Báo cáo khoa học: Oxidative stress induces a reversible flux of cysteine from tissues to blood in vivo in the rat

Oxidative stress induces a reversible flux of cysteine from tissues to blood in vivo in the rat Daniela Giustarini1, Isabella Dalle-Donne2, Aldo Milzani2 and Ranieri Rossi1

1 Department of Evolutionary Biology, Laboratory of Pharmacology and Toxicology, University of Siena, Italy 2 Department of Biology, University of Milan, Italy

Keywords cysteine; diamide; glutathione; oxidative stress; thiols

Correspondence R. Rossi, Department of Evolutionary Biology, Laboratory of Pharmacology and Toxicology, University of Siena, via A. Moro 4, I-53100 Siena, Italy Fax: +39 577 234476 Tel: +39 577 234198 E-mail: ranieri@unisi.it

(Received 11 May 2009, revised 30 June 2009, accepted 3 July 2009)

doi:10.1111/j.1742-4658.2009.07197.x

Glutathione (GSH) plays a key role in defense against oxidative stress. The availability of GSH is ensured in tissues by systems devoted to its mainte- nance in the reduced state and by the flux of GSH and cysteine between sites of biosynthesis and sites of utilization. Little is known about the effect of oxidative stress on the distribution of low-molecular-mass thiols and their exchange rate between tissues. In this study, we found that a slow infusion of diamide (a specific thiol-oxidizing compound) evoked a dra- matic increase in blood cysteine in rats. Our data suggest that inter-organ exchange of cysteine occurs, that cysteine derives from both glutathione via c-glutamyl transpeptidase and methionine via homocysteine and the trans- sulfuration pathway, and that these pathways are considerably influenced by oxidative stress.

Introduction

intracellularly and cellular environment, and many pathophysiological conditions has been shown to markedly influence both thiol and disulfide concentrations [1,10–13].

Cysteine (Cys) and glutathione (GSH) are the most abundant low-molecular-mass thiols (LMM-SH), with cysteine GSH predominating predominating in extracellular fluids [1,2]. These two compounds are metabolically inter-related, and GSH, in particular, determines the redox state and represents a defense against damage mediated by reactive oxygen species (ROS) or reactive nitrogen species.

Abbreviations CySS, cystine; CysGly, cysteinylglycine; CySSGly, cystinylglycine; GSH, glutathione; GSSG, glutathione disulfide; GSSP, mixed disulfides between protein sulfhydryl groups and glutathione; c-GT, c-glutamyl transpeptidase; Hcys, homocysteine; HcySS, homocystine; LMM-SH, low-molecular-mass thiols; mBrB, monobromobimane; NEM, N-ethylmaleimide; RBC, red blood cell; ROS, reactive oxygen species; RSSP, mixed disulfides between protein sulfhydryl groups and a low-molecular-mass thiol; TSP, trans-sulfuration pathway.

FEBS Journal 276 (2009) 4946–4958 ª 2009 The Authors Journal compilation ª 2009 FEBS

4946

Low GSH concentrations and a ratio of high gluta- thione disulfide (GSSG) to GSH have been measured in red blood cells (RBCs) of patients with various diseases, surgery, including cancer, coronary heart pre-eclampsia, and genitourinary, gastrointestinal, car- diovascular and musculoskeletal diseases [3–9]. Extra- cellular fluids such as plasma are characterized by a lower thiol : disulfide ratio in comparison to the intra- GSH is synthesized intracellularly from its constituent amino acids, cysteine, glutamate and glycine, by a two- step reaction catalyzed by the enzymes c-glutamylcyste- ine synthetase and GSH synthetase. The supply of Cys is directly related to the rate of GSH synthesis, because the cellular cysteine concentration is a key determinant in regulating the kinetics of the first reaction (i.e. the formation of c-glutamylcysteine), which represents the rate-limiting step in GSH synthesis [14]. Cysteine is rap- idly auto-oxidized to cystine (CySS) in the extracellular fluids; CySS may have a considerable physiological role as a source of Cys, as, once it has entered a different cell, it can be reduced again by GSH [15].

D. Giustarini et al.

Cysteine flux under oxidative stress

increase, particularly evident

A continuous flux of GSH and Cys exists between various tissues, and diet, starvation and pathologies that affect various organs may influence this phenome- non [2,16,17]. Many pathological conditions are known or thought to elicit oxidative stress, and it is recog- nized that this may have a role in the progression of the disease (for review, see [18]). Usually, the occur- rence of oxidative stress is evaluated by measuring sev- eral biomarkers, but little is known about the influence of oxidative stress on cysteine and glutathione exchange among tissues. Indeed, the factors that main- tain the redox homeostasis of plasma thiols as well as the mechanisms that regulate the exchange of LMM- SH among cells need to be clarified.

amount thiols total the of

In previous studies [21]. course of levels of GSH, Cys, CysGly and Hcys and the corresponding disulfide forms calculated as the sum of LMM-SS and RSSP (Fig. 1A–D). Values for total disulfides are expressed as ‘thiol equivalents’, i.e. the concentration of LMM-SH that are involved in formation of both LMM-SS and RSSP. The level of Cys significantly decreased by (cid:2) 20% during treatment with diamide, and then returned slowly to initial for values. A parallel disulfides of cysteine, followed by a decrease in the disulfide forms of both glutathione and cysteine, was also observed, with a tendency (particularly for GSH) levels (Fig. 1A,B). Disulfides of to return to basal homocysteine tended to increase during the diamide infusion and remained higher than the basal concentra- tion during the whole experiment (Fig. 1C). Con- the CysGly redox balance was minimally versely, influenced by treatment with diamide (Fig. 1D). When considering (i.e. reduced + disulfide forms) found in plasma at each time point during the whole course of the experiment (Fig. 2), we observed that the levels of total cysteine (and also total homocysteine, although to a smaller extent) were markedly increased during and after diamide infusion with respect to time zero.

Here, we have analyzed the effect of the administra- tion of azodicarboxylic acid bis-dimethylamide (dia- mide), a thiol-specific oxidizing substance, on the blood and tissue distribution of thiols and disulfides in rats. Diamide is a mild oxidizing compound that easily penetrates cell membranes and reacts quickly and spe- cifically with intracellular thiols, both LMM-SH and protein sulfhydryls [19,20], without producing free rad- icals or other ROS, as indicated by the lack of increase in lipid peroxidation observed in in vitro and in vivo experiments [22,23], we observed that a single intraperitoneal administration of diamide in rats induced selective oxidation of blood glutathione, with no evident alterations in thiol levels in other tissues. Moreover, during diamide treatment, Cys levels increased in plasma, indicating that Cys was released from an extra-hematic compartment.

As in vivo studies based on treatments with diamide are essentially limited by its reaction with thiol groups within a few seconds [24], in this paper we used slow intravenous infusion of the drug. Additionally, we monitored all physiological LMM-SH, namely Cys, cysteinylglycine (CysGly), homocysteine (Hcys), GSH and related disulfide forms in both blood and organs. Disulfide forms comprise low-molecular-mass disulfides [LMM-SS: CySS, cystinylglycine (CySSGly), homocys- tine (HcySS) and GSSG] and mixed disulfides between protein sulfhydryl groups and a low-molecular-mass thiol (RSSP). This allowed us to study relatively long- lasting effects of thiol-directed oxidative stress, and, in particular, its influence on the inter-organ exchange of LMM-SH.

Results

FEBS Journal 276 (2009) 4946–4958 ª 2009 The Authors Journal compilation ª 2009 FEBS

4947

In a first set of experiments, the kinetics of plasma thiols during and after diamide administration was evaluated. Rats were treated for 60 min and monitored for the next 2 h. Specifically, we measured the time The same samples were also analyzed for the LMM- SH, LMM-SS and RSSP content in RBCs. These cells the main stores of hematic GSH, which represent occurs mostly (> 99%) in the reduced form [25], whereas Cys and other LMM-SH represent only a minimal fraction [26]. GSH was rapidly oxidized in RBCs mainly to mixed disulfides with proteins (GSSP) and recovered (cid:2) 80% of its initial value within 2 h of the end of treatment (Fig. 3B). Cys levels decreased during the first part of the treatment (10–30 min), but the prevailing phenomenon was a dramatic reversible increase in both the reduced and, in particular, the disulfide forms, mainly mixed disulfides with proteins (CySSP) (Fig. 3A). The highest value reached for CyS- SP was (cid:2) 2.5 nmol per mg Hb (i.e. 0.75 mm), and was measured at the end of treatment with diamide. This suggests a large, reversible flux of cysteine from other tissues towards RBCs. Figure 3C shows the change in total cysteine and total glutathione in RBCs with respect to their basal levels. It is evident that the levels of total GSH, greatly and total cysteine, but not levels after oxidative reversibly increased over basal stress induction by diamide. The change in total cyste- ine concentration in both plasma and RBCs during diamide treatment and over the subsequent 2 h is sum- marized in Fig. 4. Values were normalized for the hematocrit, and the level of total cysteine greatly increased in plasma and RBCs, reaching a maximum

D. Giustarini et al.

Cysteine flux under oxidative stress

A

B

C

D

Fig. 1. Thiols and disulfides in rat plasma during and after diamide infusion. Time course of plasma LMM-SH (closed circles) and total disulfide (open circles) levels during and after diamide infusion (7.73 lmolÆmin)1Ækg)1 intravenously). (A) Cysteine; (B) glutathione; (C) homocysteine; (D) cysteinylglycine. Disulfides represent the sum of low-molecular-mass disulfides and protein mixed disulfides; values are expressed as thiol equivalents. Data are the means ± SD of four experiments. *P < 0.05 versus value at time zero; **P < 0.01 versus value at time zero.

that

FEBS Journal 276 (2009) 4946–4958 ª 2009 The Authors Journal compilation ª 2009 FEBS

4948

level 75 min after the start of diamide infusion. It is evident the cysteine that enters RBCs cannot derive solely from plasma stores. We also monitored the concentrations of thiols and disulfides in blood withdrawn from rats at 75 min after the beginning of diamide infusion (when total cysteine reached its maxi- mum observed levels): both plasma and RBCs were analyzed ex vivo over time. The observed concentra- tion changes of GSH, Cys and their oxidized forms in RBCs were similar to those observed in vivo, with thi- ols and disulfides recovering their basal levels within (cid:2) 2 h after blood collection (data not shown). The plasma Cys levels increased slightly over time, but more evident was the increase in CySS and CySSP,

D. Giustarini et al.

Cysteine flux under oxidative stress

lung and heart, a slight but increase in Hcys wa seen, both at

Fig. 2. Total thiols in rat plasma during and after diamide infusion. Time course of total plasma thiols (sum of reduced + disulfide forms) levels during and after diamide infusion (7.73 lmolÆmin)1Ækg)1 intravenously). Data are the means ± SD from the experiments shown in Fig. 1. *P < 0.05 versus value at time zero; **P < 0.01 versus value at time zero.

also observed (+16.2 ± 5.1% and organs was in liver, +20.9 ± 3.2%, respectively). Additionally, significant kidney, (P < 0.05) the end of diamide infusion (i.e. 60 min) and at the end of the experiment (i.e. 180 min) (data not shown). These data suggest that diamide has a minimal effect on the thiol ⁄ disulfide balance of tissues other than blood, and did not induce evident decreases in cyste- ine and GSH levels in the organs analyzed. Together,

probably due to oxidation of cysteine exported from RBCs (Fig. 5). Little GSH was exported during this ex vivo experiment, and the concentrations of total GSH were measured within RBCs were constant (data not shown). In animals

In

The change in total cysteine levels was found to be dose-dependent. performed with experiments various doses of diamide (dose range 1.93–9.66 lmolÆ min)1Ækg)1), the increase in the concentration of total cysteine showed an almost linear dose dependence, as shown in Fig. 6, in which the maximum total cysteine accumulated in RBCs and plasma is plotted against the diamide dose administered. these findings prompted us to further investigate the origin of cysteine that enters blood. In theory, two possible sources exist: (a) intracellular GSH is exported and converted extracellularly to Cys by c-glutamyl transpeptidase (c-GT) and dipep- tidases, with cysteine being then taken up by RBCs, or (b) methionine is first converted into Hcys, and then, through the trans-sulfuration pathway (TSP), to Cys. To assess these two hypotheses, we repeated the experiments using rats pre-treated with acivicin and ⁄ or propargylglycine. Acivicin is an inhibitor of c-GT, thus this treatment should eliminate the frac- tion of Cys that derives from GSH [27]. Propargyl- glycine is an inhibitor of cystathionase, the enzyme that forms cysteine from cystathionine in the TSP; in this case, treatment should eliminate the supply of Cys derived from methionine [28]. Pre-treatment with these inhibitors in control rats (administered with saline) completely abolished the release of Cys into treated with diamide, blood (Fig. 8). propargylglycine and acivicin significantly decreased the flux of cysteine towards the hematic compart- ment by (cid:2) 25% and 30%, respectively. Concomitant use of these inhibitors enhanced the inhibitory effect on the Cys increase in blood. This suggests that both pathways contribute to the large cysteine efflux from organs into the hematic compartment observed in our experiments.

Discussion

diamide after end of significant

FEBS Journal 276 (2009) 4946–4958 ª 2009 The Authors Journal compilation ª 2009 FEBS

4949

We further evaluated the effect of diamide infusion on various rat organs in order to assess whether dia- mide was also able to affect the thiol ⁄ disulfide bal- ance of these organs, and whether cysteine efflux to the blood was paralleled by a change in the Cys concentration in other tissues. Diamide, at 60 min from the start of administration, evoked a significant decrease in the hepatic Cys level, followed by an increase in both liver and kidney (after 2 h from the end of administration), but the GSH concentration was found to be significantly higher only in the liver 2 h administration the (P < 0.05) (Fig. 7A,B ). A slight but increase in the concentration of total disulfide forms of GSH in heart and lung but not in other analyzed In this study, a systemic response to a thiol-directed oxidative insult was induced by means of slow infu- sion of diamide in the rat, which evoked a relatively long-lasting oxidant stimulus. Little is known about the effect of oxidative stress on the distribution of LMM-SH in extracellular fluids and their exchange rate between tissues. The paucity of information in this field is due, at least in part, to the intrinsic diffi- culty in creating reliable animal models of oxidative stress. Acute carbon tetrachloride administration is frequently used as a rodent model for oxidative stress; these experiments are promising, and reports

D. Giustarini et al.

Cysteine flux under oxidative stress

B

A

C

Fig. 3. Thiols and disulfides in rat erythrocytes during and after diamide infusion. Time course of erythrocyte LMM-SH (closed circles), LMM-SS (closed squares) and RSSP (closed triangles) levels during and after diamide infusion (7.73 lmolÆmin)1Ækg)1 intravenously). (A) Cys- teine; (B) glutathione. (C) Increase in total cysteine and total glutathione levels over values at time zero. Values are the sum of the thiol and disulfide concentrations shown in (A) and (B), and are expressed as thiol equivalents. Data are the means ± SD of four experiments. *P < 0.05 versus value at time zero; **P < 0.01 versus value at time zero.

[30,31].

FEBS Journal 276 (2009) 4946–4958 ª 2009 The Authors Journal compilation ª 2009 FEBS

4950

have demonstrated which antioxidant parameters are influenced by these treatments [29,30]. However, these models are harsh, many free radicals are produced as a result of the reductive bioactivation of carbon tetra- chloride by cytochrome P450, and the animals may suffer multi-organ damage. Thus, information on the inter-organ trafficking of small thiols is difficult to interpret In contrast, diamide is a thiol- specific oxidizing agent that rapidly reacts with GSH and other thiols without production of free radicals or other ROS. Further, we performed experiments by a slow infusion of diamide to avoid the problem that

D. Giustarini et al.

Cysteine flux under oxidative stress

Fig. 4. Total cysteine in rat blood during and after diamide infusion. Time course of the total cysteine (sum of reduced thiol + disul- fides) increase over the levels at time zero in rat blood during and after diamide infusion (7.73 lmolÆmin)1Ækg)1 intravenously). The reported values were calculated by normalization of measured con- centrations of total cysteine in both plasma (black bars) and erythro- cytes (gray bars) to the relative hematocrit value. Values are expressed as thiol equivalents. Data are the means ± SD of four experiments.

Fig. 6. Dose dependence of the maximum increase of cysteine induced by diamide. Values indicate the increase of total Cys over levels at time zero in rat blood 75 min after the start of diamide infusion (60 min infusion, dose range 1.93–9.66 lmolÆmin)1Ækg)1 intravenously). The reported values were calculated by normaliza- tion of measured concentrations of total cysteine in both plasma and erythrocytes to the relative hematocrit value. Values are expressed as thiol equivalents. Data are the means ± SD of five experiments.

little effect was found for GSH,

value at

Fig. 5. Thiols and disulfides in rat plasma ex vivo after diamide infusion. Ex vivo levels of plasma Cys (triangles), CySS (squares) and CySSP (circles) after diamide infusion (7.73 lmolÆmin)1Ækg)1 intravenously) and blood withdrawal. Blood was withdrawn 15 min after the end of diamide infusion and maintained at 37 (cid:2)C. Time zero indicates the time of withdrawal. Data are the means ± SD of three experiments. *P < 0.05 versus time zero; **P < 0.01 versus value at time zero.

The main finding of our experiments was a revers- ible accumulation of cysteine within the blood (Fig. 4). Our previous studies [22,23] showed a slight increase in hematic cysteine after a single intraperitoneal diamide administration to rats, accompanied by oxidation of hematic GSH; its disulfide forms, and antioxidant enzyme levels in tis- sues other than blood [23]. Conversely, a 1 h treatment with diamide, like that performed here, evoked dra- matic changes in hematic thiols and disulfides. More- over, the features of cysteine accumulation in blood were studied in detail by separate analysis of plasma and erythrocyte compartments. The total cysteine con- centration in the plasma compartment increased from (cid:2) 100 to (cid:2) 140 lm (Fig. 2) during diamide administra- tion. In contrast, the levels of reduced cysteine did not change (Fig. 1A). This is probably the result of two phenomena: enhanced delivery of cysteine into the blood by tissues, and oxidation elicited by diamide. The total Hcys level also showed a reversible increase (Fig. 2).

FEBS Journal 276 (2009) 4946–4958 ª 2009 The Authors Journal compilation ª 2009 FEBS

4951

More dramatic effects were observed in erythrocytes, in which GSH was reversibly oxidized (Fig. 3B), and, more importantly, a large increase in the cysteine con- in both the reduced and disulfide forms, centration, was observed (Fig. 3A). Therefore, a high amount of cysteine is delivered into the plasma and rapidly taken the diamide effect is limited to only a few seconds once administered to rats, because of its high reacti- vity [22,23].

D. Giustarini et al.

Cysteine flux under oxidative stress

A

B

Fig. 7. Effect of diamide infusion on thiols in various rat tissues. Rats were administered diamide (7.73 lmolÆmin)1Ækg)1 intravenously) for 1 h and killed 60 or 180 min after the start of the treatment. The levels of Cys (A) and GSH (B) in various tissues are reported. Values at time zero are those obtained in rats treated analogously (i.e. implanted with the valve), but without any infusion. Data are the means ± SD of four experiments. *P < 0.05 versus untreated animals; *P < 0.01 versus untreated animals.

this transport system was

evidenced by constant levels of

blocked); therefore, it may require a reduced mem- brane thiol. It is also possible that, under our experi- less mental conditions, efficient during the oxidative perturbation induced by diamide. In contrast to Cys, GSH returned to basal levels probably as a result of thiol ⁄ disulfide reactions and NADPH-dependent reduction. De novo GSH syn- thesis did not occur under our experimental condi- tions, as total glutathione measured in vivo (Fig. 3) and ex vivo (data not shown). This suggests that Cys loss in RBCs does not represent a device to counterbalance de novo GSH synthesis.

FEBS Journal 276 (2009) 4946–4958 ª 2009 The Authors Journal compilation ª 2009 FEBS

4952

up by RBCs during diamide administration (Fig. 4). In RBCs, cysteine accumulated mainly as protein mixed disulfides, conceivably linked to the highly reactive b-125 cysteine residues of hemoglobin. Analogously, GSH is mostly oxidized to form mixed disulfides with hemoglobin, with a minimal increase in GSSG concen- tration [20]. This extraordinarily high accumulation of cysteine in RBCs was found to be reversible, and Cys reached its initial levels within 2 h after the end of diamide administration. Cysteine is probably reduced by the recovered GSH and exported from RBCs. In fact, it has recently been observed that Cys-enriched erythrocytes export cysteine in a time- and concentra- tion-dependent manner [32]. Indeed, in ex vivo experi- ments, we observed a significant and progressive increase in total plasma cysteine, mainly CySS and CySSP, after diamide infusion, corresponding to the cysteine concentration decrease within RBCs (Fig. 5). Given that plasma is a rather oxidizing environment, which lacks reductases [2], it is likely that the exported Cys is oxidized both to CySS and RSSP. We cannot exclude the possibility that a fraction of CySS may be exported from erythrocytes. Nevertheless, our previous data with washed RBCs indicated that, under physio- logical conditions, human erythrocytes release cysteine over time, mostly in the reduced form [26]. Although the erythrocyte transport system for Cys efflux has not yet been identified, Yildiz et al. [32] suggested that this process is carrier-mediated and is significantly attenu- is ated when GSH is depleted (or synthesis its The reason for the observed Cys accumulation, first in plasma and then in RBCs, is not clear. It is possible that the thiol–disulfide status in plasma is finely tuned, and the diamide-induced oxidation of Cys is compen- sated for by efflux of GSH and Cys from the liver and other organs to maintain the cysteine pool. GSH is known to be a critical source for maintaining a steady Cys availability as it is continuously exported out of cells into plasma and converted to circulating cysteine by the action of c-GT and dipeptidase (Fig. 9). Among the various tissues, liver has been shown to play a cen- tral role in the GSH homeostasis, serving as the princi- pal source of the GSH circulating in plasma [14,33]. The extracellular translocation of hepatic GSH and its c-GT-dependent extracellular catabolism, resulting in formation of plasma Cys and CySS (cysteine is rapidly oxidized to cystine extracellularly [34]), could be a

D. Giustarini et al.

Cysteine flux under oxidative stress

thiol equivalents. Data

are expressed as

Fig. 8. Effect of acivicin and propargylglycine on the cysteine Increase in cysteine (sum of reduced cysteine + disul- increase. fides) over levels at time zero in rat blood induced by diamide infu- sion (7.73 lmolÆmin)1Ækg)1 intravenously) after pre-treatment with acivicin (open circles) or propargylglycine (closed triangles) or both (open triangles). Data are compared with values obtained from rats treated with diamide (7.73 lmolÆmin)1Ækg)1 intravenously) without pre-treatments (closed circles), or from rats treated with acivicin (gray squares) or propargylglycine (closed squares) without diamide infusion. Disulfides represent the sum of low-molecular-mass disul- fides and protein mixed disulfides. The reported values were calcu- lated by normalization of measured concentrations of total cysteine in both plasma and erythrocytes to the relative hematocrit value. Values are the means ± SD of three experiments.*P < 0.05 versus samples trea- ted only with diamide analyzed at the same time.

necessarily produce a significant, quantifiable decrease in its levels in the organs themselves. Indeed, although a 1–2% release of GSH does not induce a significant decrease in its concentration in liver or in kidneys, it provokes a marked effect in plasma, where its levels (and in general the levels of all LMM-SH) are in the micromolar range. In contrast, as disulfide levels are very low in the intracellular compartments ((cid:2) 2000– 3000-fold lower than thiol compounds [2]), we can assume that their contribution to the increase in Cys is negligible.

FEBS Journal 276 (2009) 4946–4958 ª 2009 The Authors Journal compilation ª 2009 FEBS

4953

it of a systemic response to local oxidative conditions. Never- theless, after a 60 min diamide infusion, we observe a significant decrease in hepatic Cys levels only (10 lm), whereas change the GSH concentration did not (Fig. 7). Analogously, disulfide levels did not vary to a significant extent after diamide administration. Even thought the whole amount of cysteine that disappears from the liver is released into blood, it cannot quanti- tatively explain the dramatic rise of Cys registered within erythrocytes. Nevertheless, we should consider these phenomena as a dynamic process to which numerous factors contribute in order to maintain fairly constant thiol levels in various organs and to supply the compounds necessary to counteract the diamide challenge. In this context, a possible limit of our study is that we did not record the multi-directional flux of various thiols in ‘real time’; we could only perform our measurements at fixed times, and these values obtained are therefore the result of various events that are not completely discriminated. Additionally, given the high concentration of GSH in organs (2–10 mm), is release does not evident that an increase in its Additional factors may be considered to be involved in the observed process of cysteine accumulation (Fig. 9). Extracellular cysteine can also derive from intracellular stores of various tissues that can deliver it to the plasma; the other main source is diet. During prolonged starvation, skeletal muscle, in particular, can deliver cysteine to the plasma by degradation of pro- teins [14]. Methionine from protein degradation or from intracellular stores can also serve as a source of cysteine through the action of the TSP [35]. Hcys is synthesized from methionine by almost all cells through the activated methyl cycle. Hcys can be reconverted into methionine (with formation of tetrahydrofolate) or cysteine through the TSP; alternatively, it can be exported [36]. In our experiments, pre-treatment with acivicin or propargylglycine, which inhibit either c-GT or cystathionase, respectively, should block cysteine influx to the blood during diamide infusion. In both cases, a decrease of 25–30% in the hematic levels of cysteine was observed. The effect was strengthened by concurrent administration of both inhibitors (Fig. 8). This suggests that both sources of Cys from tissues, i.e. GSH via c-GT and methionine via Hcys and the TSP, are involved. Notwithstanding this, methionine avail- ability is unlikely to be a limiting factor for plasma cysteine flux from tissues, because, under oxidative con- ditions, many factors can contribute to deliver cysteine to blood. Interestingly, stimulation of the TSP by oxidative stress has been demonstrated previously [35,37,38], suggesting that the redox sensitivity of the trans-sulfuration pathway may be considered to be an auto-corrective response that leads to an increased level of glutathione synthesis in cells challenged by oxidative stress. Evidence also exists indicating that c-GT may be upregulated as adaptive response to an oxidative insult [39]. Experiments performed using various diamide doses (Fig. 6) showed that the process is dose-depen- dent. Therefore, the observed phenomenon may derive from an oxidative regulation of one or more proteins critical to the metabolism of cysteine, or from activa- tion systemic mechanism of Cys ⁄ GSH delivery ⁄ uptake regulation.

D. Giustarini et al.

Cysteine flux under oxidative stress

Fig. 9. Schematic diagram showing the metabolic pathways of various physiological LMM-SH. Intracellular Cys is a key determinant in regu- lating the kinetics of formation of c-glutamylcysteine (c-GluCys), the first step in GSH synthesis. Intracellular Cys may derive from diet, intra- cellular stores, or methionine (Met), which is converted into Hcys by the activated methyl cycle. Hcys may then condense with serine to produce cystathionine, which in turn may be hydrolyzed to Cys. The conversion of cystationine into Cys is catalyzed by the enzyme c-cysta- thionase, the activity of which is inhibited by propargylglycine (PGG). Both GSH and Cys can be exported from cells into plasma, where they undergo auto-oxidation. The liver and kidneys appear to have significant capacity for GSH efflux. Although it has been established that cells are able to export excess GSSG, it is not yet known whether they also possess a transport system for CySS. Plasma Cys, but neither GSH nor GSSG, can be taken up by most cells (also including RBCs). Some cell types (excluding RBCs) also possess a transport system for uptake of CySS. Extracellular GSH can be converted into CysGly and then into Cys by the combined action of the membrane enzymes c-glutamyl transpeptidase (cGT) and dipeptidases (DP). The activity of cGT is inhibited by acivicin. Alternatively, CysGly may be taken up and converted into Cys by DPs.

subsequently converted into GSH. An increase in liver GSH after an acute increase of circulating cysteine has been described previously [43].

FEBS Journal 276 (2009) 4946–4958 ª 2009 The Authors Journal compilation ª 2009 FEBS

4954

To summarize, our data suggest that diamide shifts the thiol:disulfide ratio in plasma and RBCs. To coun- teract this effect, Cys is likely to be delivered from multiple tissues; a fraction of Cys is oxidized within plasma by diamide, whereas another fraction enters RBCs, where it is oxidized to CySS and mostly CySSP. Once the diamide infusion and, consequently, the oxi- dant stimulus, is complete, cysteine is taken up de novo by tissues other than blood (Fig. 7A). We infer that ) transport system for CySS may contribute to the xc remove the cysteine accumulated within erythrocytes and plasma. In fact, it has been demonstrated that this transport system, which is widely distributed in various organs [40,41], is upregulated under oxidative condi- tions [42]. Therefore, it is reasonable to hypothesize that, under our experimental conditions, CySS uptake into various organs is enhanced, followed by CySS reduction to Cys. Indeed, a clear increase in Cys levels was observed 2 h after the end of diamide infusion in the kidneys and the liver. Additionally, a significant increase in GSH was observed in the liver at 3 h after the start of treatment (Fig. 7B), probably indicating that the excess cysteine enters this organ, where it is Maintenance of an adequate plasma thiol–disulfide balance appears to be fundamental. It has been dem- onstrated that even a minimal shift in the redox state of either the Cys ⁄ CySS or GSH ⁄ GSSG pool (e.g. 21 and 9 mV changes, respectively, as observed in aging) is sufficient to cause a large increase in the oxidized forms of intracellular proteins bearing vicinal thiols, which can influence specific signaling pathways [44]. In addition, it has been observed that the NMDA recep- tor, protein disulfide isomerase, epidermal growth fac- tor receptor and extracellular signal-regulated kinase are modulated by the cellular and ⁄ or extracellular redox state [45–48]. The shift in the thiol ⁄ disulfide redox state over the range found in vivo in human plasma is a key determinant of early events of vascular disease development [49]. In this context, it should be noted that Hcys levels in plasma (and in some other tis- sues) were also reversibly increased in our experiments by the thiol-directed oxidant, although to a minor extent compared with Cys levels (Figs 1C and 2). This

D. Giustarini et al.

Cysteine flux under oxidative stress

effect could be indicative of diamide-induced stimula- tion of methionine transformation into Cys, a pathway in which Hcys represents an intermediate. Given the well-known action of Hcys as a pro-atherogenic and cardiovascular risk factor, the fact that diamide treat- ment (and probably other events that perturbate the thiol ⁄ disulfide status of plasma) may influence homo- cysteine levels suggests that this topic deserves further investigation.

Many authors have focused their attention on the possible role of modulation of LMM-SH levels in dis- ease development and ⁄ or progression. However, only a few studies have investigated the physiological rules that govern the distribution of LMM-SH in extracellular flu- ids, the contribution of various tissues, the possible con- sequence(s) of decreased ⁄ increased efflux from a single organ because of pathophysiological conditions, and, more importantly, the influence of oxidative stress on these processes. Our data clarify some of these aspects, and suggest that thiol-directed oxidative stress consider- ably influences the inter-organ exchange of cysteine and its production from methionine and ⁄ or GSH.

Experimental procedures

acidified by 1 : 1 addition of a solution of trichloroacetic acid (TCA, 12% w ⁄ v). After separation of proteins by cen- trifugation at 15 000 g for 2 min at room temperature, 25 lL of the supernatant was brought to a pH of (cid:2) 8.0 using 5 lL of 2 m Tris, and then 0.5 lL of 40 mm mono- bromobimane (mBrB; Calbiochem, Milan, Italy) dissolved in methanol was added for LMM-SH measurement. After a 10 min incubation in the dark, samples were acidified and analyzed by HPLC as previously described [1]. A further 25 lL of plasma were added with 1 lL of 50 mm N-ethyl- maleimide (NEM; Sigma-Aldrich, Milan, Italy,) for mea- surement of both low-molecular-mass disulfides and protein mixed disulfides (total disulfides). After 10 min incubation at room temperature, NEM was extracted with four vol- umes of dichloromethane, 2 mm (final concentration) of dithiothreitol was added to reduce disulfides, and after 5 min samples were 1 : 1 diluted with 12% w ⁄ v TCA and then proteins were discarded by centrifugation at 15 000 g for 2 min at room temperature. Supernatants (30 lL) were brought to a pH of (cid:2) 8.0 using 6 lL of 2 m Tris, and then 1.2 lL of mBrB was added. After a 10 min incubation, samples were acidified with HCl and analyzed by HPLC as previously reported [1]. All samples with visible hemolysis were discarded. For determination of LMM-SS, 5 lL aliquots of plasma (after addition of NEM, as described above) were 1 : 3 diluted with 8% w ⁄ v TCA, and then proteins were discarded by centrifugation at 15 000 g for 2 min at room temperature. Supernatants (10 lL) were brought to a pH of (cid:2) 8.0 using 2 lL of 2 m Tris, then 2 mm dithiothreitol (final concentration) was added. Sam- ples were then derivatized using mBrB, and analyzed as described above. The level of protein mixed disulfides was calculated as the difference between total disulfides and LMM-SS.

Sprague–Dawley male rats (400–450 g) were purchased from Charles River Laboratories (Calco, Italy). A double valve (model 617, 20 · 20 mm; Danuso Instruments, Milan, Italy) was implanted in each animal; jugular and femoral veins were cannulated for either drug administration or blood collection, as described previously [20]. The valve was implanted under pentobarbital anesthesia two days before the experiment. Animals were allowed to freely move and fed ad libitum before and during the experiments.

The RBC pellet was washed three times with ice cold Na+ ⁄ K+ phosphate-buffered saline (pH 7.4) containing 5 mm glucose for LMM-SH analyses or 5 mm glucose and 80 mm NEM for disulfide analyses, and hemolyzed using 10 volumes of 10 mm Na+ ⁄ K+ phosphate buffer, pH 7.4. Aliquots of the hemolysate were used for LMM-SH, total disulfide or LMM-SS determinations as described above.

Rats received infusions of diamide in saline (7.73 lmol Æmin)1Ækg)1 unless otherwise specified) via the cannula implanted in the femoral vein. The infusion was started immediately after blood withdrawal for time zero measure- ments and lasted 1 h. Blood aliquots (100 lL each) were collected through the valve connected to the jugular vein and immediately processed. All animal manipulations were performed in accordance with the European Community guidelines for the use of laboratory animals. The experi- ments were authorized by the local ethics committee. Rats did not show any evidence of distress or altered behavior during the experiments.

For ex vivo experiments, after 75 min from the start of diamide infusion, (cid:2) 600 lL of blood was withdrawn from the valve (K3EDTA was used to prevent blood clotting) and maintained at 37 (cid:2)C under gentle gyratory shaking for analyses of thiols and related disulfides. At the indicated times, aliquots of blood were centrifuged at 15 000 g for 15 s. Thiols and disulfides were analyzed in both plasma and RBCs as described above.

Animals

All HPLC analyses were performed by using a Spheri- sorb C18 column (4.6 mm internal diameter · 250 mm column length) (Varian Inc, Palo Alto, CA) by means of an Agilent series 1100 HPLC (Agilent Technologies, Milan, Italy) equipped with a fluorometric detector.

Blood samples were collected in plastic tubes containing K3EDTA. Blood was immediately centrifuged at 15 000 g for 15 s. Aliquots of plasma (20 lL) were immediately

FEBS Journal 276 (2009) 4946–4958 ª 2009 The Authors Journal compilation ª 2009 FEBS

4955

Thiol and disulfide measurement in blood

D. Giustarini et al.

Cysteine flux under oxidative stress

4 Ne´ meth I, Orvos H & Boda D (2001) Blood glutathione redox status in gestational hypertension. Free Radic Biol Med 30, 715–721.

5 Nobili V, Pastore A, Gaeta LM, Tozzi G, Comparcola D, Sartorelli MR, Marcellini M, Bestini E & Piemonte F (2005) Glutathione metabolism and antioxidant enzymes in patients affected by nonalcoholic steatohep- atitis. Clin Chim Acta 355, 105–111.

6 Richards RS, Wang L & Jelinek H (2007) Erythrocyte oxidative damage in chronic fatigue syndrome. Arch Med Res 38, 94–98.

Treatments with acivicin (a c-GT inhibitor) and propargyl- glycine (a cystathionine c-lyase inhibitor) were performed as described by Zalups and Lash [50] and Shinozuka et al. [51], respectively. Propargylglycine (100 mgÆkg)1 intraperi- toneally) was administered 5 h before starting diamide infu- sion, whereas acivicin was administered in two doses (each 10 mgÆkg)1 intraperitoneally) 2.5 and 1 h before starting diamide infusion. In experiments with control rats, only saline was infused.

7 Stepniewska J, Dolegowska B, Ciechanowski K, Kwiat- kowska E, Millo B & Chlubek D (2006) Erythrocyte antioxidant defense system in patients with chronic renal failure according to the hemodialysis conditions. Arch Med Res 37, 353–359.

8 Tsuru R, Hojo Y, Gama M, Mizuno O, Katsuki T & Shimada K (2006) Redox imbalance in patients with coronary artery disease showing progression of athero- sclerotic lesions. J Cardiol 48, 183–191.

For the assessment of thiols and disulfides in rat tissues, animals were anesthetized with sodium pentobarbital (60 mgÆkg)1), blood was withdrawn from abdominal aorta, and then tissues were rapidly collected. All organs were homogenized in ten volumes of ice-cold 50 mm Na+ ⁄ K+ phosphate buffer, pH 7.4, using a glass ⁄ teflon potter. Aliquots were then either immediately acidified by addition of 0.5 volumes of 20% w ⁄ v TCA for LMM-SH analyses or added with 50 mm NEM (final concentration); NEM was extracted with four volumes of dichloromethane and then disulfides were measured by HPLC after dithiothreitol reduction and mBrB labeling as described above.

9 Yeh CC, Hou MF, Tsai SM, Lin SK, Hsiao JK, Huang JC, Wang LH, Wu SH, Hou LA, Ma H et al. (2005) Superoxide anion radical, lipid peroxides and antioxi- dant status in the blood of patients with breast cancer. Clin Chim Acta 361, 104–111.

Thiol and disulfide measurements in rat tissues

10 Moriarty-Craige SE, Adkison J, Lynn M, Gensler G, Bressler S, Jones DP & Sternberg P Jr (2005) Antioxi- dant supplements prevent oxidation of cysteine ⁄ cystine redox in patients with age-related macular degeneration. Am J Ophthalmol 140, 1020–1026.

Data are expressed as means ± SD. Differences between means were evaluated using one-way analysis of variance (ANOVA). A value of P < 0.05 was considered statistically significant.

Statistics

Acknowledgements

11 Raijmakers MT, Roes EM, Zusterzeel PL, Steegers EA & Peters WH (2004) Thiol status and antioxidant capacity in women with a history of severe pre-eclamp- sia. Br J Obstet Gynaecol 111, 207–212.

12 Rossi R, Giustarini D, Milzani A & Dalle-Donne I

(2008) Cysteinylation and homocysteinylation of plasma protein thiols during ageing of healthy humans. J Cell Mol Med doi:10.1111/j.1582-4934.2008.00417.

This work was supported by grants from the Fondazi- one Monte dei Paschi di Siena, by FIRST (Fondo Interno Ricerca Scientifica e Tecnologica, University of Milan) 2007, and by the Fondazione Ariel, Centro per le Disabilita` Neuromotorie Infantili, Milan, Italy.

References

13 Williams RH, Maggiore JA, Reynolds RD & Helgason CM (2001) Novel approach for the determination of the redox status of homocysteine and other aminothiols in plasma from healthy subjects and patients with ischemic stroke. Clin Chem 47, 1031–1039.

1 Giustarini D, Dalle-Donne I, Lorenzini S, Milzani A & Rossi R (2006) Age-related influence on thiol, disulfide, and protein-mixed disulfide levels in human plasma. J Gerontol A Biol Sci Med Sci 61, 1030–1038.

14 Ookhtens M & Kaplowitz N (1998) Role of the liver in interorgan homeostasis of glutathione and cyst(e)ine. Semin Liver Dis 18, 313–329.

15 Bannai S (1984) Transport of cystine and cysteine in

mammalian cells. Biochim Biophys Acta 779, 289–306.

2 Moriarty-Craige SE & Jones DP (2004) Extracellular thiols and thiol ⁄ disulfide redox in metabolism. Annu Rev Nutr 24, 481–509.

16 Ashfaq S, Abramson JL, Jones DP, Rhodes SD,

3 Muda P, Kampus P, Zilmer M, Zilmer K, Kairane C, Ristima¨ e T, Fischer K & Teesalu R (2003) Homocyste- ine and red blood cell glutathione as indices for middle-aged untreated essential hypertension patients. J Hypertens 21, 2329–2333.

Weintraub WS, Hooper WC, Vaccarino V, Harrison DG & Quyyumi AA (2006) The relationship between plasma levels of oxidized and reduced thiols and early atherosclerosis in healthy adults. J Am Coll Cardiol 47, 1005–1011.

FEBS Journal 276 (2009) 4946–4958 ª 2009 The Authors Journal compilation ª 2009 FEBS

4956

D. Giustarini et al.

Cysteine flux under oxidative stress

17 Blanco RA, Ziegler TR, Carlson BA, Cheng PY, Park Y, Cotsonis GA, Accardi CJ & Jones DP (2007) Diur- nal variation in glutathione and cysteine redox states in human plasma. Am J Clin Nutr 86, 1016–1023.

Ames BN, Basu S et al. (2005) Biomarkers of Oxidative Stress Study. II. Are oxidation products of lipids, proteins, and DNA markers of CCl4 poisoning?. Free Radic Biol Med 38, 698–710.

18 Dalle-Donne I, Rossi R, Colombo R, Giustarini D & Milzani A (2006) Biomarkers of oxidative damage in human disease. Clin Chem 52, 601–623.

31 Manibusan MK, Odin M & Eastmond DA (2007) Postulated carbon tetrachloride mode of action: a review. J Environ Sci Health C Environ Carcinog Ecotoxicol Rev 25, 185–209.

19 Dalle Donne I, Giustarini D, Colombo R, Milzani A & Rossi R (2005) S-glutathionylation in human platelets by a thiol–disulfide exchange-independent mechanism. Free Radic Biol Med 38, 1501–1510.

20 Rossi R, Milzani A, Dalle-Donne I, Giannerini F,

32 Yildiz D, Uslu C, Cakir Y & Oztas H (2006) l-Cysteine influx and efflux: a possible role for red blood cells in regulation of redox status of the plasma. Free Radic Res 40, 507–512.

33 Ookhtens M, Mittur AV & Erhart NA (1994) Changes in plasma glutathione concentrations, turnover, and disposal in developing rats. Am J Physiol 266, R979– R988.

34 Sengupta S, Wehbe C, Majors AK, Ketterer ME,

Giustarini D, Lusini L, Colombo R & Di Simplicio P (2001) Different metabolizing ability of thiol reactants in human and rat blood: biochemical and pharmacological implications. J Biol Chem 276, 7004– 7010.

21 Rossi R, Dalle-Donne I, Milzani A & Giustarini D (2006) Oxidized forms of glutathione in peripheral blood as biomarkers of oxidative stress. Clin Chem 52, 1406–1414.

DiBello PM & Jacobsen DW (2001) Relative roles of albumin and ceruloplasmin in the formation of homocystine, homocysteine–cysteine-mixed disulfide, and cystine in circulation. J Biol Chem 276, 46896– 46904.

22 Di Semplicio P, Giannerini F, Giustarini D, Lusini L & Rossi R (1998) The role of cysteine in the regulation of blood glutathione–protein mixed disulfides in rats trea- ted with diamide. Toxicol Appl Pharmacol 148, 56–64.

35 Mosharov E, Cranford MR & Banerjee R (2000) The quantitatively important relationship between homocysteine metabolism and glutathione synthesis by the transsulfuration pathway and its regulation by redox changes. Biochemistry 39, 13005–13011. 36 Finkelstein JD (2000) Pathways and regulation of

23 Giannerini F, Giustarini D, Lusini L, Rossi R & Di Semplicio P (2001) Responses of thiols to an oxidant challenge: differences between blood and tissues in the rat. Chem Biol Interact 134, 73–85.

homocysteine metabolism in mammals. Semin Thromb Hemost 26, 219–225.

37 Vitvitsky V, Mosharov E, Tritt M, Ataullakhanov F &

24 Kosower NS, Kosower EM, Wertheim B & Correa WS (1969) Diamide, a new reagent for the intracellular oxidation of glutathione to the disulfide. Biochem Biophys Res Commun 37, 593–596.

Banerjee R (2003) Redox regulation of homo- cysteine-dependent glutathione synthesis. Redox Rep 8, 57–63.

38 Martı´ n JA, Pereda J, Martı´ nez-Lo´ pez I, Escrig R,

25 Giustarini D, Dalle-Donne I, Colombo R, Milzani A & Rossi R (2003) An improved HPLC measurement for GSH and GSSG in human blood. Free Radic Biol Med 35, 1365–1372.

26 Giustarini D, Milzani A, Dalle-Donne I & Rossi R

Miralles V, Pallardo´ FV, Vin˜ a JR, Vento M, Vin˜ a J & Sastre J (2007) Oxidative stress as a signal to up-regu- late gamma-cystathionase in the fetal-to-neonatal transi- tion in rats. Cell Mol Biol 53, 1010–1017.

39 Liu RM, Shi MM, Giulivi C & Forman HJ (1998)

(2008) Red blood cells as a physiological source of glu- tathione for extracellular fluids. Blood Cells Mol Dis 40, 174–179.

Quinones increase c-glutamyl transpeptidase expression by multiple mechanisms in rat lung epithelial cells. Am J Physiol 274, L330–L336.

27 Smith TK, Ikeda Y, Fujii J, Taniguchi N & Meister A (1995) Different sites of acivicin binding and inactiva- tion of c-glutamyl transpeptidases. Proc Natl Acad Sci USA 92, 2360–2364.

40 Christensen HN (1990) Role of amino acid transport and countertransport in nutrition and metabolism. Physiol Rev 70, 43–77.

41 Burdo J, Dargusch R & Schubert D (2006) Distribution

28 Cresenzi CL, Lee JI & Stipanuk MH (2003) Cysteine is the metabolic signal responsible for dietary regulation of hepatic cysteine dioxygenase and glutamate cysteine ligase in intact rats. J Nutr 133, 2697–2702.

29 Griffiths HR, Møller L, Bartosz G, Bas A, Bertoni-

) in the of the cystine ⁄ glutamate antiporter system xc brain, kidney, and duodenum. J Histochem Cytochem 54, 549–557.

42 Dun Y, Mysona B, Van Ells T, Amarnath L, Ola MS,

Freddari C, Collins A, Cooke M, Coolen S, Haenen G, Hoberg AM et al. (2002) Biomarkers. Mol Aspects Med 23, 101–208.

30 Kadiiska MB, Gladen BC, Baird DD, Germolec D, Graham LB, Parker CE, Nyska A, Wachsman JT,

Ganapathy V & Smith SB (2006) Expression of the cys- tine-glutamate exchanger (xc-) in retinal ganglion cells and regulation by nitric oxide and oxidative stress. Cell Tissue Res 324, 189–202.

FEBS Journal 276 (2009) 4946–4958 ª 2009 The Authors Journal compilation ª 2009 FEBS

4957

D. Giustarini et al.

Cysteine flux under oxidative stress

43 Aebi S & Lauterburg BH (1992) Divergent effects of intravenous GSH and cysteine on renal and hepatic GSH. Am J Physiol 263, R348–R352.

48 Noiva R (1999) Protein disulfide isomerase: the multi- functional redox chaperone of the endoplasmic reticu- lum. Semin Cell Dev Biol 10, 481–493.

44 Jones DP (2002) Redox potential of GSH ⁄ GSSG cou- ple: assay and biological significance. Methods Enzymol 348, 93–112.

49 Go YM & Jones DP (2005) Intracellular proatherogenic events and cell adhesion modulated by extracellular thiol ⁄ disulfide redox state. Circulation 111, 2973–2980.

50 Zalups RK & Lash LH (1997) Depletion of

45 Choi YB & Lipton SA (2000) Redox modulation of the NMDA receptor. Cell Mol Life Sci 57, 1535–1541. 46 Finkel T & Holbrook NJ (2000) Oxidants, oxidative

glutathione in the kidney and the renal disposition of administered inorganic mercury. Drug Metab Dispos 25, 516–523.

stress and the biology of ageing. Nature 408, 239–247. 47 Kamata H, Shibukawa Y, Oka SI & Hirata H (2000) Epidermal growth factor receptor is modulated by redox through multiple mechanisms. Effects of reduc- tants and H2O2. Eur J Biochem 267, 1933–1944.

51 Shinozuka S, Tanase S & Morino Y (1982) Metabolic consequences of affinity labeling of cystathionase and alanine aminotransferase by l-propargylglycine in vivo. Eur J Biochem 124, 377–382.

FEBS Journal 276 (2009) 4946–4958 ª 2009 The Authors Journal compilation ª 2009 FEBS

4958