than 0.4 mol -SH/mol HSA, commercial defatted HSA was reduced with dithiothreitol. Prior to the reduction the content of HSA-SH group was determined. An appropriate amount of HSA was mixed with dithiothreitol at molar ratio 1:1 (molar content of oxidized thiol group:dithiothreitol) for 1 h at 37 °C in 0.1 M sodium phosphate, pH 7.4. Subsequently, dithiothreitol was washed away from HSA with 0.1 M sodium phosphate, pH 7.4, using an Ultracel-30K device (Millipore, USA). After this treatment, the HSA-SH content was approximately 0.96 mol -SH group/mol HSA. HSA prepared in this way was used for the preparation of the samples with the content of HSA-SH group from 0.4 to 0.8 mol -SH/mol HSA. Preparation of FA/HSA mixtures The stocks solutions of all FA in methanol are made. An aliquot of FA solution was dried under nitrogen and mixed thoroughly with 2 lM defatted HSA in pH 7.4 sodium phosphate buffer at 2:1 and 4:1 molar ratio FA/HSA. The mixture was incubated at 37 °C for 30 min, and after that the aliquot was subjected to fluorescence measurement. Fluorescence spectroscopy Fluorescence spectra were performed on a Fluoromax-4 Jobin Yvon (Horiba Scientific, Japan) spectrofluorimeter. The HSA concentration used for fluorescence was 2 lM. The spectra were recorded in the wavelength range of 300 to 450 nm upon excitation at 295 nm. Quartz cell (1 cm path length) and slit widths (4 nm) were used for all measurements. Each spectrum was the average of two scans and respective blanks of sodium phosphate buffer were used for the correction of all fluorescence spectra. The experiment was repeated and found to be reproducible within experimental errors. Circular dichroism spectroscopy The CD spectra were recorded on a J-815 spectrometer (Jasco Corporation, Tokyo, Japan) in a solution of defatted HSA (1 mg/ ml) in 0.01 M sodium phosphate pH 7.4 (control) and mixtures of HSA and FA (myristic, stearic and oleic acid) at a ratio 1:4. For the far-UV region (185–250 nm) measurements were made in a 0.01 cm path-length cell at 25 °C. Data were collected at 0.1nm intervals (50 nm/min) with a sensitivity of ±200 mdeg. Spectra represent the average of two accumulations and were baselinecorrected by subtraction of blank buffer. For measurements in the near-UV region (260–320 nm) a 1 cm path-length cell at 25 °C was used. Data were collected at 0.1-nm intervals (50 nm/ min) with a sensitivity of ±200 mdeg.
Thiol quantification Total serum-free thiol groups (protein and nonprotein, i.e., total thiol content) and HSA-SH content were determined spectrophotometrically according to a modified Ellman’s method [33]. 5,50 Dithiobis-(2-nitrobenzoic acid) (DTNB) reagent (100 ll of 2 mM solution) was mixed with equal volumes of sample (serum, isolated HSA, or cysteine standard solutions) and 1 M Tris buffer (pH 8.0) and brought up to 1000 ll with water. All reagents and sample were kept at room temperature 30 min before determination. Absorbance was measured after 30 min at room temperature at 412 nm against the sample and reagent blanks. The standard curve was created with cystein standard solutions concentration range from 0.0001 to 0.1 mM (in probe), which corresponded to the real sample levels of thiol groups of 0.001 to 1 mM; y = 1.36X-0.00935, r = 0.9996). Molar extinction coefficient is 13600 M1 cm1. Determination the pseudo-first-order rate constant For monitoring of the kinetics of the reaction of DTNB with the HSA-SH group commercial HSA and HSA isolated from sera of diabetic patients and healthy individuals (in the concentration of 0.1 mM) were used. DTNB reagent (100 ll of 5 mM solution) was mixed with equal volumes of isolated HSA (0.1 mM) (ratio 50:1) and 1 M Tris buffer (pH 8.0) and brought up to 1000 ll with water. All reagents and sample were previously preincubated for 5 min at 37 °C. The absorbance at 412 nm against the reagent blanks was recorded at intervals of 10 s, for 30 min at 37 °C, and corrected for the sample blank. A plot of ln [HSA-SH group] vs time (t) gives a straight line with a slope of k, which represents the value of pseudo-first-order constant for the reaction of the HSA-SH group with DTNB. Preparation of FA bound to HSA for FA analysis HSA solution (500 ll) was mixed with chloroform:methanol (2:1) at a volume ratio 1:2 and combined with the antioxidant 2,6-di-tert-butyl-4-methylphenol and internal standard (FA 13:0). The mixture was passed through a column with anhydrous ammonium sulfate. The remaining organic phase with FA was evaporated to dryness. FA analysis FA methyl esters were prepared by transmethylation, adding 1 ml sulfuric acid (1 M) in methanol to the dry residue of FA and heating at 85 °C for 2 h. FA methyl esters were then extracted three times with 0.5 ml of hexane and separated by gas chromatography (GC) using a Shimadzu (Kyoto, Japan) GC 2014 equipped with a flame ionization detector and DB-23 fused silica gel capillary column. The flame ionization detector was set at 250 °C, the injection port at 220 °C, and the oven temperature programmed from 130 to 190 °C at a heating rate of 3 °C/min. Identification of individual FA methyl esters was done by comparing sample peak retention times with authentic standards (Sigma Chemical Company) and/or the (PUFA)-2 standard mixtures (Restec, Bellefonte, PA). Statistical analysis The normality of the distribution of the data was assessed using the Shapiro–Wilk test. The statistical analysis of continuous variables was performed by Student’s t test. Continuous data are expressed as mean ± standard deviation (SD). P values <0.05 were considered significant.
Influence of HSA-bound FA on the HSA-SH reactivity / V.B. Jovanovic´ et al. / Anal. Biochem. 448 (2014) 50–57
Results and discussion Oxidative and/or carbonyl stress, which is believed to play an important role in the pathogenesis of several diseases such as liver and renal failure [18–20,34], uremia [22], diabetes mellitus [21,27], Alzheimer disease, and atherosclerosis [24]), was associated with a decrease of the HSA-SH content. In recent years, the research has focused on the modification of serum proteins’ thiol groups, especially the HSA-SH group during oxidative and/or carbonyl stress. Although HSA possesses a single exposed thiol group at Cys34 compared to 59 lysine and 24 arginine residues [3], the abundance of HSA [3] implies that this group has a great importance in the defense against oxidative stress. Therefore, monitoring the changes in HSA-SH content allows an evaluation of the extent of oxidative [20,23,27,34,35] and/or carbonyl [34] stress. In order to investigate the changes in the HSA-SH group content in patients with diabetes, HSA was isolated from serum by affinity chromatography (based on the specific binding of HSA to CB Sepharose, [26]) and HSA-SH content was determined spectrophotometrically according to a modified Ellman’s method [33]. The reliability of the method (from the isolation to quantification) was verified using 10 serum samples from diabetic patients. The amounts of isolated HSA were from 8.5 to 14.7 mg (15 to 25%) of the HSA applied to the column (40.1 to 58.7 mg). The total serum thiol content (mmol/L), surface thiol-group content of the isolated HSA (mol -SH/mol isolated HSA), and HSA-SH contribution (%) to the total thiol content were determined (Table 1). It was found that the content of –SH groups originating from the isolated HSA was between 11 and 33% higher than the total thiol content in the serum. The influence of the presence of proteins, small molecules, and ions in serum on the total thiol content determination was investigated using the standard addition method [28]. The values obtained for the recovery (99.1 ± 1.29%) and relative standard deviation (RSD of 1.37 to 1.64%) showed that the determination is accurate and precise, and that the presence of proteins, small molecules, and ions in the serum cannot be the reason for the found discrepancy. A plausible explanation could be the influence of various molecules bound to HSA (for which HSA is a transporter) which can result in an altered conformation of HSA, and therefore on its affinity for CB (applied for HSA isolation). In real samples, HSA binds various FA (with molar ratios FA/HSA from 0.1 to 6:1 [5,14]), which could lead to the alteration in HSA molecular conformation [3,10]. Therefore, the influence of binding of various FA on the HSA conformation changes was investigated by recording fluorescence and CD spectra. The fluorescence spectra of defatted HSA (control) and mixtures of HSA and FA (myristic, palmitic, stearic, oleic, eicosapentaenoic (EPA), and docosahexaenoic (DHA) acid) at ratios 1:2 and 1:4 are
53
presented in Figs. 2A and B. It can be observed that binding of all FA to HSA caused a decrease in the intensity of fluorescence at the peak wavelength (Figs. 2A and B). When FA/HSA ratio was 2:1 (Fig. 2A) the redundancy was about 20% for myristic and oleic acid, and 10% in the case of stearic, palmitic, EPA, and DHA acids. The enhancing of the FA/HSA ratio from 2:1 to 4:1 resulted in further mild changes in the fluorescent spectra (Fig. 2B), especially in the case of oleic acid. In addition, fluorescence spectra show that all FA binding to HSA molecules leads to the shift of the peak to the left and this effect becomes more prominent when FA/HSA ratio rises. In order to investigate the influence of FA binding on the secondary structure of HSA, the far-UV CD spectra of HSA in the absence (defatted HSA) and presence of FA (mixtures of HSA and myristic, stearic or oleic acid at a ratio 1:4) were recorded. As shown in Fig. 2C, the far-UV CD spectrum of HSA exhibits two negative bands in the UV region at 208 and 222 nm, which are typical for a-helical proteins [36]. The CD spectrum of the mixture of HSA and stearic acid at a ratio 1:4, as well as myristic and oleic acid (not shown), was found to be practically identical (within the experimental error ca. 5%) to the spectrum of defatted HSA. In order to quantify the different types of content of the secondary structures of HSA, far-UV CD spectra have been analyzed by the algorithm CONTINLL using reference set protein SP37. The obtained values of the a-helix, b-sheet, and other (turn and unordered) structures content of HSA were in the defatted HSA (62.6, 15.1, and 22.4%, respectively) and HSA-bound stearic acid (60.2, 17.1, and 22.7%, respectively). So, the CD analysis shows that binding FA to HSA does not induce significant changes in the secondary structure of HSA. The CD spectrum, in the near-UV region was used to probe the asymmetry of the proteins aromatic amino acid environment (Fig. 2D). Near-UV CD of defatted HSA showed characteristic minima at 262 and 268 nm with two shoulders near 283 and 290 nm [37]. The near-UV CD spectrum of the HSA-bound stearic acid at ratio 1:4 revealed that the intensity of the peaks at 262 and 268 nm increased and that they slightly changed the position, and that there is almost complete loss of two shoulders at 283 and 290 nm compared to the CD spectrum of defatted HSA. These results indicate a change in the environment of aromatic amino acid. HSA contains at least seven sites for FA binding (Fig. 1) [11,38]. The first one is located in subdomain IB, the second lies at the interface between subdomains IA and IIA. Sites 3 and 4 are both identified within subdomain IIIA, site 5 is located in subdomain IIIB, site 6 at the interface between subdomains IIA and IIB, and site 7 is associated with subdomain IIA [10]. Not all FA binding sites on the HSA molecule possess the same affinity; site 5 has the highest affinity, followed by sites 2 and 4 [39]. Tryptophan residue (Trp214) plays an important role in the formation of the IIA bind-
Table 1 Results of determination of total serum thiols (mmol/L), thiol-group content on the surface of the HSA isolated by affinity chromatography (mol -SH/mol HSA), and HSA-SH contribution (%) to the total serum thiols content, obtained for 10 serum samples from patients with type 2 diabetes. Serum sample
1 2 3 4 5 6 7 8 9 10 a
Total serum -SH content (mmol/L)
0.401 0.391 0.424 0.311 0.398 0.428 0.366 0.342 0.402 0.302
Serum HSA (mmol/L)
0.832 0.714 0.783 0.603 0.808 0.828 0.884 0.721 0.772 0.666
mol -SH/mol HSA
0.641 0.661 0.603 0.653 0.607 0.645 0.469 0.528 0.613 0.569
Contribution of HSA-SH to total serum thiols (mmol/L)a
(%)
0.533 0.472 0.472 0.393 0.490 0.534 0.415 0.381 0.473 0.379
133 121 111 127 123 125 113 111 118 125
Thiol group content (mmol/L) originated from isolated HSA is calculated according to the formula: mol -SH /mol HSA x serum HSA (mmol/L).
Influence of HSA-bound FA on the HSA-SH reactivity / V.B. Jovanovic´ et al. / Anal. Biochem. 448 (2014) 50–57
54
(A)
(B)
(C)
(D)
Fig.2. Fluorescent spectra of defatted HSA (control) and mixtures of HSA and FA (myristic (MA), palmitic (PA), stearic (SA), oleic (OA), EPA, and DHA) at ratios 2:1 (A) and 4:1 (B). Excitation at 295 nm was used. CD spectra of defatted HSA (control) and mixture of HSA and stearic acid (HSA-bound SA) at ratio 4:1 in far-UV region (C) and near-UV region (D).
ing site with limiting solvent accessibility, as well as in additional hydrophobic packing interactions at the IIA–IIIA interface [2,3]. Taking into account the affinity of FA binding sites on the HSA molecule, the position of these binding sites regarding the Trp214 residue (from which fluorescence originated) and the obtained fluorescence and near-UV CD spectra, it can be concluded that binding of FA to HSA leads to changes in the conformation of HSA molecules. These changes are determined by both the type of FA and the number of bound molecules, which is in accordance with the findings of Curry et al. [11]. Does the affinity column selectivity toward HSA molecules depend on bound FA? As noted above, prior to the determination of HSA-SH content, HSA was isolated from the serum by affinity chromatography with CB dye. Based on the structural similarity of CB and billirubin molecules [29] (which binds to the site within the subdomain IIA of HSA molecule [40]), it was postulated that CB binds at the same site, and where the Trp214 residue is also located. Alterations in HSA molecule conformation due to FA binding to HSA, especially at the afore-noted subdomain, could affect the affinity of HSA toward the dye depending on the type (and amount) of bound FA. In other words, the profile of FA bound to HSA could determine which of the HSA molecules from serum would interact with CB.
In order to test this hypothesis, commercially available HSA (0,752 mmol/L and with HSA–SH content of 0.423 mol/mol HSA) was subjected to affinity chromatography. Fluorescence spectra of the fractions which were unbound (nb-HSA) and bound (bHSA) to CB were recorded (Fig. 3) and profiles of FA in these fractions were analyzed by GC (Table 2). An emission band of b-HSA showed a hypsochromic shift with a reduction of the fluorescence intensity of 30% compared to HSA before applying to the column. The maximal fluorescence intensity of nb-HSA and HSA is quite similar. Thus, Trp214 of b-HSA is more exposed to the surrounding solvent compared to this residue in nb-HSA, indicating that there are differences in the conformation of HSA molecules in these two fractions, which may affect their binding affinity to Cibacron Blue. The FA profile of commercial HSA applied to the column reveals that palmitic, stearic, oleic, and linoleic acids comprise more than 90% of bound FA (Table 2). The FA profiles of the b-HSA and nbHSA were significantly different compared to the applied one and between each other. The percentage of total saturated FA (palmitic (16:0) and stearic (18:0) acids) in the b-HSA (58.9) was higher than that in the nb-HSA (47.4%). The percentage of monounsaturated oleic acid was almost the same in both fractions (19.4 and 17.5%, resp.), whereas the total percentage of polyunsaturated FA was significantly lower in the b-HSA than in the nb-HSA (17.6 and 33.7%, resp.). Among the polyunsaturated FA, linoleic acid has the greatest
Influence of HSA-bound FA on the HSA-SH reactivity / V.B. Jovanovic´ et al. / Anal. Biochem. 448 (2014) 50–57
Fig.3. Fluorescent spectra of commercially available HSA applied to Cibacron Blue column, bound-HSA (b-HSA) and unbound-HSA (nb-HSA). HSA concentration in all samples was 2 lM. Excitation at 295 nm was used.
Table 2 Fatty acid (FA) profiles of HSA applied to Cibacron Blue column, bound (b-HSA) and unbound (nb-HSA) HSA fractions (% of total FA), determined by GC. FA
HSA (%)
b-HSA (%)
nb-HSA (%)
16:0 16:1, 18:0 18:1, 18:1, 18:2, 18:3, 20:3, 20:4, 20:5, 22:4,
46.4 0.8 15.1 18.3 1.7 12.8 1.1 1.2 0.8 0.9 0.9
37.1 1.4 21.8 19.4 1.8 12.8 1.3 0.9 1.6 0.4 0.6
29.6 1.1 17.8 17.5 1.8 27.4 1.5 1.6 1.8 1.4 –
n-7 n-9 n-7 n-6 n-3 n-6 n-6 n-7 n-6
The results are expressed as mean value of two experiments.
impact on the obtained differences in the FA/HSA content between the b-HSA and the nb-HSA. Based on these results, it can be concluded that a high content of unsaturated FA (especially polyunsaturated) bound to HSA leads to the changes of conformation of HSA molecules and thereby to a weaker binding to CB. Thus, during the isolation of HSA from serum, there is a selection of HSA molecules that the dye will bind to, according to their conformation, which in turn depends on the FA profile.
Influence of HSA-bound FA on reactivity and/or accessibility of the HSA-SH group HSA conformational changes that occur as a result of bound FA to HSA affect the intensity of fluorescence of the Trp214 residue that is located in subdomain II. This subdomain is in the proximity of the subdomain I, where the free Cys34 is located. So, conformational changes of the HSA molecule could affect the HSA-SH group to become more or less susceptible to different reactions. Serum FA content increases in diabetes, liver, and cardiovascular disease [3,15]. Some of these diseases are accompanied by increased oxidative and/or carbonyl stress, which may render HSA-SH susceptible to modification with oxidative and/or carbonyl species. Therefore, it is very important to investigate whether FA binding to HSA affects the HSA-SH group reactivity and/or accessibility. The kinetics of the reaction of DTNB with the HSA-SH group from commercial HSA applied to a CB column, b-HSA and nb-
55
Fig.4. The kinetics curves of the HSA-SH group reaction with DTNB obtained for commercial HSA applied to Cibacron Blue column, b-HSA and nb-HSA.
HSA fractions (from the previous experiment), were investigated (Fig. 4). The obtained pseudo-first-order rate constants, obtained by using the molar ratio between HSA and DTNB 1:50, were: 4.8 103, 21.6 103, and 11.2 103 s1, resp. (Table 3). The total amounts of fatty acids (determined using internal standard of FA 13:0) bound in b-HSA and nb-HSA were 1.08 and 0.74 mol FA/mol HSA, resp. As the amount and FA profile of b-HSA is different (Table 2) compared with nb-HSA, it can be inferred that the observed significant increase (21.6 103 s1) of the b-HSA-SH reactivity is the consequence of the nature and abundance of FA molecules bound to HSA and subsequent changes in the HSA conformation. On the other hand, the thiol group content increases in b-HSA fraction (18.2% compared to the applied HSA, 11.8% compared to the nb-HSA, Table 3) obtained after 30 min of incubation (time condition applied for thiol group content determination) were statistically significant (P < 0.05). This indicates an increase of the SH group accessibility for the reaction with DTNB. Thus, the selection (binding) of HSA molecules by CB leads to the increase of HSA-SH group content and therefore to its higher contribution to total serum thiol groups (above described discrepancy). These findings were further verified by affinity chromatography of defatted HSA (with active charcoal, as reported under Materials and methods) and HSA with bound stearic acid. Four samples of commercial HSA with HSA-SH content ranging from 0.4 to 0.8 mol -SH/mol HSA were subjected to affinity chromatography on CB column, and the content of thiol groups in b-HSA fraction was determined (Fig. 5). Recovery values of the b-HSA-SH group content obtained for HSA with bound stearic acid ranged from 110 to 140%, while those for the defatted HSA are in the range from 98.5 to 101.7%. This directly validates that the FA bound to HSA (which is isolated by affinity chromatography) affects the determination of thiol groups with DTNB, such that the HSA-SH content is higher than total serum thiols. Therefore, affinity chromatography with CB is not suitable for isolation of HSA in order to determine the HSA-SH group content.
Reactivity of thiol group of HSA isolated from serum of diabetic patients and healthy individuals To check whether there is a difference in the reactivity of the HSA thiol group in real samples, HSA was isolated by affinity chromatography from serum of 20 diabetic patients (type 2) and 17
Influence of HSA-bound FA on the HSA-SH reactivity / V.B. Jovanovic´ et al. / Anal. Biochem. 448 (2014) 50–57
56
Table 3 HSA-SH content and the pseudo-first-order rate constant (k) for the reaction of HSASH group with DTNB obtained for the commercial HSA applied to Cibacron Blue column, bound (b-HSA) and unbound (nb-HSA) HSA fractions.
HSA b-HSA nb-HSA
mol -SH/mol HSA
k 103 (s1)
0.423 ± 0.011 0.500 ± 0.012* 0.447 ± 0.016
4.8 ± 0.2 21.6 ± 0.4* 11.2 ± 0.4**
The experiment was done in triplicate. * P < 0.05 compared to the applied HSA or nb-HSA fraction. ** P < 0.05 compared to the applied HSA.
Since the content of serum triglyceride (2.4 ± 0.7 mmol/L) and FA bound to HSA (1.3 ± 0.6 mol FA/mol HSA) and the reactivity of HSA-SH group (20.9 103 ± 4.4 103 s1) in the group of diabetic patients were statistically significantly higher (P < 0.05) compared to the control group (Table 4), it can be concluded that the impact of FA on the determination HSA-SH group is greater. This is supported by obtaining a stronger positive correlation (r = 0.828, P < 0.05) between the pseudo-first-order rate constants and the content of free FA bound to HSA in the diabetic group compared to the control group (r = 0.465). Acknowledgments This research was supported by the Ministry of Education, Science, and Technological Development of the Republic of Serbia (Project No. 172049) and FP7 RegPot FCUB-ERA, GA No256715. Our special thanks go to Irena Radovanovic (McGill University, Department of Biochemistry and Complex Traits Group) for help in editing. The authors declare that they have no conflict of interest. References
Fig.5. Recovery values of HSA-SH group content in bound HSA fraction, obtained after affinity chromatography (on Cibacron Blue column) of defatted HSA and HSA with bound stearic acid (HSA-bound SA). The content of -SH group in the samples ranged from 0.4 to 0.8 mol -SH/mol HSA.
Table 4 Serum triglyceride, FA/HSA content, and the pseudo-first-order rate constant (k) (for the reaction of HSA-SH group with DTNB) in patients with type 2 diabetes (n = 20) and control group (n = 17). Diabetic patients
Control group
Triglyceride (mmol/L) 2.4 ± 0.7 1.3 ± 0.4 FA/HSA (mol FA/mol HSA) 1.3 ± 0.6 0.7 ± 0.2 k (s1) 20.9 103 ± 4.4 103 12.9 103 ± 2.6 103
control individuals and the pseudo-first-order rate constants for the reaction of HSA-SH group with DTNB (at the molar ratio between HSA and DTNB 1:50) were determined. The mean value of the pseudo-first-order rate constant obtained for the control group (12.9 103 ± 2.6 103 s1) was statistically significantly lower (P <0.05) compared to the diabetic group (20.9 103 ± 4.4 103 s1). The HSA-SH group of the diabetics is more reactive compared to the control, which could be the consequence of conformational changes of the HSA molecule due to the binding of FA. These results suggest that the increased serum content of FA in the hyperglycemic state may have a positive effect on the body’s defense against oxidative and/or carbonyl stress. Due to the abundance of HSA, the Cys34 thiol group represents an important factor in the defense against oxidative stress [27], and an increase in the reactivity of the HSA-SH group in hyperglycemia could potentiate its capacity for scavenging of reactive oxidative species. A similar effect on the reactivity of the HSA-SH group can be expected during FA supplementation, which is used in various pathological conditions [41–43].
[1] D.C. Carter, J.X. Ho, Structure of serum albumin, Adv. Protein Chem. 45 (1994) 153–203. [2] X.M. He, D.C. Carter, Atomic structure and chemistry of human serum albumin, Nature 358 (1992) 209–215. [3] T. Peters, All about Albumin: Biochemistry, Genetic and Medical Applications, Academic Press, NY, 1996. [4] A.J. Stewart, C.A. Blindauer, S. Berezenko, D. Sleep, P.J. Sadler, Interdomain zinc site on human albumin, Proc. Natl. Acad. Sci. USA 100 (2003) 3701–3706. [5] A.A. Spector, Structure and lipid binding properties of serum albumin, Methods Enzymol. 128 (1986) 320–339. [6] N.A. Kratochwil, W. Huber, F. Muller, M. Kansy, P.R. Gerber, Predicting plasma protein binding of drugs: a new approach, Biochem. Pharmacol. 64 (2002) 1355–1374. [7] G. Sudlow, D.J. Birkett, D.N. Wade, The characterization of two specific drug binding sites on human serum albumin, Mol. Pharmacol. 11 (1975) 824–832. [8] G. Sudlow, D.J. Birkett, D.N. Wade, Further characterization of specific drug binding sites on human serum albumin, Mol. Pharmacol. 12 (1976) 1052– 1061. [9] G. Colmenarejo, In silico prediction of drug-binding strengths to human serum albumin, Med. Res. Rev. 23 (2003) 275–301. [10] S. Curry, Plasma albumin as a fatty acid carrier, Adv. Mol. Cell. Biol. 33 (2003) 29–46. [11] S. Curry, P. Brick, N.P. Franks, Fatty acid binding to human serum albumin: new insights from crystallographic studies, Biochim. Biophys. Acta 1441 (1999) 131–140. [12] A.A. Bhattacharya, T. Grune, S. Curry, Crystallographic analysis reveals common modes of binding of medium and long-chain fatty acids to human serum albumin, J. Mol. Biol. 303 (2000) 721–732. [13] S. Curry, H. Mandelkow, P. Brick, N. Franks, Crystal structure of human serum albumin complexed with fatty acid reveals an asymmetric distribution of binding sites, Nat. Struct. Biol. 5 (1998) 827–835. [14] R. Brodersen, S. Andersen, H. Vorum, S.U. Nielsen, A.O. Pedersen, Multiple fatty acid binding to albumin in human blood plasma, Eur. J. Biochem. 189 (1990) 343–349. [15] D.P. Cistola, D.M. Small, Fatty acid distribution in systems modeling the normal and diabetic human circulation. A 13C nuclear magnetic resonance study, J. Clin. Invest. 87 (1991) 1431–1441. [16] Jmol: an open-source Java viewer for chemical structures in 3D. http:// www.jmol.org/. [17] G.J. Quinlan, G.S. Martin, T.W. Evans, Albumin: biochemical properties and therapeutic potential, Hepatology 41 (2005) 1211–1219. [18] K. Oettl, V. Stadlbauer, F. Petter, J. Greilberger, C. Putz-Bankuti, S. Hallstrom, C. Lackner, R.E. Stauber, Oxidative damage of albumin in advanced liver disease, Biochim. Biophys. Acta 1782 (2008) 469–473. [19] 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 correlation with renal dysfunction: examination with the redox state of albumin, Kidney Int. 66 (2004) 1988–1993. [20] J. Himmelfarb, E. McMonagle, Albumin is the major plasma protein target of oxidant stress in uremia, Kidney Int. 60 (2001) 358–363. [21] K. Oettl, G. Reibnegger, O. Schmut, The redox state of human serum albumin in eye diseases with and without complications, Acta Ophthalmol. 89 (2011) 174–179. [22] M.L. Wratten, L. Sereni, C. Tetta, Oxidation of albumin is enhanced in the presence of uremic toxins, Ren. Fail. 23 (2001) 563–571.
Influence of HSA-bound FA on the HSA-SH reactivity / V.B. Jovanovic´ et al. / Anal. Biochem. 448 (2014) 50–57 [23] P. Faure, R. Tamisier, J.P. Baguet, A. Favier, S. Halimi, P. Levy, J.L. Pepin, Impairment of serum albumin antioxidant properties in obstructive sleep apnoea syndrome, Eur. Respir. J. 31 (2008) 1046–1053. [24] P.J. Thornalley, Protein and nucleotide damage by glyoxal and methylglyoxal in physiological systems—role in ageing and disease, Drug Metab. Drug Interact. 23 (2008) 125–150. [25] J. Zeng, M.J. Davies, Protein and low molecular mass thiols as targets and inhibitors of glycation reactions, Chem. Res. Toxicol. 19 (2006) 1668–1676. [26] J. Travis, J. Bowen, D. Tewksbury, D. Johnson, R. Pannell, Isolation of albumin from whole human plasma and fractionation of albumin-depleted plasma, Biochem. J. 157 (1976) 301–306. [27] A. Guerin-Dubourg, A. Catan, E. Bourdon, P. Rondeau, Structural modifications of human albumin in diabetes, Diabetes Metab. 38 (2012) 171–178. [28] V.B. Jovanovic, A.Z. Penezic-Romanjuk, I.D. Pavicevic, J.M. Acimovic, L.M. Mandic, Improving the reliability of human serum albumin-thiol group determination, Anal. Biochem. 439 (2013) 17–22. [29] R.J. Leatherbarrow, P.D. Dean, Studies on the mechanism of binding of serum albumins to immobilized cibacron blue F3G A, Biochem. J. 189 (1980) 27–34. [30] B.T. Dumas, W.A. Watson, H.G. Biggs, Albumin standards and the measurement of serum albumin with bromcresol green, Clin. Chim. Acta 258 (1997) 21–30. [31] G.R. Kingsley, The determination of serum total protein, albumin and globulin by the biuret reaction, J. Biol. Chem. 131 (1939) 197–200. [32] R.F. Chen, Removal of fatty acids from serum albumin by charcoal treatment, J. Biol. Chem. 242 (1967) 173–181. [33] G. Bulaj, T. Kortemme, D.P. Goldenberg, Ionization-reactivity relationships for cysteine thiols in polypeptides, Biochemistry 37 (1998) 8965–8972. [34] Y. Matsuyama, H. Terawaki, T. Terada, S. Era, Albumin thiol oxidation and serum protein carbonyl formation are progressively enhanced with advancing stages of chronic kidney disease, Clin. Exp. Nephrol. 13 (2009) 308–315.
57
[35] A. Hayakawa, K. Kuwata, S. Era, M. Sogami, H. Shimonaka, M. Yamamoto, S. Dohi, H. Hirose, Alteration of redox state of human serum albumin in patients under anesthesia and invasive surgery, J. Chromatogr., B: Biomed. Sci. Appl. 698 (1997) 27–33. [36] P. Manavalan Jr., W.C. Johnson, Sensitivity of circular dichroism to protein tertiary structure class, Nature 305 (1983) 831–832. [37] I. Sjoholm, I. Ljungstedt, Studies on the tryptophan and drug-binding properties of human serum albumin fragments by affinity chromatography and circular dichroism measurements, J. Biol. Chem. 248 (1973) 8434–8441. [38] J. Reichenwallner, D. Hinderberger, Using bound fatty acids to disclose the functional structure of serum albumin, Biochim. Biophys. Acta 1830 (2013) 5382–5393. [39] J.R. Simard, P.A. Zunszain, J.A. Hamilton, S. Curry, Location of high and low affinity fatty acid binding sites on human serum albumin revealed by NMR drug-competition analysis, J. Mol. Biol. 361 (2006) 336–351. [40] C.E. Petersen, C.E. Ha, K. Harohalli, J.B. Feix, N.V. Bhagavan, A dynamic model for bilirubin binding to human serum albumin, J. Biol. Chem. 275 (2000) 20985–20995. [41] M. de Lorgeril, P. Salen, P. Defaye, M. Rabaeus, Recent findings on the health effects of omega-3 fatty acids and statins, and their interactions: do statins inhibit omega-3?, BMC Med 11 (2013) 5. [42] J.H. Garman, S. Mulroney, M. Manigrasso, E. Flynn, C. Maric, Omega-3 fatty acid rich diet prevents diabetic renal disease, Am. J. Physiol. Renal Physiol. 296 (2009) 306–316. [43] B.E. Karlstrom, A.E. Jarvi, L. Byberg, L.G. Berglund, B.O. Vessby, Fatty fish in the diet of patients with type 2 diabetes: comparison of the metabolic effects of foods rich in n-3 and n-6 fatty acids, Am. J. Clin. Nutr. 94 (2011) 26–33.