Multiparametric protocol for the determination of thiol redox state in living matter

Free Radical Biology and Medicine 74 (2014) 85–98

Contents lists available at ScienceDirect

Free Radical Biology and Medicine journal homepage: www.elsevier.com/locate/freeradbiomed

Methods in Free Radical Biology and Medicine

Multiparametric protocol for the determination of thiol redox state in living matter Konstantinos Grintzalis, Ioannis Papapostolou, Dimitris Zisimopoulos, Irene Stamatiou, Christos D. Georgiou n Genetics, Cell and Developmental Biology Section, Department of Biology, University of Patras, Patras, Greece

art ic l e i nf o

a b s t r a c t

Article history: Received 15 April 2014 Received in revised form 13 June 2014 Accepted 20 June 2014 Available online 1 July 2014

Thiol redox state (TRS) evaluation is mostly restricted to the estimation of GSH and GSSG. However, these TRS parameters can estimate the GSSG/GSH potential, which might be useful for indicating abnormalities in redox metabolism. Nonetheless, evaluation of the multiparameric nature of TRS is required for a more accurate assessment of its physiological role. The present protocol extends the partial assessment of TRS by current methodologies. It measures 15 key parameters of TRS by two modular subprotocols: one for the glutathione (GSH)- and cysteine (CSH)-based nonprotein (NP) thiols/ mixed disulfides (i.e., GSH, GSSG, GSSNP, CSH, CSSNP, NPSH, NPSSNP, NPxSHNPSSNP, NPxSHNPSH), and the other for their protein (P) thiols/mixed disulfides (i.e., PSH, PSSG, PSSC, PSSNP, PSSP, NPxSHPSSNP). The protocol eliminates autoxidation of GSH and CSH (and thus overestimation of GSSG and CSSNP). Its modularity allows the determination GSH and GSSG also by other published specific assays. The protocol uses three assays; two are based on the photometric reagents 4,40 -dithiopyridine (DTP) and ninhydrin (NHD), and the third on the fluorometric reagent o-phthaldialdehyde (OPT). The initial assays employing these reagents have been extensively modified and redesigned for increased specificity, sensitivity, and simplicity. TRS parameter values and their standard errors are estimated automatically by sets of Exceladapted algebraic equations. Protocol sensitivity for NPSH, PSH, NPSSNP, PSSP, PSSNP, CSH, CSSNP, PSSC, NPxSHNPSSNP, and NPxSHNPSH is 1 nmol –SH/CSH, for GSSNP 0.2 nmol, for GSH and GSSG 0.4 nmol, and for PSSG 0.6 nmol. The protocol was applied on human plasma, a sample of high clinical value, and can be also applied in any organism. & 2014 Elsevier Inc. All rights reserved.

Keywords: Thiol redox state Glutathione Cysteine Nonprotein thiols Protein thiols

Introduction Evaluating cellular thiol redox state (TRS) is very important in a wide variety of biological processes [1,2]. TRS comprises many

Abbreviations: TRS, thiol redox state; GSH, glutathione; GSSG, oxidized glutathione; GSSNP, mixed disulfides of glutathione with nonprotein (NP) thiols; CSH, cysteine; CSSNP, mixed disulfides of cysteine with nonprotein (NP) thiols; NPSH, reduced nonprotein (NP) thiols; NPSSNP, mixed disulfides of nonprotein thiols (NPS); NPxSHNPSSNP, the NP thiol component of NPSSNP, excluding the NPSH components GSH and CSH; NPxSHNPSH, the NP thiol component of NPSH, excluding the NPSH components GSH and CSH; PSH, reduced protein (P) thiols; PSSG, mixed disulfides of glutathione (GS) with protein thiols (PS); PSSC, mixed disulfides of cysteine (CS) with protein (PS) thiols; PSSP, mixed disulfides of protein thiols (PS); PSSNP, mixed disulfides of nonprotein (NP) with protein (PS) thiols; NPxSHPSSNP, the NP thiol component of PSSNP, excluding the NPSH components GSH and CSH; Note: abbreviations of the reagents used in the present study are listed in the section of Materials under “Reagents.” n Corresponding author. Fax: þ 30 2610 997840. E-mail addresses: [email protected] (K. Grintzalis), [email protected] (I. Papapostolou), [email protected] (D. Zisimopoulos), [email protected] (I. Stamatiou), [email protected] (C.D. Georgiou). http://dx.doi.org/10.1016/j.freeradbiomed.2014.06.024 0891-5849/& 2014 Elsevier Inc. All rights reserved.

protein (P) and nonprotein (NP) thiol parameters such as glutathione (GSH), cysteine (CSH), and reduced NP and P thiols (NPSH and PSH, respectively), as well as their symmetric and mixed disulfides. The commonly used indicators for the assessment of TRS are restricted to GSH and its symmetric oxidized disulfide (GSSG). The role of GSH in intracellular redox state signaling, enzymic antioxidant defense, and in response to environmental factors is well established [3,4]. Additionally, GSH regulation is related to many physiological processes, in which CSH and PSH and their mixed and symmetric protein disulfides (PSSNP and PSSP, respectively) are also involved [5]. Glutathione disulfide (GSSG) can activate many enzymes (e.g., glucose-6-phosphatase and acid phosphatase) and inhibit others (e.g., glycogen synthetase, pyruvate kinase, adenylate cyclase, phosphorylase/phosphatase, and ribonucleotide reductase) [4]. Aerobic organisms need a system to restore key sulfhydryl groups to their reduced state after exposure to oxidant stress. Without a process to reduce protein disulfides, vulnerable cysteinyl residues of essential enzymes might remain oxidized, leading to changes in catalytic activity. This function is fulfilled by the thiol-disulfide exchange catalyzed by thiol transferases in the

86

K. Grintzalis et al. / Free Radical Biology and Medicine 74 (2014) 85–98

presence of GSH. Many proteins are activated or inhibited in vitro by the disulfide exchange between the protein and the GSH or other small-molecule disulfides (R0 SHþRSSR2R0 SSR þRSH). Whereas many proteins are active when the key sulfhydryls are in the thiol form, others require them to be in the oxidized, disulfide form. Because the thiol-transferase reaction is bidirectional, the equilibrium is determined by the redox state of the cell. Moreover, cells avoid CSH autooxidation to CSSC, producing potentially toxic oxygen radicals by storing most of the nonprotein CSH as GSH [4]. The role of NP thiols other than GSH and GSSG in mixed disulfides (NPSSNP), such as coenzyme A (CoASH), in the TRS of cells is usually underestimated, although they have significant contribution to the physiological consequences of oxidative stress [5,6]. Such examples are the increase of CoASSG mixed disulfide levels during hydroperoxide metabolism [7], its inhibitory effect on enzymes such as GSSG reductase (GR), phosphofructokinase and fatty acid synthase, and its activating action on fructose 16-bisphosphatase [4]. PSH, PSSP, and PSSNP are also important indicators of oxidative stress. Although PSSNP are often designated in the literature as PSSG [8] due to GSH contribution (glutathionylation), this designation is limiting since they may also contain CSH (designated as PSSC; cysteinylation)—a particular form of protein S-thiolation is cysteyl-glycylation [9]—and to a lesser degree other minor NP thiols (NPxSHPSSNP, e.g., CoASH). Interdisulfide bridges between different protein thiols as well as intradisulfide bridges within proteins undergoing oxidative tertiary structure modification can produce symmetric disulfides (PSSP) [10]. These can be significant thiol redox markers since they are involved in inactivation of enzymes, transporters, and transcriptional factors during oxidative stress conditions [10], and also in pathological processes associated with oxidative stress in animal and plant cells under stress [10–13]. Protein thiols/disulfides can be extracellular and part of membrane and subcellular structures, contributing to protein stability [14] and the regulation of redox homeostasis [1]. Of special metabolic interest is the involvement of the ROS-generated protein cysteine sulfenic acid in the formation of mixed GSH disulfides via thiol-disulfide exchange reactions [15–18]. Redoxdependent signaling events involving the posttranslational oxidative modification of proteins have now been accepted as an important regulatory process, controlled by a number of redoxdependent modifications of some protein cysteinyl thiols, including an interchange between the reduced thiol and several different oxidized disulfide states [19]. It has been proposed that catalytically important sulfhydryl groups in PSH are protected from oxidative stress by reacting reversibly with GSH to form PSSG [20]. PSSG can inactivate enzymes (e.g., fructose-1,6-biphosphate aldolase) and make proteins more or less susceptible to proteolysis [21]. Also, specific oxidation/reduction of particular protein thiols may represent an important event in cellular and oxidative signaling cascades [22,23]. In addition, protein thiols could react with nitric oxide radical to form S-nitrosothiols, which are involved in signal transduction and posttranslational protein modification [24]. All these render the simultaneous quantitative assessment of thiols in organisms very important, since cellular TRS is associated with normal as well as abnormal metabolic processes.

TRS evaluation: Methodologies and limitations TRS is normally evaluated by the total cellular thiol content [25] or by GSH and GSSG [8,26,27]. GSH is estimated by photometric assays based on Ellman’s reagent (5,50 -dithiobis-2-nitrobenzoic acid, or DTNB) [28], which are not GSH specific since they do not

discriminate it from CSH [29]. GSSG is usually determined by enzymatic assays [8,30,31], while GSH has been also evaluated more specifically by HPLC after derivatization with N-ethylmaleimide (NEM) [30]. GSH and GSSG are also evaluated by the o-phthaldialdehyde (OPT)-based fluorometric assay [32]. However, if proteins in the sample are not effectively removed (e.g., by precipitation) they could cause interference because OPT, as with GSH, can bind covalently with closely spaced (  3 Å) sulfhydryl and ε-amino groups of cysteine and lysine residues [33,34]. Adopting this assay in the present protocol (see next section), we modified it to eliminate such protein interference and increase its specificity and sensitivity. Similarly, PSSG is underestimated by photometric assays because of oxidation of the measured GSH during the procedure [35,36]. There is also lack of assays for the simultaneous determination of CSH, nonprotein oxidized disulfides of CSH (cystine, CSSC, CSH-GSH disulfide, CSSG), and PSSC. The available methodologies, mainly HPLC based, measure the CSH component of the P/NP-CSH disulfides (via derivatized thiol detection by LC-MS after tri-n-butylphosphine, TBP, reduction [37], or by LC fluorescence following TBP [38] and tris(2-carboxyethyl)phosphine, TCEP [39], disulfide reduction). Concerning protein thiols, there are many methods available for measuring changes in the reduced protein thiol status of cells and tissues [19]. In general, these methods make use of chemicals that react with thiol groups with a variety of tags or labels, which allow the extent of incorporation to be measured. There are chemical methods involving thiol derivatization by thiol-labeling agents based on maleimide, iodoacetamide, iodoacetate, and thiosulfates. These agents may take the form of an affinity label such as biotin, a fluorophore, a radionucleotide, or a label that changes the molecular weight of the protein. Other protein thiollabeling agents include alkyl halides, arylating agents, thiosulfates, and disulfide compounds (such as DTNB). Antibodies to GSH have been used to investigate the thiol oxidation state of samples in enzyme-linked immunosorbent assays (ELISA). Moreover, cells can be metabolically labeled by incubation with [35S]cysteine, and can then undergo a variety of treatments, and S-thiolation (normally interpreted as S-glutathiolation) can be measured [19]. Normally, the PSH component of PSSNP, PSSG, and PSSC has been determined by Ellman’s reagent-based assays [40,41], which do not discriminate the NP thiol component of PSSG, PSSC, and PSSNP [41]. Other methodologies quantify protein thiols without correlating the PS thiol component of PSSG, PSSC, PSSNP, and PSSP to its corresponding NP thiol component, which they do not measure [42]. However, simultaneous determination of incomplete sets of P- and NP-TRS parameters based on GSH and CSH has been performed by GC-MS [37], laser-induced fluorescence capillary electrophoresis [43], and HPLC methodologies [38,39,42,44–46], as well as by fluorometric and photometric assays [47–52]. However, in a case where plasma total thiols were determined after reduction with TCEP, the involved HPLC methodology [39] does not discriminate P from NP thiols (e.g., CSH and GSH resulting from the reduction of PSSC and PSSG, respectively).

Advantages of a multiparametric TRS protocol Restricting TRS evaluation to the estimation of GSH and GSSG can offer a calculation of the GSSG/GSH potential, which might be useful at best as an analytical tool to disclose disturbances in redox metabolism [53]. Even so, a new technique of noninvasively measuring thiol redox potentials has shown that the GSH/GSSG ratio is 50,000 rather than about 100 in the cytosol [54], and that the cytosolic GSSG concentration is much more tightly regulated than expected (GSSG is rapidly transported into the vacuole by the ABC-C transporter Ycf1 in yeast), providing also a mechanistic

K. Grintzalis et al. / Free Radical Biology and Medicine 74 (2014) 85–98

explanation for the discrepancy with conventional measurements [54]. Given the partial assessment of TRS by current methodologies, the evaluation of its multiparameric nature is required for the accurate assessment of its physiological role. This study presents a multiparametric protocol that estimates 15 of the most important P and NP thiol/disulfide TRS parameters. The protocol is modular to adapt to any experimental needs as they relate to TRS. The protocol represents a major extension and redesigning of two TRS protocols previously developed by our lab [50,51]. Although limited in TRS parameters these protocols may have been, the scientific attention they have received is a reflection of the usefulness in biological research of TRS evaluation. In general, the protocol differs in offering a more extensive set of TRS parameters that are identified with increased specificity, sensitivity, and simplicity, while their estimation (and their statistical analysis) is automated by Excel-adapted algebraic sets of equations (see Supplementary Material). More specifically, the major differences and innovations of the present protocol are the following:

 It eliminates the artificial oxidation of GSH during sample



processing (and subsequent overestimation of GSSG), by using a modification of the NEM-derivatization procedure for GSH reported in the (GSH-GSSG-restricted) Giustarini et al. 2013 protocol [30]. However, the present protocol does not adopt the procedures for GSH and GSSG (via HPLC, and the GR recycling assay, respectively) of the aforementioned protocol, because they are somewhat complex and time consuming for an also time consuming multiparametric TRS protocol. Nonetheless, the modularity of the present protocol allows their selection (as well as that of the GR recycling assay in Rahman et al. 2006 protocol [31]) as optional. Given the applicability of the GSHNEM derivatization step on other experimental models besides blood [30,55], its adoption by the present protocol extents its use to any organism as well. It eliminates autoxidation of CSH (and thus overestimation of CSSNP) by inclusion of catalase (CAT) in the sample











87

homogenization buffers. CSH autoxidation to cystine can be caused by H2O2 in the presence of transition metals [56,57]. It measures higher numbers of TRS parameters and of quite different thiol types, and with much higher specificity and simplicity than the Patsoukis and Georgiou 2004 and 2005 protocols [50,51]. For example, the protocol also measures CSH and its P/NP disulfides, excluding the less TRS important N-acetylcysteine that is measured by the Patsoukis and Georgiou 2004 protocol [50]. It measures GSH and GSSG with high specificity (equaling that of the Giustarini et al. 2013 protocol [30]; data not shown) by using a simple chemical assay that is based on the fluorescent probe OPT. The Patsoukis and Georgiou 2005 and 2005 protocols, instead, measure GSH and GSSG nonspecifically by the photometric DTNB and the fluorescent monobromobimane probes, respectively. It measures general TRS thiols with the photometric probe 4,40 dithiopyridine (DTP), a highly specific reagent for -SH groups, and not with the prone to interference DTNB probe (which has been used in other assays and also by the Patsoukis and Georgiou 2004 protocol). It uses TBP or TCEP to reduce thiol disulfides by a nearly 100% efficiency instead of the less effective disulfide reductants sodium borohydride (NaBH4) and NaBH4/dithithiothreitol (DTT) used in the Patsoukis and Georgiou 2004 and 2005 protocols, respectively. It replaces the ineffective trichloroacetic acid (TCA) protein precipitation step with a certain combination of deoxycholate (DOC) and TCA, which precipitates proteins as low as 2–5 mg by Z90% effectiveness [58,59]. Protein precipitation with TCA alone is used in the Patsoukis and Georgiou 2004 and 2005 protocols, and is normally used in most thiol protocols including the Giustarini et al. 2013 protocol [30].

The present protocol consists of two subprotocols: one for the quantification of the NP-TRS and the other for the P-TRS parameters after appropriate reduction of involved disulfide bonds (Fig. 1).

Fig. 1. Schematic representation of the quantification of nonprotein (NP) and protein (P) thiol parameters of TRS by the present protocol. The main NP and P TRS parameters (estimated after reduction of the P/NP symmetric/mixed disulfides with TBP or TCEP) are highlighted in green and orange bold, respectively. Abbreviations in parenthesis show the reagents OPT, NHD, and DTP that are involved in the quantification of the indicated TRS parameters; optional methods for GSH (as GS-NEM) and GSSG quantification by HPLC and the GR recycling assay, respectively, are also stated (in parentheses). TRS parameters shown inside parentheses without the equality symbol are those derived by corresponding algebraic equations (see section “Estimation of the TRS parameters”), which involve certain fractions (A to L) quantified by the protocol. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

88

K. Grintzalis et al. / Free Radical Biology and Medicine 74 (2014) 85–98

In addition to GSH and GSSG, the NP-TRS protocol quantifies the NP-thiols/disulfides GSSNP (¼ GSHGSSNP), CSH, CSSNP (¼ CSHCSSNP), NPSH (¼ GSHþCSHþ NPxSHNPSH), NPxSHNPSH (the unidentified NP thiol component/s of NPSH, excluding GSH and CSH), NPSSNP (¼ GSSGþGSSNPþCSSNPþ NPxSHNPSSNP), and NPxSHNPSSNP (the unidentified NP thiol component/s of NPSSNP, excluding GSH and CSH); GSH, GSSNP, NPSSNP, CSSNP, NPxSHNPSH, and NPxSHNPSSNP are derived algebraically. The P-TRS subprotocol measures the P-thiols/disulfides PSH (containing reduced protein cysteines), PSSG, PSSC, PSSNP (¼ PSSGþPSSCþNPxSHPSSNP), NPxSHPSSNP (the unidentified NP thiol component/s of PSSNP, excluding GSH and CSH), and PSSP (intra- plus interprotein disulfides); PSSP and NPxSHPSSNP are derived algebraically. The two subprotocols are based on three different thiol probes. These are the photometric reagents DTP and ninhydrin (NHD), and the fluorometric reagent OPT (Fig. 2). Although the mechanism of OPT reaction with GSH is already known (i.e., via formation of covalent bonds, one with the sulfhydryl and the other with the εamino groups of cysteine and lysine, respectively [33]), the fact that OPT also reacts specifically with GSSG at pH Z 12 [32,33,60] has not been explained. We propose a possible mechanism involving GSSG reaction with OPT at alkaline pH Z12, where it is hydrolyzed to GS̄ (Fig. 2) as similarly are hydrolyzed protein disulfides and low molecular weight aliphatic and aromatic disulfides (2RSSR þ4OH̄ -3RS̄ þRSO2̄ þ2 H2O [61]). Then, GS̄ would react with OPT to form a GS-OPT fluorophore [33]. It should be noted that GS-OPT remains stable at the pH range 8-12, as indicated by the stability of its fluorescence [60]. Thus, it is stable at pHs 8 and 12 used for GSH and GSSG determination,

Fig. 2. Reaction mechanisms of thiol/mixed disulfide-based -SH groups, cysteines, and GSSG/GSH with the respective DTP [40], NHD [71], and OPT reagents used in the multiparametric TRS protocol. OPT use in the GSSG determination involves alkaline hydrolysis of GSSG (at pH 12), where the resulting GS̄ reacts with OPT to form the GS-OPT fluorophore (as also GSH does) [33] (a detailed explanation of the OPT-based GSSG assay reaction mechanism is presented in the Introduction).

respectively [32]. Below pH 8, GS-OPT is unstable as shown by the decrease of its fluorescence intensity [60]. The initial DTP-, NHD-, and OPT-based assays have been modified in the present protocol for increased specificity, stability, and ease of use. The DTP reagent quantifies the NP thiols NPSH, NPSSNP, PSH, PSSNP, PSSP, NPxSHNPSH, NPxSHNPSSNP, and NPxSHPSSNP. For the determination of the -SH groups in the P thiols PSH and NPSH, in the PS- part of PSSP and PSSNP (after their reduction), and in the NPS- part of PSSNP and NPSSNP (after their reduction), we chose DTP instead of DTNB for the following reasons: DTP reacts with thiols in a thiol-disulfide exchange reaction accompanied by the stoichiometric release of the chromogenic compound 4-thiopyridone. Because the electrostatic milieu of cysteine thiols in proteins may have substantial effects on the rates of thiol-disulfide exchange reactions, DTP has a great advantage over DTNB due to its hydrophobic nature and somewhat small size. Protonation of the pyridyl nitrogens increases the reactivity of DTP dramatically. This almost completely compensates for the decrease in reaction rate at low pH due to protonation of the nucleophilic thiolate anion. Therefore, the reagent can be used at low pH where thiol-disulfide exchange reactions normally do not occur. DTP gives off only two 4-PS moles when reacting with NPSH or mixed (P/NP) disulfides (after their reduction), and is not involved in SH/SS exchange reactions, making it the best -SH reagent [10,11]. On the other hand, DTNB-based assays have the following disadvantages: 5-thio-2-nitrobenzoate, the product of DTNB reaction with -SH groups, can be also formed by the hydrolysis of DTNB, making DTNB not a safe quantitative indicator of thiols. DTNB instability due to hydrolysis has been estimated as an artificial 1.2 μM h–1 increase in thiol concentration at pH 8.0 [40]. Another limitation in using DTNB is that unwanted oxidation of thiols occurs at pH values above 7.0 (DTNB assays are performed usually at pH 7.5), because SH/SS exchange reactions of the final product are generally occurring at this pH range [45]. The NHD reagent quantifies specifically free CSH and the CS part of CSSNP and PSSC after their reduction. NHD allows the discrimination of CSH in mixtures of NP thiols that are determined by the nonspecific DTP and DTNB assays. The initial NHD-based assay [51,62] has been modified in the present protocol in order to stabilize the absorbance of the chromophore and increase the sensitivity of the method. The OPT reagent quantifies specifically GSH and GSSG, and the GS part of GSSNP and PSSG after their reduction. The initial assay [32] has been modified for use in the present protocol in order to become more specific by removing interfering proteins [33,34]. For this, we replaced the metaphosphoric acid step in the initial assay by use of a more effective precipitation step; we used a certain DOC-TCA combination which precipitates as slow as 5 mg ml  1 proteins with Z95% efficiency [58,59] (thus also allowing the measurement of very low protein samples). This step is followed by a protein pellet and aqueous fraction washing step with diethyl ether, in order to remove any TCA traces (besides DOC), which interfere with the OPT assay. These modifications achieve a  4000-fold increase in sensitivity for GSH and GSSG over the initial assay, due to the high sample dilutions it requires in order to eliminate metaphosphoric acid interference. Moreover, the protein thiol TRS parameters can be evaluated in samples with low protein concentration, as it is estimated by an ultrasensitive (100–200 ng) Coomassie brilliant blue (CBB) G-250 assay [63]. The determination of NP and P thiol disulfides requires prior reduction of -SS- bonds, which is commonly done by dithiothreitol. However, the reduction reaction is rather time-consuming and most importantly there is an absolute requirement for the complete removal of dithiothreitol prior P/NP thiol quantification [40]. An alternative reductant, borohydride, has been also employed, taking advantage of the easy removal of its excess by acid

K. Grintzalis et al. / Free Radical Biology and Medicine 74 (2014) 85–98

neutralization. However, its reductive effectiveness is low at the alkaline pH required for certain parameters of TRS [50,51]. We chose trialkylphosphines because they are very strong -SS- group reductants (more powerful than dithiothreitol [48]), reduce disulfides stoichiometrically and irreversibly, and are nonreactive toward many other functional groups [49]; TCEP [39] and TBP [37,38] have been already used in disulfide bond reduction. Hydrophobic TBP was selected for the DTP-based assays in the protocol because it can be easily removed by extraction in butanolcyclohexane (5:1 v/v). On the other hand, the water-soluble TCEP was preferred for the NHD- and OPT-based assays because it does not interfere with them. The present protocol was applied in human plasma, a sample of established clinical value, for two main reasons: (1) As any extracellular fluid, blood plasma is considered a “difficult” sample because of its low concentration levels on GSH and GSSG (both present at  1 to 2 μM) [30], and (2) because the evaluation of the other 13 parametres of TRS could be also of clinical usefulness.

Materials Instrumentation Balance (Kern, 770/65/6 J) Bench-top centrifuge (Hermle, Model Z206A) Glass Pasteur pipettes (internal diameter 0.5 cm, 22 cm long, by Hirschmann Laborgeräte GmbH & Co, Germany) Microcentrifuge clear tubes, 1.5 and 2 ml (VWR, Cat. No. 89000028) Micropipettes (adjustable volume) 2.5 μl, 10 μl, 20 μl, 100 μl, 200 μl, 1 ml, and tips (Eppendorf Research) Microcuvette for absorbance measurements (12.5  12.5  45 mm external dimensions, 4 mm internal window, and 9 mm bottom, 1.16 ml, quartz; Starna 9/B/9/Q/10) Microcuvette for fluorescence measurements (45  4 mm, 0.5 ml, quartz; Starna SOG/Q), with its FCA4 adapter pH meter (Metrohm, 827 pHlab) Spectrofluorometer (Shimadzu, Model RF-1501) Spectrophotometer (Hitachi, Model UV-VIS U-1800) Reagents Acetic acid (glacial) 100% (Merck, Cat. No. 1.00063.1011)! CAUTION Corrosive. Acetone (AC; Merck, Cat. No. 01-6300117)! CAUTION Highly flammable and harmful. Acivicin (Sigma-Aldrich, Cat. No. SML0312). Aldrithiol (4,40 -dithiopyridine or DTP; Sigma-Aldrich, Cat. No. 143057)! CAUTION Irritant. Catalase from bovine liver (CAT; Sigma, Cat. No. C-9322). Chloroform (CHCl3; Merck, Cat. No. 1.02445)! CAUTION Highly flammable and harmful. Cumene hydroperoxide, CumOOH (Sigma, Cat. No. C-0524)! CAUTION Oxidizing, Corrosive. Cyclohexane (Sigma-Aldrich, Cat. No. 179191)! CAUTION Highly flammable and harmful. L-Cysteine (CSH; Sigma-Aldrich, Cat. No. C-7352). Deoxycholic acid, sodium salt (DOC; Sigma-Aldrich, Cat. No. D-6750). Diethyl ether (DE; Merck, Cat. No. 1.00921)! CAUTION Highly flammable and harmful. Dimethyl sulfoxide (DMSO; Sigma-Aldrich, Cat. No. 276855)! CAUTION Harmful. Ethylenediaminetetraacetic acid (Na2EDTA; Merck, Cat. No. 324503)! CAUTION Irritant, dangerous for the environment.

89

Formamide (100% w/w or 25 M, Merck, Cat. No. 1.09684.1000)! CAUTION Very toxic. Glutathione (GSH; Sigma-Aldrich, Cat. No.G-4251). L-Glutathione oxidized disodium salt (GSSG; Sigma-Aldrich, Cat. No. G-4626). Glutathione peroxidase from bovine erythrocytes (GPx; SigmaAldrich, Cat. No. G-6137). HCl conc. ( Z37% w/w; Fluka, Cat. No. 84415)! CAUTION Corrosive. Isobutanol (Merck, Cat. No. 1.00984.1000)! CAUTION Irritant, dangerous for the environment. Methanol (MetOH; 100%) for HPLC (Sigma-Aldrich, Cat. No. 34860)! CAUTION Highly flammable and harmful. N-Ethylmaleimide (NEM; Sigma-Aldrich, Cat. No. E-3876)! CAUTION Highly flammable and corrosive. Ninhydrin (NHD; Serva, Cat. No. 30410). N,N-Dimethylformamide (DMF, Merck, Cat. No. 1.03053.1000)! CAUTION Very toxic. Sodium phosphate (Νa2HPO4; Merck, Cat. No. 30412). o-Phthaldialdehyde (OPT; Sigma-Aldrich, Cat. No. P-1378)! CAUTION Irritant. NaOH (Merck, Cat. No. 567530)! CAUTION Corrosive. Tri-n-butylphosphine 97% (TBP; Sigma-Aldrich, Cat. No. 247049)! CAUTION Highly flammable and corrosive. Tris(2-carboxyethyl)phosphine hydrochloride, powder, Z 98% (TCEP; Sigma-Aldrich, Cat. No. C4706)! CAUTION Highly flammable and corrosive. Trichloroacetic acid (TCA; Merck, Cat. No. 1.00807.0250)! CAUTION Corrosive, dangerous for the environment. Urea (Sigma-Aldrich, Cat. No. U1250). Water (ddH2O; purified by a Milli-Q system, Millipore Corp). Reagent setup

 100% AC: Keep at -20 1C.  500 mM phosphate buffer, 1 mM acivicin, 30 mM EDTA, pH 7.5

 

    



(500PAE buffer, pH 7.5): Prepare fresh by dissolving 0.891 g Na2 HPO4.2H2O, 111 mg Na2EDTA, 1.8 mg in 8 ml ddH2O. Adjust the pH to 7.5 and final volume of the solution to 10 ml. 2000 Units CAT ml-1: Prepare fresh by dissolving 1 mg CAT in 1 ml ddH2O. Keep on ice. 500 mM phosphate buffer, 1 mM acivicin, 30 mM EDTA, 300 mM NEM, pH 7.5 (500PAEN buffer, pH 7.5): Prepare fresh by dissolving 37.5 mg NEM in 1 ml 500PAE buffer, pH 7.5. Keep on ice. 300 Units GPx ml-1: Prepare fresh by dissolving 1 mg GPx in 1 ml ddH2O. Keep on ice. 10 mM CumOOH: Dilute CumOOH stock (5.52 M) to 50 mM with 100% MetOH. Dilute further this stock to 10 mM with ddH2O. 50, 100% TCA: Prepare the 100% TCA fresh by dissolving 10 g TCA in ddH2O to a final 10 ml. Prepare the 50% TCA by 2X dilution of 100% TCA with ddH2O. 1% DOC: Prepare fresh by dissolving 0.01 g DOC in ddH2O to final 1 ml. 10% TCA, 0.02% DOC, 50PAE buffer, pH 7.5, washed with DE (10TCA-DOC-50PAE-DE): Prepare fresh by mixing 2.5 ml 100% TCA with 19.5 ml ddH2O, 2.5 ml 500PAE buffer, pH 7.5, and 0.5 ml 1% DOC, followed by centrifugation at 16,000g for 2 min. Then, wash 3  with 25 ml DE by vortexing, followed by centrifugation at 16,000g for 2 min and discard the upper DE phase. 10% TCA, 0.02% DOC, 50PAEN buffer, pH 7.5, washed with DE (10TCA-DOC-50PAEN-DE): Prepare fresh by mixing 0.5 ml 100% TCA with 3.9 ml ddH2O, 0.5 ml 500PAEN buffer, pH 7.5, and 0.1 ml 1% DOC, followed by centrifugation at 16,000g for 2 min.

90



 

   



        

K. Grintzalis et al. / Free Radical Biology and Medicine 74 (2014) 85–98

Then, wash 3  with 5 ml DE by vortexing, followed by centrifugation at 16,000g for 2 min and discard the upper DE phase. 4, 0.1 M Acetic acid, pH 4.5: For preparing the 4 M solution, dilute the concentrated stock (16.6 M) 4.165X with ddH2O and adjust to pH 4.5 with 10 M NaOH. For preparing the 0.1 M solution dilute the 4 M stock solution 40X with ddH2O, and adjust to pH 4.5 with 1 M NaOH. 12 Μ Formamide: Prepare fresh by mixing 0.25 ml 4 M acetic acid, pH 4.5, with 25.75 ml ddH2O and 24 ml 100% formamide, and adjust pH to 4.5 with 1 M NaOH. 0.5, 1 M phosphate buffer, pH 8.0: Prepare the 1 M solution by dissolving 7.12 g Na2 HPO4.2H2O in ddH2O to final 40 ml and adjust the pH to 8.0 (with 10 M HCl). The 0.5 M solution is made by 2X dilution of the 1 M stock. 50 mM TBP: Prepare fresh by diluting 80X with DMF the commercial TBP stock (4 M). 2.5 Μ Urea: Prepare fresh by dissolving 7.5 g urea in a final volume 50 ml ddH2O. 50 mM TCEP: Prepare fresh by dissolving 14 mg TCEP in 1 ml ddH2O. The reagent must be prepared fresh before use. 10% TCA, 0.02% DOC, 7 M formamide, 0.25 M phosphate buffer (pH 8.0), 2 mM TBP (TCA-DOC-F-P-T): Prepare fresh (20.4 ml) by mixing 12 ml 12 Μ formamide with 5.2 ml 1 M phosphate buffer, pH 8.0, 800 μl 50 mM TBP, 400 ml 1% DOC, and 2 ml 50% TCA, followed by centrifugation at 16,000g for 2 min. 10% TCA, 0.02% DOC, 2 M urea, 8 mM phosphate buffer (pH 8.0), 2 mM TCEP (TCA-DOC-U-P-T): Prepare fresh (12 ml) by mixing 10 ml 2.5 M urea with 200 μl 0.5 M phosphate buffer, pH 8.0, 500 μl 50 mM TCEP, 240 ml 1% DOC (final 0.02%), 1200 μl 100% TCA (final 10%), followed by centrifugation at 16,000g for 2 min. Then, wash 3  with equal volume DE by vortexing, followed by centrifugation at 16,000g for 2 min and discard the upper DE phase. 3 mM DTP: Prepare fresh by dissolving 7 mg DTP in 1 ml 100% DMSO and dilute 10X with ddH2O. 10% DMSO: Dilute 10X with ddH2O the 100% DMSO commercial stock. Acetic acid-HCl solution: Prepare fresh mixing 45 ml 100% acetic acid with 30 ml conc. HCl. 100 mM NHD: Prepare fresh by dissolving 0.79 g NHD in 40 ml acetic acid-HCl solution. 1 mg ml-1 OPT: Prepare fresh by dissolving 1 mg OPT in 1 ml 100% MetOH. Keep on ice and light-protected. 0.2 M NaOH: Prepare by dissolving 0.16 g NaOH in ddH2O to final 20 ml. 5 mM GSSG: Prepare fresh by dissolving 3.3 mg GSSG in 1 ml 10TCA-DOC-50PAE-DE. Keep on ice. 5 mM GSH: Prepare fresh by dissolving 3 mg GSH in 2 ml 10TCADOC-50PAE-DE. Keep on ice (stable for at least 2 weeks at -20 1C). 5 mM CSH: Prepare fresh by dissolving 3 mg CSH in 0.5 ml 0.5 M HCl, and then add 4.5 ml 10TCA-DOC-50PAE-DE. Keep on ice (stable for at least two weeks at -20 1C).

Construction of standard curves Standard curve for -SH groups in fractions A, B, H, I, and J (for definitions of fractions A, B, H, I, and J see Method procedure): From the 5 mM GSH solution, prepare a series of GSH standards (1 to 17 μM, in 10TCA-DOC-50PAE-DE) and mix 900 μl of each with 50 μl 4 M acetic acid, pH 4.5, and 50 μl 3 mM DTP. As reagent blank, 900 μl 10TCA-DOC-50PAE-DE in place of the GSH solution is used. After 10 min incubation at room temperature (RT), the absorbance of the mixtures is measured at 325 nm against the reagent blank using a photometric quartz microcuvette. Sensitivity: 1 nmol -SH.

Standard curve for fractions C, and D (for definitions of fractions C and D see Method procedure): From the 5 mM CSH solution, prepare a series of CSH standards (1 to 12 μM, in 10TCA-DOC50PAE-DE) and mix 2 ml of each with 2 ml 100% acetic acid and 2 ml 100 mM NHD. As reagent blank, 2 ml 10TCA-DOC-50PAE-DE in place of the CSH solution is used. After 10 min incubation at 100 1C, let the tubes cool, add to each 1 ml chloroform, vortex vigorously, centrifuge at 16,000g for 5 min, transfer the chloroform bottom phase into a glass microcuvette, and measure absorbance at 560 nm against the reagent blank. Sensitivity: 1 nmol CSH. CRITICAL: If the chloroform phase that results after the centrifugation step is not clear, you should not measure its absorbance because you will get erroneously high readings. The slight turbidity is due to the presence of minute water droplets entrapped in the chloroform phase during vortexing because the NHD reaction mixtures were not cooled well (to RT) before adding chloroform. To remove turbidity either recentrifuge the chloroform phase after recooling in ice water bath, or warm it shortly at 30 1C. Standard curve for fraction K (¼ PSSC) (for definition of fraction K see Method procedure): From the 5 mM CSH solution, prepare CSH standards (1 to 12 μM in TCA-DOC-F-P-T) and mix 2 ml of each with 2 ml 100% acetic acid and 2 ml 100 mM NHD. As reagent blank, 2 ml TCA-DOC-F-P-T in place of the CSH solution is used. After 10 min incubation at 100 1C, let the tubes cool, add to each 1 ml chloroform, vortex vigorously, centrifuge at 16,000g for 5 min at RT. Isolate the chloroform phase and wash 1 ml of it with 4 ml 4 M acetic acid by vortexing, followed by centrifugation at 16,000g for 5 min at RT, then isolate the acetic acid-washed chloroform (cleared from the color interference due to formamide), transfer it into a glass microcuvette, and measure its absorbance at 560 nm against the reagent blank. Sensitivity: 1 nmol CSH. CRITICAL: Same as in Standard curve for fraction C, D. CAUTION: Avoid plastic photometric cuvettes as they might be eroded by the chloroform. Standard curve for fractions E ( ¼ GSSG), and F (for definitions of fractions E and F see Method procedure): From the 5 mM GSSG solution, prepare a series of GSSG standards (2 to 20 μM, in 10TCADOC-50PAE-DE) and mix 230 μl of each with 110 μl 0.2 M NaOH and 20 μl 1 mg ml-1 OPT. As reagent blank, 230 μl 10TCA-DOC50PAE-DE in place of the GSSG solution is used. After 20 min incubation at RT in the dark, the fluorescence at ex/em 340/ 420 nm is measured against the reagent blank using a fluorometric quartz microcuvette (with the spectrofluorometer in use set at low sensitivity and slit width 10 nm). Sensitivity: 0.4 nmol GSSG. Standard curve for fraction G (for definition of fraction G see Method procedure): From the 5 mM GSH solution, prepare a series of GSH standards (1 to 15 μM, in 10TCA-DOC-50PAE-DE) and mix 190 μl of each with 20 μl 50 mM TCEP, 130 μl 0.5 M phosphate buffer, pH 8.0, and 20 μl 1 mg ml-1 OPT. As reagent blank, 190 μl 10TCA-DOC-50PAE-DE in place of the GSH solution is used. After 20 min incubation at RT in the dark, the fluorescence at ex/em 340/420 nm is measured against the reagent blank using a fluorometric quartz microcuvette (with the spectrofluorometer in use set at low sensitivity and slit width 10 nm). Sensitivity: 0.2 nmol GSH. Standard curve for fraction L (¼PSSG) (for definition of fraction L see Method procedure): From the 5 mM GSH solution, prepare a series of GSH standards (3 to 35 μM, in TCA-DOC-U-P-T) and mix 200 μl with 130 μl 0.5 M phosphate buffer, pH 8.0, and 20 μl 1 mg ml-1 OPT is added. As reagent blank, 200 μl TCA-DOC-U-P-T in place of the GSH solution is used. After 20 min incubation at RT in the dark, the fluorescence at ex/em 340/420 nm is measured

K. Grintzalis et al. / Free Radical Biology and Medicine 74 (2014) 85–98

against the reagent blank using a fluorometric quartz microcuvette (with the spectrofluorometer in use set at low sensitivity and slit width 10 nm). Sensitivity: 0.6 nmol GSH.

Method procedure The protocol procedure (outlined in Fig. 3) is as follows: Sample treatment 1. Split sample (internal organs, solid tissues, cells, etc; for blood see as follows) in 1/5 and 4/5 portions (expressed in weight or volume for tissue/cell or liquid samples, respectively) and homogenize as follows: (a) The 1/5-sample is initially mixed with an appropriate volume of cold ddH2O (as to keep the final homogenization volume as low as possible), and to the resulting mixture volume 1/10 vol 500PAEN buffer, pH 7.5, is added, followed by homogenization (with a homogenization method appropriate for the sample type), and incubation for 5 min at room temperature (RT). The blood plasma 1/5-sample is collected in microcentrifuge tubes prefilled with 1/10 vol 500PAEN buffer, pH 7.5, and mixed by tilting four to five times for 5 min at RT. The 1/5-sample homogenate is centrifuged at 16,000g for 10 min at 4 1C, the resulting clear NEM-treated homogenate supernatant (designated 1/5-NEM-hs) is collected, and its protein content is measured as stated elsewhere [63] in order to express its TRS parameters per protein amount. PAUSE POINT: Samples can be stored at  80 1C because compared with results from fresh samples they did not affect these results even when stored for 3 months.

91

NOTE: Alternatively, the 1/5-NEM-hs can be used to quantify both GSSG and GSH by the more complicated (although  5 fold more sensitive) methods described in the Giustarini et al. 2013 protocol [30]; GSSG is determined by the nonendpoint GR recycling assay as also described elsewhere [31], and GSH as GS-NEM by HPLC, starting from substeps A or B in step 2 of the “Analytical procedures” section [30]. (b) The 4/5-sample is also mixed with an appropriate (for effective homogenization) volume of cold ddH2O, and to the resulting mixture 1/10 vol 500PAE buffer, pH 7.5, and 1/10 vol 2000 Units CAT ml  1 are added, followed by homogenization; the blood plasma 4/5-sample is collected in microcentrifuge tubes prefilled with 1/10 vol 500PAE buffer, pH 7.5, and 1/10 vol 2000 Units CAT ml  1, and mixed by tilting four to five times. The 4/5-sample homogenate is centrifuged at 16,000g for 10 min at 4 1C, the clear untreated homogenate supernatant is collected (designated 4/5-untr-hs), and its protein content is measured as stated in step 1a. NOTE: Acivicin inhibits γ-GT [64], thus preventing this enzyme from utilizing GSH as substrate, while NEM inhibits GR (and limits artifactual ex vivo reduction of GSSG) and prevents GSH autoxidation by forming a GS-NEM adduct, thus allowing the accurate determination of GSSG [30,65]. We did not use serine/borate to inhibit γ-GT in order to minimize possible interference by the homogenization buffers in the assays of the protocol. Instead, we counterbalanced the slow inactivation potential of acivicin by using 4.5-fold higher concentration (100 mM) than the one used in the Giustarini et al. 2013 protocol [30]. NOTE: CAT was added to eliminate CSH autoxidation caused by H2O2 in the presence of transition metals [56,57].

Fig. 3. General outline of the multiparametric TRS protocol. TRS parameters are enclosed in white boxes, and are determined by the indicated steps (described in detail in the text). Text in italics designates indicative starting sample fractions processed during the protocol procedure, as they are described in detail the protocol procedure.

92

K. Grintzalis et al. / Free Radical Biology and Medicine 74 (2014) 85–98

We verified this by spiking the sample homogenate with known concentrations of CSH, and testing for recovery (Method Procedure, step 9, fraction C treatment; data not shown). NOTE: The 1/5 and 4/5 sample portion splitting is based on the need of a small fraction (  1/5) for the determination of GSSG. It is a good starting approximation for most samples, and should be set experimentally considering also sample GSSG concentration. CRITICAL: When cell lysis homogenization buffers are used, they should not contain -SH groups (e.g., mercaptoethanol, dithiothreitol). Moreover, if they contain organic solvents and detergents these should be tested for interference with the DTP, NHD and OPT assays of the protocol. 2. The 4/5-untr-hs (from step 1b) is further split into two 1/4 and 3/4 vol portions. The 3/4 vol portion is not treated (designated 3/4-untr-hs), while to the 1/4 vol portion 1/10 vol 10 mM CumOOH and 1/10 vol 300 units GPx ml  1 are added and incubated at RT for 15 min (designated 1/4-GPx-hs). NOTE: In this step, GPx will oxidize GSH specifically to GSSG. However, to ensure complete oxidation of GSH in the 1/4 vol fraction, at least three dilutions of it (with 10X diluted 500PAE buffer, pH 7.5) should be CumOOH/GPx-treated and give the same total GSSG (¼ GSSG þGSSGGSH) concentration. 3. 1/5-NEM-hs, 1/4-GPx-hs, and 3/4-untr-hs are subjected to protein precipitation by DOC-TCA (DT) [58,59]. Specifically, per 1 ml 1/5-NEM-hs, 1/4-GPx-hs, and 3/4-untr-hs 20 μl 1% DOC (final 0.02%) are added and incubated for 10 min at RT, followed by addition of 115 μl 100% TCA (final 10%), incubation for 15 min in an ice-water bath, protein precipitation by centrifugation at 16,000g for 5 min at 4 1C, and collection of the corresponding protein free supernatants (pfsup) 1/5-DT-NEM-pfsup, 1/4-DTGPx-pfsup, and 3/4-DT-untr-pfsup. The resulting corresponding protein pellets (pp), designated 1/4-DT-GPx-pp and 3/4-DT-untrpp, are also collected (to be used in step 4). NOTE: The minimum protein concentration with Z90% efficiency by DOC-TCA precipitation is 2 mg ml-1 (for BSA [58,59]). Then, the 1/5-DT-NEM-pfsup, 1/4-DT-GPx-pfsup, and 3/4-DTuntr-pfsup are washed 3  with equal volume diethyl ether (DE) by vortexing, followed by centrifugation at 16,000g for 5 min at 4 1C (and discarding of the upper DE phase). Then, the three washed bottom aqueous phases (aqph), designated 1/5-DT-NEM-DE-aqph, 1/4-DT-GPx-DE-aqph, and 3/4DT-untr-DE-aqph, are collected (to be used in subsequent steps) and their exact volume is measured for normalizing the quantification of their nonprotein thiols per supernatant protein (before the DT-DE treatments). NOTE: DE-washing of 1/5-DT-NEM-pfsup, 1/4-DT-GPx-pfsup, and 3/4-DT-untr-pfsup removes DOC and TCA (thus facilitating their pH adjustment) and also any remains of MetOH and CumOOH from the 1/4-DT-GPx-pfsup, all of which may interfere in the assays applied on these aqueous phases in subsequent steps. PAUSE POINT: Although sample immediately is recommended, DE-washed, protein free supernatants can be stored at -80 1C for analyzing them after maximum 2 days. 4. The 1/4-DT-GPx-pp and 3/4-DT-untr-pp (from step 3) are washed once with 0.5 ml cold 100% acetone (AC) by vortexing, followed by centrifugation at 16,000g for 5 min at 4 1C, and airdrying of pellet remnant AC for 10 min at RT. PAUSE POINT: Although it is generally recommended to analyze samples immediately, AC-washed protein pellets

can be stored for maximum 1 week at -80 1C. The resulting AC-washed protein pellets, designated 1/4-DTGPx-AC-pp and 3/4-DT-untr-AC-pp, are treated as follows: (a) The 3/4-DT-untr-AC-pp is solubilized in minimum 6 ml 12 M formamide, centrifuged at 16,000g for 5 min to collect the clear solubilizate (sol), designated 3/4-DT-untr-AC-ppsol (to be used in step 19), and its protein concentration is measured in order to express its TRS parameters per protein. (b) The 1/4-DOC-TCA-GPx-AC-pp is solubilized in 0.2 ml 6 M urea, and then adjusted to 2.5 M urea by appropriate dilution with ddH2O and centrifuged at 16,000g for 5 min. The resulting clear solubilizate (sol), designated 1/4-DTGPx-AC-ppsol, is collected (to be used in step 29) and its protein content is measured in order to express its TRS parameters per protein amount. Subprotocol for nonprotein thiols Fraction A [ ¼ NPSH ( ¼ GSH þ CSH þNPxSHNPSH)] 5. Mix two 900-μl portions (for sample blank, SB, and sample, S) 3/4-DT-untr-DE-aqph (obtained from step 3; further diluted if needed with 10TCA-DOC-50PAE-DE) with the reagents shown in Table 1: 6. Incubate reagent mixtures for 10 min at RT and measure the absorbance of the mixtures at 325 nm, using a photometric quartz microcuvette. Subtract from the S value the values of the SB and RB, and the resulting net absorbance is converted to -SH group (GSH equivalent) concentration from the Standard curve for -SH groups in fractions A, B, H, I, J (see “Reagent setup” section). Fraction B {¼ NPSH (¼ Fraction A)þ NPSHNPSSNP [ ¼ GSHGSSG (¼ 2GSSG)þ GSHGSSNP ( ¼ GSSNP) þNPxSHGSSNP þCSHCSSNP ( ¼ CSSNP)þ NPxSHCSSNP þNPxSHNPxSSNPx]}, for indirect NPSSNP determination 7. Mix 2700 μl 3/4-DT-untr-DE-aqph (obtained from step 3; further diluted if needed with 10TCA-DOC-50PAE-DE) with 150 μl 1 M phosphate buffer, pH 8.0, and 150 μl 50 mM TBP incubate for 30 min at 50 1C. As reagent blank, 2700 μl 10TCA-DOC-50PAEDE in place of the sample is used. To both mixtures add 3 ml butanol-cyclohexane (BC) (5:1 v/v), vortex, and centrifuge at 16,000g for 5 min to isolate the corresponding bottom aqueous phases, designated sample-BC-aqph and RB-CB-aqph. Mix two 900-μl portions (for SB and S) sample-BC-aqph and 900 ml RBBC-aqph with the reagents shown in Table 2: 8. Incubate the reagent mixtures for 10 min at RT and measure the absorbance of the mixtures at 325 nm. Subtract from the S value the values of the SB and RB, and the resulting net absorbance is converted to GSH concentration from the Standard curve for -SH groups in fractions A, B, H, I, J (see “Reagent setup” section). Fraction C ¼CSH 9. Mix two 2000-ml portions (for SB and S) 3/4-DT-untr-DE-aqph (obtained from step 3; further diluted, if needed, with 10TCADOC-50PAE-DE) with the reagents shown in Table 3: Table 1 Reagents (μl)

Reagent blank (RB) Sample blank (SB) Sample (S)

3/4-DT-untr-DE-aqph 10TCA-DOC-50PAE-DE 4.5 M acetic acid, pH 4.5 3 mM DTP 10% DMSO

– 900 50 50 –

900 – 50 – 50

900 – 50 50 –

K. Grintzalis et al. / Free Radical Biology and Medicine 74 (2014) 85–98

Table 2 Reagents (μl)

Table 4 Reagent blank (RB) Sample blank (SB) Sample (S)

Sample-BC-aqph – RB-BC-aqph 900 4.5 M acetic acid, pH 4.5 50 3 mM DTP 50 10% DMSO –

900 – 50 – 50

900 – 50 50 –

Table 3 Reagents (ml)

Reagent blank (RB)

Sample blank (SB)

Sample (S)

3/4-DT-untr-DE-aqph 10TCA-DOC-50PAE-DE 100% acetic acid 100 mM NHD Acetic acid-HCl solution

– 2000 2000 2000 –

2000 – 2000 – 2000

2000 – 2000 2000 –

10. Incubate the reagent mixtures for 10 min at 100 1C, let the tubes cool in an ice-water bath, and add 1 ml chloroform. Vortex the tubes and centrifuge them to separate the chloroform (bottom) phase. Measure the absorbance of the chloroform phase at 560. Subtract from the S value the values of the SB and RB, and the resulting net absorbance is converted to CSH concentration from the Standard curve for fractions C, D (see “Reagent setup” section). CRITICAL: Same as in Standard curve for fractions C, and D. Fraction D [ ¼ CSH (¼ Fraction C) þCSHCSSNP ( ¼ CSSNP)], for indirect CSSNP determination 11. Mix two 1800-ml portions (for SB and S) 3/4-DT-untr-DE-aqph (obtained from step 3; further diluted, if needed, with 10TCADOC-50PAE-DE) with the reagents shown in Table 4: 12. Incubate the reagent mixtures for 10 min at 100 1C, let the tubes cool in an ice-water bath, and add 1 ml chloroform. Vortex the tubes and centrifuge them to separate the chloroform (bottom) phase. Measure the absorbance at 560. Subtract from the S value the values of the SB and RB, and the resulting net absorbance is converted to CSH concentration from the Standard curve for fractions C, D (see “Reagent setup” section).

13.

14.

15.

16.

93

CRITICAL: Same as in Standard curve for fractions C, and D. Fraction E ¼GSSG Mix two 230-μl portions (for SB and S) 1/5-DT-NEM-DE-aqph (obtained from step 3; further diluted, if needed, with 10TCADOC-50PAEN-DE) with the reagents shown in Table 5: Incubate the reagent mixtures for 20 min at RT in the dark, and measure the fluorescence of the mixtures at ex/em 340/ 420 nm. Subtract from the S value the values of the SB and RB, and the resulting net absorbance is converted to GSSG concentration from the Standard curve for fractions E ( ¼ GSSG), F (see “Reagent setup” section). Fraction F [ ¼ GSSG (¼ Fraction E)þGSSGGSH ( ¼ 1/2GSH)], for indirect GSH determination Mix two 230-μl portions (for SB and S) 1/4-DT-GPx-DE-aqph (obtained from step 3; further diluted if needed with 10TCADOC-50PAE-DE) with the reagents shown in Table 6: Incubate the reagent mixtures for 20 min at RT in the dark, and measure the fluorescence of the mixtures at ex/em 340/ 420 nm. Subtract from the S value the values of the SB and RB, and the resulting net absorbance is converted to GSSG concentration from the Standard curve for fractions E ( ¼ GSSG), F (see “Reagent setup” section). Fraction G [¼ GSH þGSHGSSG (¼ 2GSSG ¼2Fraction E) þ GSHGSSNP (¼ GSSNP)], for indirect GSSNP determination

Reagents (ml)

Reagent blank (RB)

3/4-DT-untr-DE-aqph – 10TCA-DOC-50PAE-DE 1800 1 M phosphate buffer, pH 8.0 100 50 mM TCEP 100 Incubate for 30 min at 50 1C 100% acetic acid 2000 100 mM NHD 2000 Acetic acid-HCl solution –

Sample blank (SB)

Sample (S)

1800 – 100 100

1800 – 100 100

2000 – 2000

2000 2000 –

Table 5 Reagents (μl)

Reagent blank (RB)

Sample blank (SB)

Sample (S)

1/5-DT-NEM-DE-aqph 10TCA-DOC-50PAEN-DE 0.2 M NaOH 1 mg ml  1 OPT 100% MetOH

– 230 110 20 –

230 – 110 – 20

230 – 110 20 –

Table 6 Reagents (μl)

Reagent blank (RB)

Sample blank (SB)

Sample (S)

1/4-DT-GPx-DE-aqph 10TCA-DOC-50PAE-DE 0.2 M NaOH 1 mg ml  1 OPT 100% MetOH

– 230 110 20 –

230 – 110 – 20

230 – 110 20 –

Table 7 Reagents (μl)

Reagent blank (RB)

Sample blank (SB)

Sample (S)

3/4-DT-untr-DE-aqph 10TCA-DOC-50PAEN-DE 0.5 M phosphate buffer, pH 8.0 50 mM TCEP Incubate for 30 min at 50 1C 1 mg ml  1 OPT 100% MetOH

– 190 130

190 – 130

190 – 130

20

20

20

20 –

– 20

20 –

17. Mix two 190-μl portions (for SB and S) 3/4-DT-untr-DE-aqph (obtained from step 3; further diluted if needed with 10TCADOC-50PAE-DE) with the reagents shown in Table 7: 18. Incubate the reagent mixtures for 20 min at RT in the dark, and measure the fluorescence of the mixtures at ex/em 340/ 420 nm. Subtract from the S value the values of the SB and RB, and the resulting net absorbance is converted to GSH concentration from the Standard curve for fraction G (see “Reagent setup” section).

Subprotocol for protein thiols Fraction H¼PSH 19. Mix two 900-μl portions (for SB and S) 3/4-DOC-TCA-untr-ACppsol (obtained from step 4a; diluted appropriately, if needed, with 12 M formamide) with the reagents shown in Table 8:

94

K. Grintzalis et al. / Free Radical Biology and Medicine 74 (2014) 85–98

Table 8

Table 9

Reagents (μl)

Reagent blank (RB)

Sample blank (SB)

Sample (S)

Reagents (μl)

Reagent blank (RB)

Sample blank (SB)

Sample (S)

3/4-DOC-TCA-untr-ACppsol 12 M formamide 0.1 M acetic acid, pH 4.5 3 mM DTP 10% DMSO

–

900

900

–

900

900

900 50 50 –

– 50 – 50

– 50 50 –

3/4-2DT-untr-AC-ppsol-TBP-BCpfsup RB-formamide-TBP-BC-DT 0.1 M acetic acid, pH 4.5 3 mM DTP 10% DMSO

900 50 50 –

– 50 – 50

– 50 50 –

20. Incubate the reagent mixtures for 10 min at RT and measure the absorbance of the mixtures at 325 nm. Subtract from the S value the values of the SB and RB, and the resulting net absorbance is converted to GSH concentration from the Standard curve for -SH groups in fractions A, B, H, I, and J (see “Reagent setup” section).

Reduction of protein -SS- groups (in PSSP, PSSNP, PSSC, PSSG) 21. Mix 4.5 ml 3/4-DT-untr-AC-ppsol (obtained from step 4a; diluted appropriately using 12 M formamide) with 1.95 ml 1 M phosphate buffer, pH 8.0, and 0.3 ml 50 mM TBP and incubate for 30 min at 50 1C (designated 3/4-DT-untr-ACppsol-TBP). For the two reagent blanks (RB; one used in the subsequent step 22a and the other in the step 22b) use for both 4.5 ml 12 M formamide (in place of the 3/4-DOC-TCA-untrAC-ppsol) mixed with 1.95 ml 1 M phosphate buffer, pH 8.0, and 0.3 ml 50 mM TBP. The final volume of the reagent blank solution for both reagent blanks is 6.75 ml, and is designated RB-formamide-TBP. 22. The 3/4-DT-untr-AC-ppsol-TBP (prepared in step 21) is split in 2.7 and 3.6 ml portions, which are treated as follows: (a) To 2.7 ml 3/4-DT-untr-AC-ppsol-TBP portion and to 2.7 ml RB-formamide-TBP (both from step 21) add an equal volume butanol-cyclohexane (BC) (5:1 v/v), vortex and centrifuge at 16,000g for 5 min to isolate the bottom aqueous phase (free of TBP). To precipitate proteins in the resulting 2.7 ml BC-washed bottom aqueous phase, 60 ml 1% DOC (final 0.02%) is added and incubated for 10 min at RT, and then 300 μl 100% TCA (final 10%) is added and incubated for 15 min in an ice-water bath. Finally, proteins are precipitated by centrifugation at 16,000g for 5 min at 4 1C, and the protein free supernatant (pfsup) is collected (it contains NPSHPSSNP, which is the sum CSHPSSNP þGSHPSSNP þNPxSHPSSNP), its volume is measured, and it is designated 3/4-2DT-untr-AC-ppsol-TBP-BCpfsup. Similarly, the BC-washed RB-formamide-TBP is brought to 0.02% DOC and 10% TCA by addition of 60 ml 1% DOC and 300 μl 100% TCA, respectively, followed by centrifugation at 16,000g for 5 min at 4 1C, and the supernatant is designated RB-formamide-TBP-BC-DT. Two 900-ml portions 3/4-2DT-untr-AC-ppsol-TBP-BC-pfsup, one for sample blank (SB) and one for sample (S), and 900 ml from RBformamide-TBP-BC-DT (for RB; the remaining can be used for making two more RB repeats) will be used in step 23. (b) The 3.6 ml 3/4-DT-untr-AC-ppsol-TBP portion (from step 21) is mixed with 80 ml 1% DOC (final 0.02%), incubated for 10 min at RT, then mixed with 400 μl 100% TCA (final 10%), and incubated for 15 min in an ice-water bath. Then, proteins are precipitated by centrifugation at 16,000g for 5 min at 4 1C, and the protein pellet (containing all -SSgroups reduced) is collected (for use in step 25) and designated 3/4-2DT-untr-AC-ppsol-TBP-pp. The resulting protein free supernatant (pfsup), designated 3/4-2DT-

Table 10 Reagents (μl)

Reagent blank (RB)

Sample blank (SB)

Sample (S)

3/4-2DT-untr-AC-TBP2ppsol 12 M formamide 0.1 M acetic acid, pH 4.5 3 mM DTP 10% DMSO

–

900

900

900 50 50 –

– 50 – 50

– 50 50 –

untr-AC-ppsol-TBP-pfsup, is also collected (it contains NPSHPSSNP, which is the sum CSHPSSNP þ GSHPSSNP þ NPxSHPSSNP) and its volume is measured. Similarly, a 3.6 ml portion of RB-formamide-TBP (from step 21) is brought to 0.02% DOC and 10% TCA by addition of 80 ml 1% DOC and 400 μl 100% TCA, respectively, followed by centrifugation at 16,000g for 5 min at 4 1C, and the supernatant is designated RB-formamide-TBP-DT. Two 2000-ml portions from 3/4-2DT-untr-AC-ppsol-TBP-pfsup, one for sample blank (SB) and one for sample (S), and 2000 ml from RB-formamide-TBP-DT (for RB; the remaining can be used for making one more RB repeat) will be used in step 27.

23. 24.

25.

26.

27.

Fraction I [ ¼ NPSHPSSNP (¼ CSHPSSNP þGSHPSSNP þNPxSHPSSNP)], for PSSNP determination Mix the 900 ml SB, S and RB fractions from step 22a with the reagents shown in Table 9: Incubate the reagent mixtures for 10 min at RT and measure the absorbance of the mixtures at 325 nm. Subtract from the S value the values of the SB and RB, and the resulting net absorbance is converted to GSH concentration from the Standard curve for -SH groups in fractions A, B, H, I, and J (see “Reagent setup” section). Fraction J [¼ totPSH¼PSH (¼ Fraction H) þ2PSHPSSP þPSHPSSNP (¼ Fraction I)], for PSSP determination Solubilize the 3/4-2DT-untr-AC-ppsol-TBP-pp (obtained in step 22b) in a minimum volume (up to 1.8 ml) 12 M formamide, centrifuge at 16,000g for 5 min to collect the clear protein solubilizate (designated 3/4-2DT-untr-AC-TBP-2ppsol) and measure its protein concentration in order to express the result of the PSH sum per protein amount. Mix two 900-μl portions (for SB and S) 3/4-2DT-untr-AC-TBP-2ppsol (diluted if needed with 12 M formamide) with the reagents shown in Table 10: Incubate the reagent mixtures for 10 min at RT and measure the absorbance of the mixtures at 325 nm. Subtract from the S value the values of the SB and RB, and the resulting net absorbance is converted to GSH concentration from the Standard curve for -SH groups in fractions A, B, H, I, and J (see “Reagent setup” section). Fraction K [¼ CSHPSSNP ( ¼ PSSC)] Mix the 2000 ml SB, S and RB fractions from step 22b with the reagents shown in Table 11:

K. Grintzalis et al. / Free Radical Biology and Medicine 74 (2014) 85–98

Table 11 Reagents (ml)

Reagent blank (RB)

Sample blank (SB)

Sample (S)

3/4-2DT-untr-AC-ppsol-TBPpfsup RB-formamide-TBP-DT 100% Acetic acid 100 mM NHD Acetic acid-HCl solution

–

2000

2000

2000 2000 2000 –

– 2000 – 2000

– 2000 2000 –

95

31. Incubate the reagent mixtures for 20 min at RT in the dark, and measure the fluorescence of the mixtures at ex/em 340/ 420 nm using a quartz microcuvette (with the spectrofluorometer in use set at low sensitivity). Subtract from the sample value the values of the sample and reagent blanks and the resulting net absorbance is converted to GSH concentration from the Standard curve for fraction L ( ¼PSSG) (see “Reagent setup” section). Estimation of the TRS parameters

Table 12 Reagents (μl)

Reagent blank (RB)

Sample blank (SB)

Sample (S)

1/4-2DT-GPx-AC-ppsol-TCEP-DEpfsup RB-urea-TCEP-DT-DE 0.5 M phosphate buffer, pH 8.0 1 mg ml  1 OPT 100% MetOH

–

200

200

200 130 20 –

– 130 – 20

– 130 20 –

28. Incubate the reagent mixtures for 10 min at 100 1C, let the tubes cool in an ice-water bath and add 1 ml chloroform. Vortex the tubes and centrifuge them to separate the chloroform phase (bottom). Remove the interference from formamide by isolating the chloroform phase after vortexing with 4 ml acetic acid, followed by centrifugation at 16,000g for 5 min, and measure absorbance at 560 nm. Subtract from the S value the values of the SB and RB, and the resulting net absorbance is converted to CSH concentration from the Standard curve for fraction K (¼ PSSC) (see “Reagent setup” section). CRITICAL: Same as in Standard curve for fractions C, and D. Fraction L [ ¼ GSHPSSNP ( ¼ PSSG)] 29. One ml 1/4-DT-GPx-AC-ppsol (obtained from step 4b; diluted appropriately with 2.5 M urea) is mixed with 20 μl 0.5 M phosphate buffer, pH 8.0, and reduced with 50 μl 50 mM TCEP by incubation for 30 min at 50 1C. As reagent blank (RB), 1 ml 2.5 M urea (in place of 1/4-DT-GPx-AC-ppsol) is mixed with 20 μl 0.5 M phosphate buffer, pH 8.0, and 50 μl 50 mM TCEP, and designated RB-urea-TCEP. 30. In the above 1.07 ml of the TCEP-reduced 1/4-DT-GPx-AC-ppsol proteins is precipitated by addition of 24 ml 1% DOC (final 0.02%) and incubation for 10 min at RT, then addition of 120 μl 100% TCA (final 10%) and incubation for 15 min in an ice-water bath, and finally centrifugation at 16,000g for 5 min at 4 1C. The protein free supernatant (pfsup) is washed 3  with equal volume DE (by vortexing and centrifugation at 16,000g for 5 min at 4 1C, and discarding the upper DE phase). The resulting DE-washed supernatant is collected (and designated 1/4-2DT-GPx-AC-ppsol-TCEP-DE-pfsup) and its volume is measured (it contains NPSHPSSNP, which is the sum CSHPSSNP þGSHPSSNP þNPxSHPSSNP). Similarly, the 1.07 ml RBurea-TCEP portion (for RB from step 29) is brought to 0.02% DOC and 10% TCA by addition of 24 μl 1% DOC and 120 μl 100% TCA, respectively, followed by centrifugation at 16,000g for 5 min at 4 1C. The resulting clear supernatant is washed 3  with equal volume DE (by vortexing and centrifugation at 16,000g for 5 min at 4 1C, and discarding the upper DE phase), and the resulting DE-washed supernatant is designated RBurea-TCEP-DT-DE. Two 200-ml portions 1/4-2DT-GPx-AC-ppsolTCEP-DE-pfsup, one for sample blank (SB) and one for sample (S), and 200 ml RB-urea-TCEP-DT-DE (for RB), are mixed with the reagents as shown in Table 12:

The following equations were derived from the experimental values for Fractions A to L as determined in the protocol procedure (for their Excel-adapted automated estimation see Supplementary Material). Nonprotein parameters 1. NPSH¼ A ( ¼ GSH þCSH þNPxSHNPSH) 2. NPSSNP¼½(B - A), deduced from B ¼A þ NPSHNPSSNP [ ¼ GSHGSSG ( ¼ 2GSSG) þGSHGSSNP (¼ GSSNP) þNPxSHGSSNP þCSHCSSNP (¼ CSSNP) þ NPxSHCSSNP þ NPxSHNPxSSNPx] ¼ A þ2NPSSNP 3. CSH¼ C 4. CSSNP ¼D - C, deduced from D ¼CSH (¼ C) þCSHCSSNP (¼ CSSNP) 5. GSSG ¼E 6. GSH¼2(F - E), deduced from F¼GSSG (¼ E)þ GSSGGSH (¼ 1/ 2GSH) 7. GSSNP¼ G-2F, deduced from G ¼ GSH [ ¼ 2(F - E)] þGSHGSSG ( ¼ 2GSSG ¼2E)þ GSHGSSNP (¼ GSSNP) 8. NPxSHNPSH ¼A - C - 2(F - E), deduced from equalities (1), (3), (6) 9. NPxSHNPSSNP ¼ B þ C - A - D - G - 2(E þF) (defined as the sum NPxSHGSSNP þNPxSHCSSNP þNPxSHNPxSSNPx). It is deduced from B¼ A þNPSHNPSSNP [ ¼ GSHGSSG (¼ 2GSSG) þGSHGSSNP ( ¼ GSSNP) þNPxSHGSSNP þCSHCSSNP (¼ CSSNP) þ NPxSHCSSNP þ NPxSHNPxSSNPx], and from equalities (1), (4), (5), (7) Protein parameters 10. PSH¼H 11. PSSNP ¼I ( ¼ NPSHPSSNP ¼CSHPSSNP þGSHPSSNP þNPxSHPSSNP) 12. PSSP ¼½(J - I - H), deduced from J¼ totPSH¼PSH (¼ H)þ 2PSHPSSP þPSHPSSNP ( ¼ I) 13. PSSC¼ K (¼ CSHPSSNP) 14. PSSG ¼L ( ¼ GSHPSSNP) 15. NPxSHPSSNP ¼I - K - L, deduced from equalities (11), (13), (14)

Algebraic preconditions For the validity of the above experimental fraction values of the TRS components, and of those derived as algebraic functions of experimentally determined fraction values, the following not strict inequalities should hold true: NPSH ZGSHþ CSH NPSSNP ZGSSG 2NPSSNPZ CSSNP þGSSNP þGSSG PSSNPZPSSG þPSSC Statistical treatment of raw data TRS parameters were calculated in 20 plasma samples (10/10 males/females, covering a 50-60 year age span) as mean

96

K. Grintzalis et al. / Free Radical Biology and Medicine 74 (2014) 85–98

values 7 SE of the mean. Parameters GSH, NPSSNP, CSSNP, GSSNP, PSSP, NPxSHNPSSNP, NPxSHNPSH, and NPxSHPSSNP are derived as mean values determined by the algebraic functions of the mean values of the involved fractions (see section “Estimation of the TRS parameters”). These functions were used in conjunction with the individual SE of the involved fractions to estimate the SE of each of these parameters by the following general SE function adapted from elsewhere [50]: SETRS ¼ 7 √{[(∂TRS/∂A)SEA]2 þ[(∂TRS/∂B)SEB]2 þ[(∂TRS/∂C)SEC]2 þ[(∂TRS/∂D)SED]2 þ…} The SE equations of these TRS parameters are the following (for their Excel-adapted automated estimation see Supplementary Material): SENPSSNP ¼ 7 √(SE2A þSE2B) SECSSNP ¼ 7 √(SE2D þSE2C) SEGSSNP ¼ 7 √(SE2G þSE2F ) SEGSH ¼ 7 √(SE2F þSE2E) SENPxSHNPSH ¼ 7 √(SE2A þSE2C þSE2F þ SE2E SENPxSHNPSSNP ¼ 7 √(SE2B þSE2C þ SE2A þSE2D þ SE2G þSE2E þ SE2F SEPSSP ¼ 7 √(SE2J þSE2I þSE2H) SENPxSHPSSNP ¼ 7 √(SE2I þSE2K þSE2L For determining the minimum statistical variation of the TRS parameters quantified by the present protocol, human plasma samples were analyzed the same day of blood collection. The TRS parameters of each plasma sample were derived from the analysis of at least three successive sample dilutions, and their mean value and SE were calculated. The TRS parameters were also statistically analyzed for precision, assessed during a single analytical run (within-run, within-day precision or repeatability), and with time (between-run, between day repeatability, also named intermediate precision). For calculating the minimum statistical variation of TRS parameters quantified by the present protocol, at least three successive dilutions of human plasma samples were analyzed the same day of blood collection, and their mean value was calculated. The within-day % coefficient variation is calculated as Standard Deviation (SD)x100/mean, and the variance of intermediate precision (σ2total) is defined as the sum of between day variance (σ2between) that is associated with the day-to-day variation, and the variance of repeatability (σ2within). The within-day coefficient variation of the TRS protocol for all its parameters is 4 4% and the between day repeatability is 4 5.5%.

Protocol timing and planning Timing: The timing listed in the following subheadings of the protocol procedure represents the minimum time required for processing 10 samples as follows: Sample treatment (steps 1–4): total 1 h. Subprotocol for nonprotein thiols (steps 5–18): treatment of fractions A to G, total 4 h; A, 30 min; B, 40 min; C, 30 min; D, 50 min; E, 30 min; F, 30 min; G, 70 min. Subprotocol for protein thiols (steps 19–31): treatment of fractions H to L, total 5 h; H, 30 min; I, 45 min; J, 45 min; K, 90 min; L, 120 min. Planning: Because of the multiparametric analysis offered by the protocol, handling of many samples (e.g., 10) in the same day may be difficult. In such case, the protocol can be executed with all required reagents prepared for treating all samples up to step 4, by making the pose points suggested in protocol steps 3 and 4. However, it is recommended to extend the sample treatment up to step 12, for the determination of CSH and CSSNP. Nonetheless, P-thiols in the AC-washed protein pellets from step 4 are normally stable up to 1 week. Then, the Subprotocol for protein thiols can be performed next. Here the experimenter could start with the ureasolubilized protein pellet from step 4b, e.g., for PSSG determination; P-thiols in the formamide-solubilized pellet from step 4a can be analyzed next.

Calculations and expected results This protocol assesses quantitatively 15 parameters of TRS in organisms by three different assays applied on sample fractions A to L. The first assay uses the DTP reagent to quantify NPSH, NPxSHNPSH, and PSH, as well as their disulfides (NPSSNP, NPxSHNPSSNP, PSSNP, NPxSHPSSNP, PSSP); the second assay specifically measures CSH, CSSNP, and PSSC by the NHD reagent; the third assay determines GSH, GSSG, GSSNP, and PSSG by the OPT reagent. The sensitivity of the protocol for NPSH, PSH, NPSSNP, PSSP, PSSNP, CSH, CSSNP, PSSC, NPxSHNPSSNP, and NPxSHNPSH is 1 nmol -SH/CSH, for GSSNP 0.2 nmol, for GSH and GSSG 0.4 nmol, and for PSSG 0.6 nmol. When spiking the 1/5- and 4/5-blood plasma fractions, in steps 1(a) and 1(b), respectively (or in the same fractions of any nonfluid sample before homogenization), with known amounts of GSH, GSSG, and CSH, the achieved recoveries were near 90% (data not shown). The protocol is designed to be modular and adjustable to the experimenter’s research priorities on some or all TRS parameters. It may be performed separately for nonprotein and protein thiols (fractions A to G and fractions H to L, respectively), or it can be focused on individual TRS parameters (Table 13). Given the existence of lipoic acid and other lipophilic thiols in various experimental samples, these lipophilic thiols are not counted by the protocol for the following reasons: The protocol separates the sample homogenate in an aqueous and a protein fraction (for determining the nonprotein and protein thiols, respectively), by a protein precipitation step that includes homogenate pretreatment with deoxycholate (DOC) followed by TCA precipitation. Then, there is a diethyl ether-wash step (3) for the protein-free supernatant, and a cold acetone wash step (4) for the protein pellet. Thus, lipoic acid or any other lipophilic thiols are removed from the supernatant by diethyl ether (for lipoic acid solubility in diethyl ether see [66]), while any remnants of lipophilic residues as well the TCA-precipitated DOC and any TCA remnants in the pellet are also removed by the acetone wash. Another factor to consider is the applicability of the protocol in organisms such as plants, fungi, and yeast, which can synthesize phytochelatins for heavy metal detoxification [67]. These are glutathione (GSH) related peptides, which, because of their relatively small molecular weight (MW), could be measured by the protocol as GSH if present in the aqueous fraction of the sample. Nonetheless, phytochelatins are also isolated by protein precipitation in 6:1 vol cold ethanol (e.g., from the basidiomycete fungus Boletus edulis) [68]. However, the present protocol employs a far

Table 13 Modularity of TRS protocol. TRS parameter Nonprotein parameters GSH GSSG GSSNP CSH CSSNP NPSH NPSSNP NPxSHNPSH NPxSHNPSSNP Protein parameters PSSG PSSC PSH PSSNP PSSP NPxSHPSSNP

Fractions required

E, F E G, F C C, D A A, B A, C, F, E A, B, C, D, E, F, G L K H I H, I, J I, J, K

K. Grintzalis et al. / Free Radical Biology and Medicine 74 (2014) 85–98

more effective protein precipitation method (a mixture of deoxycholate and TCA), which achieves Z 90% recovery of small peptides such as ribonuclease A, even in the presence of 1 M guanidine-HCl [58]. Moreover, the protocol uses a diethyl ether step to wash the aqueous fraction of the sample, which renders peptides even more insoluble. Although we cannot exclude the possibility of very small MW peptides to escape precipitation, we expect the majority of phytochelatin peptides to end up in the protein pellet fraction, where they will be quantified as PSH or PSSP by the DTP reagent of the protocol. The protocol was applied to human plasma, results were compared with data from alternative methodologies reported in the literature, and a good correspondence was attained (Table 14). However, no comparison data were available for the TRS parameters GSSNP, CSSNP, NPSH, NPSSNP, PSSP, PSSNP, NPxSHNPSSNP, NPxSHPSSNP, and NPxSHNPSH. This is indicative of the partial evaluation of TRS by the current methodologies. We limited our measurements to normal samples to display the applicability of the present assay and also set approximate normal levels mainly for the new TRS parameters introduced by the protocol. Before testing pathological samples, extensive studies are needed to optimize normal levels by taking also into account sample variability factors such as degree of hemolysis. However, care was taken in this study not to use hemolyzed samples, which are defined as having free hemoglobin in serum or plasma below 50 mg/dl [69].

97

Caveats A near zero net absorbance or fluorescence value of a certain thiol parameter suggests that its concentration in the sample is below the assay sensitivity limit. Increasing the initial amount of sample, and/or keeping its initial homogenate at a minimum volume could address this. If the absorbance or fluorescence value of a thiol parameter falls out of the upper limit of the linear range of the corresponding thiol parameter standard curve, this means that its concentration in the sample is very high. In this case, the sample must be diluted appropriately. High statistical variation in a thiol parameter value among replicate samples suggests high variation in its concentration in the experimental organism. Increasing the number of different sample replicates can solve this problem. If the algebraic preconditions defined for this protocol (in the subsection “Estimation of the TRS parameters”) are not met, this is likely due to experimental errors in its execution. The suggested solution to this problem is repetition of the TRS parameter measurements with new and fresh samples. For example, NPSH must be greater or equal to the sum of GSH and CSH because, if detected, their values are included in the value of NPSH. Similarly, NPSH should be greater or equal to GSH or CSH alone, as long as CSH or GSH are near zero, respectively.

Conflict of Interest The authors declare that they have no patents or financial interests in the method or instruments.

Table 14 TRS parameters of human plasma. Parameters

(nmol mg-1)

(μΜ)

Nonprotein parameters GSH 0.023 7 0.004

3.17 0.26

GSSG

0.0137 0.002

1.8 7 0.21

GSSNP CSH

0.0417 0.006 0.08 7 0.01

4.3 7 0.62 11.17 0.9

1.55 7 0.12 0.197 0.018 1.917 0.15 0.09 7 0.02 2.43 7 0.35

2157 14 25.17 1.5 255 7 34 10.9 7 1.8 3167 48

5.17 0.6 987 11 0.0197 0.002

6807 75 128007 1500 2.6 7 0.28

PSSC

1.157 0.22

1507 19

PSSNP NPxSHPSSNP

5.93 7 0.65 4.767 0.69

795 7 88 6427 90

CSSNP NPSH NPSSNP NPxSHNPSH NPxSHNPSSNP Protein parameters PSH PSSP PSSG

From indicative bibliography (mM)

3.5 7 0.3 [43] 2–5 [70] 1.4–6 [44] 1.5 7 0.5 [38] 2–3 [46,55] 3.73 [39] 2 (3% RSD) [37] 1.23 7 0.44 [46] 2.2 (5.1% RSD) [37] Not available (NA) 11.7 7 0.95 [43] 8.3–10.7 [70] 5.9–12.6 [44] 9.96 7 1.29 [46] NA NA NA NA NA 400–600 [70] NA 2.5 7 0.76 [43] 0.7–1.9 [70] 5.3–11 [44] 2.317 1.18 [46] 142.17 8.4 [43] 145–176 [70] 178–227 [44] 156.82 728.44 [46] NA NA

Note: TRS parameters were calculated in plasma samples from 10 males and 10 females. All values are mean7 SE after checking for equality of error variances between values of males and females (Levene's test, SPSS Inc.) with two-way ANOVA analysis of variance (to identify significant differences between values), and with the parametric post hoc multiple comparison test (Bonferroni test, P o 0.05). There were no statistical differences in the TRS parameters between genders.

Acknowledgments This work was financially supported by the Greek Ministry of Education. We thank Prof. Dr. Helmut Sies (Institute for Biochemistry and Molecular Biology I, Heinrich-Heine-University Düsseldorf) for his very important and thoughtful comments.

Appendix A.

Supplementary Information

Supplementary data associated with this article can be found in the online version at http://dx.doi.org/10.1016/j.freeradbiomed. 2014.06.024. References [1] Sies, H. Glutathione and its role in cellular functions. Free Radic. Biol. Med. 27:916–921; 1999. [2] Halliwell, B.; Gutteridge, C. M. J. Free radicals in biology and medicine. Oxford: Oxford University Press; 1999. [3] Dickinson, D. A.; Forman, H. J. Cellular glutathione and thiols metabolism. Biochem. Pharmacol. 64:1019–1026; 2002. [4] Wang, W.; Ballatori, N. Endogenous glutathione conjugates: occurrence and biological functions. Pharmacol. Rev. 50:335–356; 1998. [5] Giles, I. G.; Tasker, M. K.; Jacob, C. Hypothesis: the role of reactive sulfur species in oxidative stress. Free Radic. Biol. Med. 31:1279–1283; 2001. [6] Li, J.; Huang, L. F.; Huang, K. -P. Glutathiolation of proteins by glutathione disulfide S-oxide derived from S-nitrosoglutathione. J. Biol. Chem. 276:3098–3105; 2001. [7] Crane, D.; Häussinger, D.; Sies, H. Rise of coenzyme A-glutathione mixed disulfide during hydroperoxide metabolism in perfused rat liver. Eur. J. Biochem. 127:575–578; 1982. [8] Akerboom, M. P. T.; Sies, H. Assay of glutathione, glutathione disulfide and glutathione mixed disulfides in biological samples. In: Colowick, S. P., Kaplan, O. N., editors. Methods in enzymology. New York: Academic Press; 1981. p. 373–382. [9] Corti, A.; Paolicchi, A.; Franzini, M.; Dominici, S.; Casini, A. F.; Pompella, A. The S-thiolating activity of membrane gamma-glutamyltransferase: formation of cysteinyl-glycine mixed disulfides with cellular proteins and in the cell microenvironment. Antioxid. Redox Signal. 7:911–918; 2005.

98

K. Grintzalis et al. / Free Radical Biology and Medicine 74 (2014) 85–98

[10] Gilbert, F. H. Thiol/disulfide exchange equilibria and disulfide bond stability. In: Packer, L., editor. Methods in enzymology. New York: Academic Press; 1995. p. 8–28. [11] Ratbbun, B. W. Glutathione metabolism in the mammalian ocular lense. In: Vina, J., editor. Glutathione: Metabolism and physiological functions. Boca Raton: CRC Press; 1990. p. 193–206. [12] Miquel, J.; Weber, H. Aging and increased oxidation of the sulfur pool. In: Vina, J., editor. Glutathione: Metabolism and physiological functions. Boca Raton: CRC Press; 1990. p. 187–192. [13] Levitt, J. Responses of plants to environmental stresses. New York: Academic Press; 1972. [14] Trivedi, M. V.; Laurence, J. S.; Siahaan, T. J. The role of thiols and disulfides on protein stability. Curr. Protein Pept. Sci. 10:614–625; 2009. [15] Gupta, V.; Carroll, K. S. Sulfenic acid chemistry, detection and cellular lifetime. Biochim. Biophys. Acta 1840:847–875; 2014. [16] Alvarez, B.; Carballal, S.; Turell, L.; Radi, R. Formation and reactions of sulfenic acid in human serum albumin. Methods Enzymol. 473:117–136; 2010. [17] Requejo, R.; Hurd, T. R.; Costa, N. J.; Murphy, M. P. Cysteine residues exposed on protein surfaces are the dominant intramitochondrial thiol and may protect against oxidative damage. FEBS J. 277:1465–1480; 2010. [18] Carballal, S.; Alvarez, B.; Turell, L.; Botti, H.; A., F. B.; Radi, R. Sulfenic acid in human serum albumin. Amino Acids 32:543–551; 2007. [19] Eaton, P. Protein thiol oxidation in health and disease: techniques for measuring disulfides and related modifications in complex protein mixtures. Free Radic. Biol. Med. 40:1889–1899; 2006. [20] van der Vliet, A.; Cross, E. C.; Halliwell, B.; O'Neill, A. C. Plasma protein sulfhydryl oxidation: effect of low molecular weight thiols. In: Packer, L., editor. Methods in enzymology. New York: Academic Press; 1995. p. 448–455. [21] Cotgreave, A. I.; Weis, M.; Atzori, L.; Moldéus, P. Glutathione and protein function. In: Vina, J., editor. Glutathione: metabolism and physiological functions. Boca Raton: CRC Press; 1990. p. 155–175. [22] Brigelius-Flohé, R. Tissue-specific functions of individual glutathione peroxidases. Free Radic. Biol. Med. 27:951–965; 1999. [23] Finkel, T. Signal transduction by reactive oxygen species. J. Cell Biol. 194:7–15; 2011. [24] Ewing, F. J.; Janero, R. D. Specific S-nitrosothiol (thionitrite) quantification as solution nitrite after vanadium(III) reduction and ozone-chemiluminescent detection. Free Radic. Biol. Med. 25:621–628; 1998. [25] Täger, M.; Piecyk, A.; Köhnlein, T.; Thiel, U.; Ansorge, S.; Welte, T. Evidence of a defective thiol status of alveolar macrophages from COPD patients and smokers. Free Radic. Biol. Med. 29:1160–1165; 2000. [26] David, L. L.; Shearer, R. T. State of sulfhydryl in selenite cataract. Toxicol. Appl. Pharmacol. 74:109–115; 1984. [27] Tietze, F. Enzymic method for quantitative determination of nanogram amounts of total and oxidized glutathione: applications to mammalian blood and other tissues. Anal. Biochem. 27:502–522; 1969. [28] Ellman, L. G. Tissue sulfhydryl groups. Arch. Biochem. Biophys. 82:70–77; 1959. [29] Redegeld, F. A. M.; Koster, A. S.; van Bennekom, W. P. Determination of tissue glutathione. In: Vina, J., editor. Glutathione: metabolism and physiological functions. Boca Raton: CRC Press; 1990. p. 11–20. [30] Giustarini, D.; Dalle-Donne, I.; Milzani, A.; Fanti, P.; Rossi, R. Analysis of GSH and GSSG after derivatization with N-ethylmaleimide. Nat. Protoc. 8:1660–1669; 2013. [31] Rahman, I.; Kode, A.; Biswas, S. K. Assay for quantitative determination of glutathione and glutathione disulfide levels using enzymatic recycling method. Nat. Protoc. 1:3159–3165; 2006. [32] Hissin, P. J.; Hilf, R. A fluorometric method for determination of oxidized and reduced glutathione in tissues. Anal. Biochem. 74:214–226; 1976. [33] Puri, R. N.; Roskoski Jr R. Reaction of low molecular weight aminothiols with o-phthalaldehyde. Anal. Biochem. 173:26–32; 1988. [34] Lenton, K. J.; Therriault, H.; Wagner, J. R. Analysis of glutathione and glutathione disulfide in whole cells and mitochondria by postcolumn derivatization high-performance liquid chromatography with ortho-phthalaldehyde. Anal. Biochem. 274:125–130; 1999. [35] Harrap, R. K.; Jackson, C. R.; Riches, G. P.; Smith, A. C.; Hill, T. B. The occurrence of protein-bound mixed disulfides in rat tissues. Biochim. Biophys. Acta (Prot. Struct.) 310:104–110; 1973. [36] Ball, R. C. Estimation and identification of thiols in rat spleen after cysteine or glutathione treatment: relevance to protection against nitrogen mustards. Biochem. Pharmacol. 15:809–816; 1966. [37] Salazar, J. F.; Schorr, H.; Herrmann, W.; Herbeth, B.; Siest, G.; Leroy, P. Measurement of thiols in human plasma using liquid chromatography with precolumn derivatization and fluorescence detection. J. Chromatogr. Sci. 37:469–476; 1999. [38] Ferin, R.; Pavao, M. L.; Baptista, J. Methodology for a rapid and simultaneous determination of total cysteine, homocysteine, cysteinylglycine and glutathione in plasma by isocratic RP-HPLC. J. Chromatogr. B Analyt. Technol. Biomed. Life Sci. 911:15–20; 2012. [39] Kusmierek, K.; Bald, E. Reversed-phase liquid chromatography method for the determination of total plasma thiols after derivatization with 1-benzyl-2chloropyridinium bromide. Biomed. Chromatogr. 23:770–775; 2009. [40] Riener, C. K.; Kada, G.; Gruber, H. J. Quick measurement of protein sulfhydryls with Ellman's reagent and with 4,40 -dithiodipyridine. Anal. Bioanal. Chem. 373:266–276; 2002.

[41] Giustarini, D.; Dalle-Donne, I.; Lorenzini, S.; Selvi, E.; Colombo, G.; Milzani, A.; Fanti, P.; Rossi, R. Protein thiolation index (PTI) as a biomarker of oxidative stress. Free Radic. Biol. Med. 53:907–915; 2012. [42] Hoff, S.; Larsen, F. H.; Andersen, M. L.; Lund, M. N. Quantification of protein thiols using ThioGlo 1 fluorescent derivatives and HPLC separation. Analyst 138:2096–2103; 2013. [43] Carru, C.; Deiana, L.; Sotgia, S.; Pes, G. M.; Zinellu, A. Plasma thiols redox status by laser-induced fluorescence capillary electrophoresis. Electrophoresis 25:882–889; 2004. [44] Bald, E.; Chwatko, G.; Glowacki, R.; Kusmierek, K. Analysis of plasma thiols by high-performance liquid chromatography with ultraviolet detection. J. Chromatogr. A 1032:109–115; 2004. [45] Hansen, R. E.; Ostergaard, H.; Norgaard, P.; Winther, J. R. Quantification of protein thiols and dithiols in the picomolar range using sodium borohydride and 4,40 -dithiodipyridine. Anal. Biochem. 363:77–82; 2007. [46] Rossi, R.; Giustarini, D.; Milzani, A.; Dalle-Donne, I. Cysteinylation and homocysteinylation of plasma protein thiols during ageing of healthy human beings. J. Cell. Mol. Med. 13:3131–3140; 2009. [47] Chin, C. C. Q.; Wold, F. The use of tributylphosphine and 4-(aminosulfonyl)-7fluoro-2,1,3-benzoxadiazole in the study of protein sulfhydryls and disulfides. Anal. Biochem. 214:128–134; 1993. [48] Kirley, T. L. Reduction and fluorescent labelling of cyst(e)ine-containing proteins for subsequent structural analyses. Anal. Biochem. 180:231–236; 1989. [49] Krijt, J.; Vacková, M.; Kozich, V. Measurement of homocysteine and other aminothiols in plasma: advantages of using tris(2-carboxyethyl)phosphine as reductant compared with tri-n-butylphosphine. Clin. Chem. 47:1821–1828; 2001. [50] Patsoukis, N.; Georgiou, C. D. Determination of the thiol redox state of organisms: new oxidative stress indicators. Anal. Bioanal. Chem. 378:1783–1792; 2004. [51] Patsoukis, N.; Georgiou, C. D. Fluorometric determination of thiol redox state. Anal. Bioanal. Chem. 383:923–929; 2005. [52] Owusu-Apenten, R. Colorimetric analysis of protein sulfhydyl groups in milk: applications and processing effects. Crit. Rev. Food Sci. Nutr. 45:1–23; 2005. [53] Flohé, L. The fairytale of the GSSG/GSH redox potential. Biochim. Biophys. Acta 1830:3139–3142; 2013. [54] Morgan, B.; Ezerina, D.; Amoako, T. N.; Riemer, J.; Seedorf, M.; Dick, T. P. Multiple glutathione disulfide removal pathways mediate cytosolic redox homeostasis. Nat. Chem. Biol. 9:119–125; 2012. [55] Rossi, R.; Milzani, A.; Dalle-Donne, I.; Giustarini, D.; Lusini, L.; Colombo, R.; Di Simplicio, P. Blood glutathione disulfide: in vivo factor or in vitro artifact? Clin. Chem. 48:742–753; 2002. [56] Saez, G.; Thornalley, J. P.; Hill, O. A. H.; Hems, R.; Bannister, V. J. The production of free radicals during the autoxidation of cysteine and their effect on isolated rat hepatocytes. Biochim. Biophys. Acta 719:24–31; 1982. [57] Munday, R.; Munday, C. M.; Winterbourn, C. C. Inhibition of copper-catalyzed cysteine oxidation by nanomolar concentrations of iron salts. Free Radic. Biol. Med. 36:757–764; 2004. [58] Arnold, U.; Ulbrich-Hofmann, R. Quantitative protein precipitation from guanidine hydrochloride-containing solutions by sodium deoxycholate/trichloroacetic acid. Anal. Biochem. 271:197–199; 1999. [59] Bensadoun, A.; Weinstein, D. Assay of proteins in the presence of interfering materials. Anal. Biochem. 70:241–250; 1976. [60] Cohn, V. H.; Lyle, J. A fluorometric assay for glutathione. Anal. Biochem. 14:434–440; 1966. [61] Donovan, J. W.; White, T. M. Alkaline hydrolysis of the disulfide bonds of ovomucoid and of low molecular weight aliphatic and aromatic disulfides. Biochemistry 10:32–38; 1971. [62] Ogwu, V.; Cohen, G. A simple colorimetric method for the simultaneous determination of N-acetylcysteine and cysteine. Free Radic. Biol. Med. 25:362–364; 1998. [63] Georgiou, C. D.; Grintzalis, K.; Zervoudakis, G.; Papapostolou, I. Mechanism of Coomassie brilliant blue G-250 binding to proteins: a hydrophobic assay for nanogram quantities of proteins. Anal. Bioanal. Chem. 391:391–403; 2008. [64] Obrador, E.; Carretero, J.; Ortega, A.; Medina, I.; Rodilla, V.; Pellicer, J. A.; Estrela, J. M. Gamma-glutamyl transpeptidase overexpression increases metastatic growth of B16 melanoma cells in the mouse liver. Hepatology 35:74–81; 2002. [65] Dubler, R. E.; Anderson, B. M. Simultaneous inactivation of the catalytic activities of yeast glutathione reductase by N-alkylmaleimides. Biochim. Biophys. Acta 14:70–85; 1981. [66] Witt, W.; Rüstow, B. Determination of lipoic acid by precolumn derivatization with monobromobimane and reversed-phase high-performance liquid chromatography. J. Chromatogr. B Biomed. Sci. Appl 705:127–131; 1998. [67] Cobbett, C.; Goldsbrough, P. Phytochelatins and metallothioneins: roles in heavy metal detoxification and homeostasis. Annu. Rev. Plant. Biol. 53:159–182; 2002. [68] Collin-Hansen, C.; Andersen, R. A.; Steinnes, E. Isolation and N-terminal sequencing of a novel cadmium-binding protein from Boletus edulis. J. Phys. IV 107:311–314; 2003. [69] Zhao, J.; Kan, Q.; Wen, J.; Li, Y.; Sheng, Y.; Yang, L.; Wu, J.; Zhang, S. Hemolysis of blood samples has no significant impact on the results of pharmacokinetic data. J. Bioequiv. Availab 4:82–85; 2012. [70] Turell, L.; Radi, R.; Alvarez, B. The thiol pool in human plasma: the central contribution of albumin to redox processes. Free Radic. Biol. Med. 7:244–253; 2013. [71] Friedman, M. Applications of the ninhydrin reaction for analysis of amino acids, peptides, and proteins to agricultural and biomedical sciences. J. Agric. Food Chem. 52:385–406; 2004.