This article appeared in a journal published by Elsevier. The attachedcopy is furnished to the author for internal non-commercial researchand education use, including for instruction at the authors institution

and sharing with colleagues.

Other uses, including reproduction and distribution, or selling orlicensing copies, or posting to personal, institutional or third party

websites are prohibited.

In most cases authors are permitted to post their version of thearticle (e.g. in Word or Tex form) to their personal website orinstitutional repository. Authors requiring further information

regarding Elsevier’s archiving and manuscript policies areencouraged to visit:

http://www.elsevier.com/authorsrights

Author's personal copy

Original Contribution

S-D-Lactoylglutathione can be an alternative supplyof mitochondrial glutathione

Tatiana Armeni a,n, Laura Cianfruglia a, Francesco Piva a, Lorena Urbanelli b,Maria Luisa Caniglia a, Armanda Pugnaloni c, Giovanni Principato a

a Department of Clinical Sciences, Section of Biochemistry, Biology, and Physics, Università Politecnica delle Marche, Ancona, Italyb Department of Experimental Medicine and Biochemical Sciences, Università di Perugia, Perugia, Italyc Department of Molecular and Clinical Sciences, Università Politecnica delle Marche, Ancona, Italy

a r t i c l e i n f o

Article history:Received 16 July 2013Received in revised form5 December 2013Accepted 5 December 2013Available online 12 December 2013

Keywords:S-D-LactoylglutathioneMitochondriaGlutathioneGlyoxalase II enzymeGlyoxalase systemThiols

a b s t r a c t

The mitochondrial pool of GSH (glutathione) is considered the major redox system in maintaining matrixredox homeostasis, preserving sulfhydryl groups of mitochondrial proteins in appropriate redox state, indefending mitochondrial DNA integrity and protecting mitochondrial-derived ROS, and in defendingmitochondrial membranes against oxidative damage. Despite its importance in maintaining mitochondrialfunctionality, GSH is synthesized exclusively in the cytoplasm and must be actively transported intomitochondria. In this work we found that SLG (S-D-lactoylglutathione), an intermediate of the glyoxalasesystem, can enter the mitochondria and there be hydrolyzed from mitochondrial glyoxalase II enzyme toD-lactate and GSH. To demonstrate SLG transport from cytosol to mitochondria we used radiolabeledcompounds and the results showed two different kinetic curves for SLG or GSH substrates, indicatingdifferent kinetic transport. Also, the incubation of functionally and intact mitochondria with SLG showedincreased GSH levels in normal mitochondria and in artificially uncoupled mitochondria, demonstratingtransport not linked to ATP presence. As well mitochondrial-swelling assay confirmed SLG entrance intoorganelles. Moreover we observed oxygen uptake and generation of membrane potential probably linked toD-lactate oxidation which is a product of SLG hydrolysis. The latter data were confirmed by oxidation of D-lactate in mitochondria evaluated by measuring mitochondrial D-lactate dehydrogenize activity. In this workwe also showed the presence of mitochondrial glyoxalase II in inter-membrane space and mitochondrialmatrix and we investigated the role of SLG in whole cells. In conclusion, this work showed new alternativesources of GSH supply to the mitochondria by SLG, an intermediate of the glyoxalase system.

& 2013 Elsevier Inc. All rights reserved.

Introduction

GSH (reduced glutathione) is present in cells of all organismsand it is not homogeneously distributed among cellular compart-ments; in fact, in confluent cells, most cellular GSH is found in thecytoplasm (80–85%) whereas in mitochondria an independentlycontrolled redox pool is present at 10–15% of total GSH [1].

Since cellular respiration occurs in mitochondria, they are themajor source of ROS (reactive oxygen species) and, in turn, theyare the organelles most exposed to damage by oxygen radicals [2].Although mitochondria are exposed to this constant generation ofoxidant species, they remain functional thanks to the presence ofan efficient antioxidant defense system [3]. Especially, the mainredox buffering system in mitochondria is constituted of glu-tathione, glutaredoxin, and thioredoxin [4]. In particular theGSH/GSSG couple is considered the most important redox systemin maintaining mitochondria matrix redox homeostasis, in preser-ving the redox state of mitochondrial proteins, and in defendingthe integrity of mitochondrial DNA [5–7]. mGSH (mitochondrialGSH) is also the primary defense against oxidative damage tomitochondrial membranes by ensuring the reduction of hydroper-oxides on phospholipids and other lipidic peroxides. Moreover,because mitochondria lack catalase, the neutralization of thehydrogen peroxide is mainly assigned to mGSH [8–10]. Throughits involvement in the metabolism of oxidant species and main-tenance of the appropriate redox state and of the mitochondrial

Contents lists available at ScienceDirect

journal homepage: www.elsevier.com/locate/freeradbiomed

Free Radical Biology and Medicine

0891-5849/$ - see front matter & 2013 Elsevier Inc. All rights reserved.http://dx.doi.org/10.1016/j.freeradbiomed.2013.12.005

Abbreviations: AA, antimycin; ASC, ascorbate; BSA, bovine serum albumin; CN-1,cyanide; D-LDH, D-lactate dehydrogenase; DCIP, dichloroindophenol; DTNB,5,50-dithiobis(2-nitrobenzoic) acid; JC1, nernstian dye 5,5′,6,6′-tetrachloro-1,1′,3,3′-tetraethylbenzimidazolcarbocyanine iodide; Δψ, membrane potential; FCCP,carbonyl cyanide p-trifluoromethoxyphenylhydrazone; Glo I, glyoxalase I; Glo II,glyoxalase II; GSH, glutathione; mGSH, mitochondrial glutathione; MG, methyl-glyoxal; P/O ratio, the ratio of moles of ATP synthesized to moles of oxygen atomsreduced to water during oxidative phosphorylation; RLM, rat liver mitochondria;ROS, reactive oxygen species; ROT, rotenone; SLG, S-D-lactoylglutathione; TMPD,N,N,N0 ,N0-tetramethyl-p-phenylenediamine; Tx100, Triton X-100

n Corresponding author. Fax: þ39 071 2204609.E-mail address: t.armeni@univpm.it (T. Armeni).

Free Radical Biology and Medicine 67 (2014) 451–459

Author's personal copy

protein sulfhydryl groups, mGSH is critical in maintaining orga-nelle functionality and therefore cell survival [11,12]. It is impor-tant to underline, as previously shown, that selective depletion ofmitochondrial GSH induced necrosis and apoptosis, while deple-tion of cytoplasmatic GSH is not a threat for cell survival ifmitochondrial GSH level is sustained [13–15].

Since mitochondria do not have the enzymes for GSH synthesisit must be transported into mitochondria from cytosol. Thecharged nature of the GSH molecule suggests that it cannotpassively diffuse across the mitochondrial inner membrane. Infact, GSH is a negatively charged molecule at physiological pH andthe matrix space is negative compared to the cytoplasm so GSHmust be transported actively or be exchanged for the other anion[16,17]. To date, two anion carriers in the membrane (dicarbox-ylate, DIC Slc25a10; and 2-oxoglutarate, OGC Slc25a11) are shownto play a role in the mitochondrial uptake of GSH from thecytoplasm [18,19]. DIC and OGC together accounted for only45–50% of total glutathione uptake in liver mitochondria, com-pared to 70–80% contribution in kidney mitochondria [13,20,21].These studies suggest tissue-specific differences in the role ofcarriers in glutathione uptake into mitochondria and also raise thepossibility of the existence of some yet “unknown” glutathionetransporters in mitochondria.

Numerous approaches can and have been used to alter con-centrations of GSH in cells. The most common approach toincreasing cellular GSH concentrations is to incubate cells withGSH, but as it cannot cross the cell membrane, it is first brokendown into amino acids and then resynthesized in the cell by theconsecutive actions of gamma-glutamylcysteine (GCS) and GSHsynthetases. However, since GCS is inhibited by feedback fromGSH, achievable levels of GSH inside cells have an upper limit.Many researchers have proposed the use of novel molecules ableto raise the intracellular GSH level. These pro-GSH molecules canbe either a GSH carrying a hydrophobic group to make cellularentry easier or a source of thiol groups from which GSH issynthesized intracellularly (N-acetylcysteine, GSH monoethylester,S-acetylglutathione, N-butanoyl GSH, etc.). But if one wants tospecifically increase GSH content in a particular subcellular orga-nelle (e.g., the mitochondria), providing the extracellular mediumwith GSH or GSH precursors will not be very effective [22–24]. Forthese reasons it is important to study the mitochondrial supply ofGSH and in this work we considered SLG (S-D-lactoylglutathione),as a candidate molecule to supply GSH in mitochondria.

SLG, an intermediate of the glyoxalase system, is a GSH estersynthesized in living cells. The glyoxalase pathway is involved incellular detoxification of α-ketoldehydes and includes Glo I (EC4.4.1.5., GlxI, glyoxalase I, lactoyglutathione lyase) that catalyzesthe formation of SLG from hemithioacetal (MeCOCH (OH)-SG)formed nonenzymatically from MG (methylglyoxal) and GSH,and Glo II (EC 3.1.2.6., GlxII, glyoxalase II, hydroxyacylglutathionehydrolase) that catalyzes the hydrolysis of SLG in D-lactate,regenerating the GSH consumed in the first reaction [25–27]. Inthis study we hypothesized SLG entry in mitochondria fromcytosol and its hydrolysis by mitochondrial Glo II with consequentGSH release into mitochondria. The aim of this work was toestablish if GSH transport by means of SLG from cytoplasm intomitochondria can be alternative to the way already described.

Materials and methods

Chemicals

All chemical reagents were obtained by Sigma Aldrich (Sigma,St Louis, MO, USA). Rotenone and FCCP were dissolved in ethanol.S-D-Lactoylglutathione was synthesized and purified as described

[28]. [Glycine-2-3H]glutathione (specific activity: 20–50 Ci(740–1850 GBq)/mmol) was purchased from PerkinElmer (Watham,MA, USA). S-Lactoylglutathione labeled with [glycine-2-3H]glu-tathione was synthesized in a mixture reaction containing MG and5 UI Glo I in buffer potassium phosphate 75 mM, pH 6.8. S-Lactoyl-glutathione labeled with [14C]methylglyoxal was synthesized in amixture reaction containing GSH and 5 UI Glo I in buffer potassiumphosphate 75 mM, pH 6.8. Carboxy-H2DCFDA (C400) and JC1 weresupplied by Invitrogen (Invitrogen, Carlsbad, CA, USA). All culturereagents were obtained by Euroclone.

Animals and mitochondrial preparation

Rat liver mitochondria (RLM) (n¼25) were isolated from adultmale Wistar rats (200–250 g body weight) supplied by INRCA(Istituto Nazionale Ricerca e Cura Anziani, Ancona, Italy). Thepurification of mitochondria from rat liver was carried out asfollows: aliquots of liver tissue (4–5 g) were homogenized 1:10(w:v) in ice-cold buffer, pH 7.5, containing 75 mM sucrose,225 mM mannitol, 1 mM EDTA, 5 mM Hepes, and 0.5 mg/ml fattyacid-free bovine serum albumin. The homogenate was centrifugedfor 10 min at 600 g at 4 1C. Sediment was discarded and thesupernatant was centrifuged for 20 min at 1200 g at 4 1C. Themitochondrial pellet obtained was washed twice and purifiedmitochondria were suspended in the same medium to obtain60–70 mg of proteins/ml. The mitochondrial protein content wasdetermined by the Lowry method [29]. This investigation wascarried out in conformity with the Guide for the Care and Use ofLaboratory Animals [30] and was approved by the local EthicsCommittee. Purity of isolated mitochondria was checked bymeasuring Glo I activity, a cytosolic marker, in the mitochondrialsuspension.

Mitoplast isolation

Mitoplasts were isolated using RLM from adult male Wistar ratsisolated in medium as described above. The final mitochondrialpellet was suspended in 5 ml of medium to which 5 ml ofdigitonin solution (6 mg/ml in medium) was added. The mixturewas shaken at 0 1C for 15 min. BSA (0.5%) of BSA was added to themedium. Mitoplasts collected by centrifugation (10,000 g for10 min) were subsequently suspended in this medium, centrifugedagain under the same conditions, and finally resuspended inmaximum volume of 1 ml. Supernatants of the first and secondcentrifugations were centrifuged (105,000 g for 60 min, 4 1C) toobtain intermembrane space and mitochondrial-outer membranefractions. The absence of intact mitochondria in the sample waschecked by assaying adenylate kinase (EC 2.7.4.3); a markerenzyme of the intermembrane space and mitoplast integrity wereconfirmed by verifying the lack of glutamate dehydrogenase(EC 1.4.1.3) activity in the suspension.

LSG mitochondrial transport

In order to follow the entry of SLG or GSH into mitochondriaafter incubation of substrates, the solution with radioactive com-pounds was diluted in 10 ml H2O. SLG quantity was determinedin mitochondrial homogenates by measuring the radioactivityafter 1, 2, 5, 10, and 20 min of incubation. Isolated mitochondriawere in state 1 (without additional substrates) and three differentmitochondrial suspensions were combined with either 14CLSG or3HLSG or 3HGSH. Radioactivity of the samples was determinedusing a liquid scintillation counter (Beckman LS-6500). Radio-activity of each sample was calculated using the calibrationof sample quenching, radioactive decay, and background.The background value was the radioactivity of the hydrophilic

T. Armeni et al. / Free Radical Biology and Medicine 67 (2014) 451–459452

Author's personal copy

scintillating solution. Each experiment was repeated three timesand average values were given.

Incubation of SLG and glutathione determination

Functionally intact mitochondria were incubated with SLG(1 mM) or GSH (1 mM) for 15 min in respiration medium andthen washed twice to 6000 g at 5 1C. Before incubation withsubstrates the mitochondria were untreated (state 1) or treatedwith 1 mM ADP (state 2) or with 16 mM FCCP (uncoupled), washed,and then analyzed. For sample preparation, mitochondrial frac-tions were immediately deproteinized in ice-cold perchloric acidand centrifuged (15,000 g for 15 min at 4 1C), and supernatants wereused for GSH determinations. Perchloric acid was removed bydeproteinized samples by neutralisation with potassium carbonate.For GSSG determination one aliquot was treated with 2-vinilpiridine(5%) and triethanolamine (1%) to mask the GSH present in the extract.

Total GSH was evaluated using the DTNB glutathione reductaseassay developed by Akerboom and Sies [31]. Briefly, the absor-bance at 412 nm generated by DTNB was measured and then totalGSH concentration was calculated using a calibration line obtainedwith known concentrations of GSH. Pellets were dissolved in 1 Msodium hydroxide and used for protein determinations using BSA(bovine serum albumin) as standard [29]. The amount of totalglutathione is reported as nanomole per minute per milligram ofproteins of the mitochondrial fraction.

Enzyme activities

Mitochondrial pellets were dissolved in buffer containing 0.5%(v:v) TX100 (Triton X-100) and centrifuged to 6000 g for 15 min,and the supernatant was utilized for enzymatic determinations.Determinations of enzymatic activities were carried out at aconstant temperature of 20 1C.

Glo II (EC 3.1.2.6) activity was determined at 412 nm using0.9 mM SLG as substrate in 100 mM MOPS buffer, pH 7.2, contain-ing 0.2 mM DTNB [32]. Enzyme activity is given as micromole perminute per milligram proteins.

Glo I (EC 4.4.1.5) activity was determined at 240 nm [33] using1 mM GSH/methylglyoxal hemithioacetal as substrate in 100 mMsodium phosphate buffer, pH 6.8. Hemithioacetal is generatedin situ by preincubation of MG with GSH in sodium phosphatebuffer at 37 1C and this step is essential for avoiding conditionswhere the formation of hemithioacetal is rate limiting [34].

D-LDH (D-lactate dehydrogenase) assay was performed photo-metrically at 600 nm, as described in [35], at 25 1C. Briefly themitochondrial sample was incubated for 2 min in 2 ml of astandard medium consisting of 0.2 mM sucrose, 10 mM KCl,20 mM Hepes/Tris, pH 7.2, 1 mM MgCl2 in the presence of 30 mMphenazine methosulfate and 50 mM DCIP (dichloroindophenol).D-LDH activity was assayed by measuring the decrease in A600 dueto DCIP reduction that occurs when 15 mM D-lactate is added. Theactivity was expressed as nanomole of DCIP reduced per minuteper milligram of proteins.

Swelling experiments

Mitochondrial swelling was measured spectrophotometrically(Beckman DU-640, Fullerton, CA, U.S.A.) by monitoring thedecrease in absorbance at 540 nm 25 1C for 10 min, similar topreviously described methods [36]. Mitochondria (1 mg of pro-tein) were incubated in 2 ml of medium containing 210 mMmannitol, 70 mM sucrose, 3 mM Hepes (pH. 7.4), 10 mM succinate,and 1 μM rotenone. SLG (0.1 M) was added after 30 s, and thedecrease in the absorbance continuously recorded from 2 to 5 min.

Sucrose (0.25 M) was used as negative control and 0.1 M ammo-nium phosphate (NH4Pi) was used as positive control.

Oxygen-uptake studies and membrane potential

Oxygen-uptake measurements were carried out at 25 1C in1.5 ml of a medium consisting of 210 mM mannitol, 70 mMsucrose, 0.1 mM EDTA, 20 mM Tris/HCl, pH 7.4, 3 mM MgCl2,5 mM KH2PO/K2HPO4 by means of a Gilson 5/6 oxygraph using aClark electrode.

Membrane potential, by safranine O response, was monitoredas described in [37]. Measurements were carried out at 25 1C in2.0 ml of RLM isolation medium containing 9.6 M safranine O and1 mg of mitochondrial protein.

Cell culture and treatment

Mouse embryonic fibroblast (MEF) cells were cultured inDulbecco’s modified Eagle’s medium (DMEM) high glucose withL-glutamine medium supplemented with 10% fetal bovine serum,100 U/ml penicillin, and 100 mg/ml streptomycin at 37 1C in ahumidified atmosphere containing 5% (v/v) CO2. For the treatmentcells were subcultured in 6-well plates at a concentration of3�105 cell/well. For induced stress cells were incubated with500 mM H2O2 in PBS for 30 min, washed in PBS, and thenincubated with GSH and SLG-encapsulated liposomes for 2 h. Afterincubation mitochondrial potential and radical levels from flowcytometry were analyzed (Beckman Coulter Epics XL) using JC-1and carboxy-H2DCFDA as probes, respectively.

Liposome preparation

Lipids used for liposome preparation were EGG-PC (phophati-dylcholine) (Sigma-Aldrich) Liposomes were prepared by extru-sion technique [38] using a small extruder apparatus (Avanti PolarLipids Inc. (Alabaster, AL). Prior to extrusion, a lipid film wasprepared in a glass vial by mixing the lipid-soluble stock solutionin chloromethane, with subsequent evaporation of the solventunder nitrogen stream. The lipid film was hydrated with a solutioncontaining SLG or GSH 150 mM. Solubilized large multilamellarvesicles were then extruded through a polycarbonate membranewith 100-nm-size pores (Avanti Polar Lipids Inc. Alabaster, AL) at50 1C.

In cell experiments, antioxidant liposomes containing 150 mMGSH or SLG encapsulated were diluted 1:15 in cultured serum-freemedium, thus providing a final concentration of 10 mM.

Reactive oxygen species (ROS) detection

Intracellular ROS levels were detected by flow cytometry usingcarboxy-H2DCFDA (C400), as probe, according to the manufac-turer’s instructions. Within cells, carboxy-H2DCFDA is hydrolyzedby esterase to form a nonfluorescent polar derivative, which isoxidized by intracellular ROS to form the fluorescent compound20,70-dichlorofluorescein (DCF), whose maximum emission is at520 nm. Cells were treated or not with 500 mM H2O2 for 30 min atroom temperature, and then they were incubated with SLG orGSH-encapsulated liposomes (final concentration 10 mM) for 2 h.Cells were trypsinized, washed twice with cold PBS and incubatedfor 30 min at 37 °C in prewarmed PBS containing the probe at aworking concentration of 10 µM. Fluorescence of labeled cells wasmeasured on a Beckman Coulter EPICS XL flow cytometer using anexcitation wavelength of 488 nm.

T. Armeni et al. / Free Radical Biology and Medicine 67 (2014) 451–459 453

Author's personal copy

Mitochondrial membrane potential assay

Mitochondrial membrane potential (ΔΨm) was analyzed usingthe JC-1 probe (Invitrogen). This cyanine dye is a lipophilic cationthat is able to selectively enter the mitochondria. The probe existsin a monomeric form emitting at 530 nm (green) on excitation at488 nm. However, depending on ΔΨm, JC-1 is able to form J-aggregates that are associated with a large shift in emission to590 nm (red/orange). Thus, the color of the dye changes reversiblyfrom green to orange as the mitochondrial membrane becomesmore polarized (above values 80–100 mV). Cells were harvestedand resuspended in diluted 0.3 ml JC-1 in complete medium (finalconcentration 4 μg/ml) and incubated in the dark for 15 min at37 1C. Cells were finally washed in PBS and analyzed on theCoulter flow cytometer using FL-1 and FL-2 emissions detectedby PMT set at 572 and 580 V, respectively. Mitochondrial depolar-ization was evaluated in terms of green/yellow fluorescence ratio.

Statistical analysis

For ethical reasons we were able to use a total of 25 rats for all theexperiments. Results are reported as means of at least 3 separateexperiments 7 standard deviation. Statistical comparison of differ-ences among groups of datawas carried out using one-way analysis ofvariance (ANOVA) on various biological parameters to test differencesbetween various groups. P values r 0.05 were considered statisticallysignificant and P values r 0.01 were considered highly significant.Homogeneity of variance was tested by Cochran C and mathematicaltransformation applied if necessary; post hoc comparison (Newman-Keuls) was used to discriminate between means values.

Results

SLG mitochondrial transport

In order to assess if cytoplasmic SLG was able to enter rat livermitochondria and, in that case, for how long its metabolites wereretained into these organelles we used radiolabeled compounds.By using 14CLSG and 3HLSG, two different marked substrates wereobtained from SLG to evaluate product metabolism of SLG hydro-lysis. GSH was then marked with 3HGSH to compare its kinetictransport versus that of 3HLSG.

Both mitochondria incubated with H-labeled SLG (3HLSG) andwith C-labeled SLG (14CLSG) showed a considerable radioactivityincrease that reached a maximum value in 10 and 15 min fromincubation, respectively, for 14CLSG and 3HLSG. Surprisingly, C-labeled SLG (14CLSG) mitochondria rapidly lost radioactivitywithin the following few minutes, unlike H-labeled SLG (3HLSG)mitochondria which kept their radioactivity. The distinct radio-activity kinetic curves of the two used radioisotopes indicated thatthe marked SLG was hydrolyzed inside mitochondria and itsproducts were differently retained (Fig. 1a).

In order to compare the kinetic transport of SLG and GSH,mitochondria were incubated with 3HLSG and 3HGSH. As shown inFig. 1b, the kinetic curves highlight a different uptake velocity andalso a diverse amount of compound stored. In particular, althoughboth compounds reach maximum cpm (count per minute) valuesin 15 min, 3HGSH arrives at about 2500 cpm whereas 3HLSGarrives at about 1650 cpm.

Mitochondria SLG incubation determines mitochondrial swellingand increases mitochondrial GSH levels

In order to demonstrate if SLG could enter mitochondria,swelling experiments were carried out. RLM was suspended in

0.1 M SLG and spontaneous swelling occurred which demon-strated SLG entry into isolated intact mitochondria. In order toprove the correctness of the experimental procedure, we used anegative control with 0.25 M isotonic sucrose and a positivecontrol with 0.1 M NH4Pi (ammonium phosphate) that respec-tively showed no swelling and spontaneous mitochondrial swel-ling (Fig. 2a).

To determine SLG contribution in mitochondrial GSH levels wemeasured total GSH content in not-incubated mitochondria (con-trol), in mitochondria incubated with 1 mM SLG, and in mitochon-dria incubated with GSH. The experiments were performed withmitochondria maintained in state 1 (without additional sub-strates), in state 2 (with additional ADP), and uncoupled (withadditional FCCP) (Fig. 2b). The external addition of 1 mM SLG inmitochondria in state 1 caused a significant increase in mitochon-drial GSH levels (þ 40%) compared to the control (not incubatedmitochondria). Also incubation with 1 mM GSH determinedsignificant increases of mitochondrial GSH levels (þ 55%) comparednot only to the control but also to the sample treated with SLG. TheATP addition did not alter previous results suggesting a transport ofthe two compounds (SLG and GSH) not ATP dependent. On the otherhand, in uncoupled mitochondria (treated with FCCP) all GSH levelssignificantly decreased compared to mitochondria in state 1. Never-theless, uncoupled mitochondria treated with both SLG and GSHsignificantly increased GSH levels compared to the uncoupled controlby 56 and 60%, respectively (Fig. 2b). GSSG (oxidized glutathione)was also evaluated in mitochondria at state 1 after addition of 1 mMGSH and 1 mM SLG. Data did not show statistical differencesbetween the groups (GSSG level as nmol/mg proteins: 0.4570.06,0.4370.05, 0.4070.03, respectively for control, mitochondria

Fig. 1. (a) Intact RLM was incubated with S-D-lactoylglutathione radiolabeled in3HSLG or in 14CSLG in order to follow the entry of SLG and its metabolism inmitochondria. Reported results are mean values 7 standard deviation (n¼3).The asterisk indicates significant differences (*Po0.05) between the correspon-dence points of the two curves. (b) Intact RLM was incubated with S-D-lactoylglu-tathione radiolabeled in 3HSLG or with GSH radiolabeled in 3HGSH in order tocompare kinetic transport of the two considered compounds. Reported results aremean values 7 standard deviation (n¼3). The asterisk indicates significantdifferences (*Po0.05) between the correspondence points of the two curves.cpm: counts per minute.

T. Armeni et al. / Free Radical Biology and Medicine 67 (2014) 451–459454

Author's personal copy

incubated with 1 mM SLG, or mitochondria incubated with GSH).These data showed that mitochondria did not suffered to oxidativestress after incubation with SLG and that most glutathione derivedfrom SLG remained in the reduced form within mitochondria.

Since it is known that also cytosolic Glo II, besides themitochondrial one, hydrolyzes SLG to yield GSH and D-lactate, itwas necessary to exclude the presence of the latter in ourincubation experiments. Therefore we measured Glo II activity inthe medium of purified mitochondria and confirmed enzymeabsence. Since this assay is not able to distinguish cytosolic andmitochondrial Glo II forms, our test proved not only the absence ofthe cytosolic form but also mitochondrial intactness.

Oxygen uptake and membrane potential by RLM added with SLG

In order to verify that SLG hydrolysis products were respiratorysubstrates, we measured mitochondrial oxygen consumption byoxygraphic analysis. We started to acquire oxygen uptake ofmitochondria suspended in respiration medium, then we addedSLG (5 mM) and recorded the oxygen uptake at state 4, subse-quently we added ADP (0.1 mM) and registered the oxygen uptakeat state 3, and finally we added FCCP to uncoupled mitochondria.Oxygen uptake was found with a respiratory control index (state3 rate/state 4 rate ratio) of 3.5 and a P/O ratio (the ratio of mol ofATP synthesized to mol of oxygen atoms reduced to water duringoxidative phosphorylation) of 2.3 (Fig. 3a). These experimentsconfirmed the hypothesis that one SLG product (expectedD-lactate) consisted of respiratory substrates and that SLG washydrolyzed in mitochondria. We compared oxygen uptake of SLG,D-lactate, and succinate and we repeated each experiment sixtimes. No significant differences were found in P/O ratios

calculated for SLG (2.370.2), D-lactate (2.170.2), and succinate(1.9870.2) (tracks not shown). These P/O values, as close to 2,suggested that SLG metabolites, such as the already knownD-lactate and succinate, could induce a flavine reduction. In orderto give value to this indication, we performed further oxygenuptake assays using electron respiratory chain inhibitors. Sincerotenone, a Complex I inhibitor, completely blocked oxygen uptakecaused by the substrate glutamatemalate pair (5 mM each) but didnot prevent the oxygen uptake due to SLG administration, there-fore oxidation of the compound produced by SLG hydrolysisinduced flavine reduction. In confirmation of the latter hypothesis,the addition of antimycin A, a powerful inhibitor of Complex III,completely inhibited oxidation in state 4. Oxygen consumptionwas restored by adding ascorbate (5 mM) plus TMPD (0.5 mM),but completely inhibited by cyanide (1 mM), a potent inhibitor ofComplex IV (Fig. 3a).

Since respiration induction is expected to generate membranepotential (Δψ), we evaluated the mitochondrial Δψ alterationdue to the SLG addiction, by using safranin O as Δψ probe. InFig. 3b we show that SLG (5 mM) generated Δψ increase, asmonitored by a decrease of safranin O absorbance. Δψ generationwas prevented by antimycin A and completely abolished bythe uncoupler FCCP. In another experiment, succinate (5 mM)

Fig. 2. Externally added SLG to intact RLM. (a) Mitochondrial swelling of sucrose(used as negative control), of 0.1 M ammonium phosphate (used as positivecontrol), of 0.1 M D-lactate, and of 0.1 M SLG. In the table the means 7 standarddeviation of the initial swelling rates are reported (n¼4). Mitochondrial swellingwas monitored as described under Materials and methods. (b) Total GSH level wasdetermined in mitochondrial homogenate after incubation of external 1 mM GSHor 1 mM SLG on intact mitochondria in state 1; with added 1 mM ADP or withadded 16 mM FCCP was determined. The amount of GSH was normalized for proteincontent. Reported results are mean values 7 standard deviation (n¼4). Differentletters indicate significant differences between the means of each group (NewmanKeuls post hoc) Po0.01.

Fig. 3. Oxygen uptake and membrane potential by intact RLM added with SLG.(a) Oxygen uptake by RLM (1 mg protein) added with SLG was measured as afunction of time. Arrows indicate the following additions: S-D-lactoylglutathione(5 mM SLG), ADP (1 mM), FCCP (16 mM), rotenone (ROT) was added 1 min beforeSLG, antimycin A (AA) that blocked oxygen uptake, ascorbate þ TMPD (ASC 5 mMþ TMPD 0.5 mM) that restored oxygen uptake, and cyanide (CN- 1 mM) thatcompletely inhibited oxygen consumption. Numbers along the curves are rates ofoxygen uptake expressed as natom O/min � mg proteins. (B) The ability of SLG togenerate membrane potential (Δψ) from the mitochondrial matrix was checked byusing safranine as Δψ probe. Δψ generation was found as a result of adding SLG(5 mM) to RLM, as monitored by a decrease of safranine absorbance. Δψ generationwas prevented by antimycin A (SLGþAA). In the same experiment, both succinate(5 mM) and pyruvate (5 mM) were monitored. Succinate generated Δψ at a higherrate than SLG in an inhibited way by antimycin. Pyruvate generated Δψ at a lowerrate than SLG in an inhibited way by rotenone.

T. Armeni et al. / Free Radical Biology and Medicine 67 (2014) 451–459 455

Author's personal copy

generated Δψ slightly greater than SLG and this potential wasabolished by antimycin A addiction, as in the previous assay.

Mitochondrial LSG metabolism and GII localization

In order to prove the hydrolysis of SLG in mitochondria toD-lactate and GSH, D-lactate dehydrogenase activity was mea-sured in mitochondria after external addition of SLG. Fig. 4a showsthat D-lactate oxidation is directly dependent on the concentrationof external SLG and saturation kinetics were found with Km andVmax values respectively of 2.8 mM and 48 nmol/min per mg ofproteins. This experiment demonstrated that D-lactate was aproduct of SLG hydrolysis.

Since SLG hydrolysis is catalyzed from Glo II we measured itsenzyme activity in intact mitochondria, mitoplast, and mitochon-drial matrix in order to determine its localization. Since intactmitochondria did not show Glo II activity, we deduced the absenceof this enzyme in external mitochondrial membranes (Fig. 4b).The absence of contamination from cytoplasmic proteins was assuredby lack of Glo I activity (a cytosolic marker) in mitochondrialsuspension. Mitoplasts showed minimal Glo II activity (10 nmol/min/mg proteins). The mitoplast supernatant, containing intermem-brane space and mitochondrial outer membrane fraction, showed aGlo II activity of 40 nmol/min/mg proteins. The addition of Tx100(0.5%), a detergent which dissolves mitochondrial membranes, inmitoplast showed Glo II activity of 70 nmol/min/mg proteins (Fig. 4b).Therefore we proved that Glo II is mainly located in matrix and outermitochondrial inner membranes.

Intracellular ROS detection on whole cells after incubation with SLGor GSH

The antioxidant effect on ROS production was evaluated byanalyzing flow cytometry intracellular ROS levels. Cells incubatedwith H2O2 showed higher levels of ROS compared with controlsample (Fig. 5a). SLG treatment did not show significant differ-ences with H2O2-treated sample (positive control), or untreatedsample (negative control); this is important to note because SLGalone did not increase ROS production and consequently mito-chondrial oxidative stress. On the another hand, GSH treatmentconfirmed the positive antioxidant effect against ROS production;in fact, ROS levels were clearly decreased in both H2O2-untreatedcontrol and more so in H2O2-treated control (Fig. 5a)

Mitochondrial membrane potential on whole cells after incubationwith SLG or GSH

Mitochondrial membrane potential was measured in terms ofJC-1 fluorescence. The extent of depolarization was derived byevaluating a decrease in yellow fluorescence and increase in greenfluorescence (Fig. 5b, right panel). Data reported in Fig. 5b showthat H2O2-untreated samples are comparable in relation to theirmitochondrial membrane potential, independently of the presenceof antioxidant-liposomes. On the other hand, H2O2 exposure leadsto a significant decrease in ΔΨm in all samples (Po 0.01), but thepresence of SLG or GSH protects mitochondria from membranepotential depolarization (H2O2 þ SLG Po0.05; H2O2 þGSHPo 0.01).

Discussion

The glyoxalase system is a metabolic pathway found wide-spread throughout biological systems. The glyoxalase pathway isinvolved in cellular detoxification of α-ketoldehydes producedduring glycolysis and it catalyzes conversion of 2-oxoaldehydesto the corresponding 2-hydroxyacids, via intermediate S-2-hydroxyacylglutathione. The glyoxalase system consists of twoenzymes, Glo I and Glo II, and a catalytic amount of GSH ascofactor. Glo I catalyzes the formation of SLG from hemithioacetal(MeCOCH(OH)-SG) formed nonenzymatically from MG and GSH;Glo II catalyzes the hydrolysis of SLG in D-lactate, regenerating theGSH consumed in the first reaction [25] (Fig. 6). The glyoxalasesystem is essential in cell detoxification but it also has a key role inthe control of cell proliferation [39,40]. Glyoxalase enzymes aredistributed in the cytosol of all aerobic organisms and this can beexplained by their correlation to GSH and the ubiquitary glycolysisof which the glyoxalase system constitutes a parallel pathway [41].Glo I enzyme is found only in cytoplasmic space while Glo IIenzyme is found in different cellular compartments (nucleus,endoplasmatic reticulum, and mitochondria). In eukaryotes multipleforms of Glo II are found in mitochondria (both in intermembranespace and in the matrix) while only one form of cytosolic Glo IIexists, which resembles the intermembrane space form [42].Glo II is a metal-dependent β-lactamase and shows a characteristicZn2þ-binding motif, conserved in all known sequences, and a Zn(II)Fe(II) center [43,44]. The biological meaning of the presence of Glo IIand the absence of Glo I in mitochondria is still unclear.

In this work, using various experimental approaches, we showthat the intermediate of the glyoxalase system, SLG, can entermitochondria and be hydrolyzed to D-lactate with GSH release.This substrate in mitochondria may have different implications.First, SLG substrate may justify the presence of mitochondrial GloII which catalyzes the D-lactate and GSH release reaction insidethe mitochondria. D-Lactate formed by mitochondrial Glo II from

Fig. 4. (a) D-Lactate oxidation in mitochondria. SLG was added at the indicateconcentrations to mitochondria with the rate (v) of DCPI reduction expressed asnmol/min/mg proteins. Saturation kinetics were found with Km and Vmax values of2.8 mM and 48 nmol/min/mg proteins. (b) Mitochondrial Glo II localization. Glo IIactivity was determined spectrophotometrically at 412 nm following DTNB oxida-tion. Glo II activity was measured in intact mitochondria, mitoplast, intermembranespace and outer membrane, and mitochondrial matrix. Enzyme activity is given asmmol/min/mg proteins.

T. Armeni et al. / Free Radical Biology and Medicine 67 (2014) 451–459456

Author's personal copy

SLG is oxidized to pyruvate via D-LDH and as a consequenceelectrons flow to oxygen, producing energy and ATP synthesis [45].Pyruvate sits at the intersection of key pathways of energymetabolism and the fate of the newly synthesized pyruvate couldbe both mitochondrial and cytosolic. In the matrix, pyruvate canbe converted by the pyruvate dehydrogenase complex to acetyl

CoA or be converted by the pyruvate carboxylase to oxalacetate,which can enter the Krebs cycle and give reducing equivalents.Moreover, oxalacetate may be reduced to malate via malatedehydrogenase, with concomitant oxidation of NADH but in thefollowing reaction malate, by malic enzyme, may again formpyruvate with a concomitant formation of NADPH from NADPþ .

Fig. 5. (a) ROS production in MEF cells. Carboxy-H2DCFCDA (C400) probe was used to analyze ROS concentration in control and treated cells by flow cytometry. In the leftpanel geometric mean value of ROS production was reported for control cells (CTRL), cells treated with GSH (GSH) or SLG (SLG) encapsulated in liposomes, cell treated with500 mM H2O2, and cells incubated with GSH or SLG encapsulated in liposomes after treatment with 500 mM H2O2. Reported results are mean values 7 standard deviation(n¼4). Different letters indicate significant differences between groups means, Po0.01. In the right panel a cytometric graph was reported for negative control cells (blackon the left) and positive control treated with 500 mM H2O2 cells (black on the right) in comparison with cells treated with H2O2 and then incubated with GSH (light gray) orSLG (dark gray) encapsulated liposomes. Not all data are shown in the cytometric graph. (b) Mitochondrial potential analysis in MEF cells. JC-1 probe was used to analyzemitochondrial membrane depolarization (ΔΨ). In the left panel the yellow/green signal ratio was reported as arbitrary units. Reported results are mean values 7 standarddeviation (n¼4). Different letters indicate significant differences between groups means, Po0.01. The right panel shows the variations on green and yellow fluorescence inSLGþH2O- and GSHþH2O2-treated cells compared with untreated control and 500 mM H2O2-treated cells. Not all data are shown in the cytometric graph.

Fig. 6. Representative scheme of SLG metabolism. 3PGA, glyceraldehyde-3-phosphate; DHAP, dihydroxyacetone phosphate; TPI, triose-phosphate isomerase; MG,metylglyoxal; SLG, S-D-lactoylglutathione; GSH, glutathione; GloI, glyoxalase I; GloII, glyoxalase II; mGloII, mitochondrial glyoxalase II.

T. Armeni et al. / Free Radical Biology and Medicine 67 (2014) 451–459 457

Author's personal copy

These pathways yield reducing equivalents and consequently theyform mitochondrial membrane potential, hence energy for mitochon-drial activity. Our experiment showed that SLG mitochondria incuba-tion generated mitochondrial membrane potential and determinedoxygen consumption (Fig. 3a and 3b). Therefore, we presumed thatthis was derived from SLG hydrolysis and subsequent mitochondrialD-lactate metabolism. To confirm this hypothesis we measuredD-lactate dehydrogenase activity after SLG mitochondria incubationand showed a correlation between D-lactate dehydrogenase activityand SLG concentration (Fig. 4a).

However, in this study, the radioactivity assay (Fig. 1a) showedthat the SLG marked on carbon of D-lactate was metabolized andafter a few minutes exited partially from mitochondria, while theSLG marked on GSH was maintained within mitochondria, thusshowing the biological role of these molecules in mitochondria.These data highlight the different fate of SLG hydrolysis products.As a matter of fact D-lactate could be oxidized to pyruvate andmetabolized outside mitochondria too [45]. Instead GSH appearsto remain inside mitochondria to demonstrate the role of LSG insupplying the mitochondria with GSH. The kinetic transport of SLGand GSH showed different uptake curves (Fig. 1b), demonstratingdiverse carrier affinity or distinct carrier utilization.

GSH plays an important role as an antioxidant and cofactor formany enzymes involved in mitochondrial redox homeostasis, detox-ification reactions, and mitochondrial signaling. Despite the key roleof mitochondrial GSH, much less is known about the modality of itsimport in mitochondria from the cytosol [16,17]. Our data suggest anew supplementary mechanism for GSH supply to mitochondria.Incubation in intact mitochondria with radiolabeled compounds, aftersome minutes, showed radioactivity within mitochondria and thelevels were maintained in mitochondria incubated with radiolabeledSLG marked on the GSH molecule. These data show the capacity ofSLG to enter mitochondria and its capacity for supplying GSH. Ourdata were confirmed by the increase of total GSH levels in mitochon-dria after incubation of SLG (Fig. 2b). In fact mitochondrial GSHincreased, compared to the control, after mitochondria incubationwith SLG and also GSH. The increase was obtained independentlyfrom addition of ADP or FCCP. These data suggest a transport which isnot ATP dependent, but also surprisingly not dependent on mito-chondrial membrane potential. We could even suppose that themembrane potential was formed contextually from entry to LSG orGSH, but additional findings are necessary for these speculations to beconfirmed.

In this context it is also possible to justify the regulation ofcytosolic and mitochondrial Glo II. In effect Scirè et al. [46] showedthat cytosolic and mitochondrial Glo II bind specifically to nega-tively charged liposomes. The binding, however, inhibited only theenzymatic activity of the cytosolic Glo II enzyme. The binding ofcytosolic Glo II to cell membranes could lead to an increase of SLGin the cytosol. The increased cytosolic SLG concentration wouldallow SLG to be imported into mitochondria where it would behydrolyzed by mitochondrial Glo II. As a consequence of SLGhydrolysis, the concentration of GSH would increase and moreGSH would be available to perform its biological roles (e.g.,protection against oxidative stress and mitochondrial protein S-glutathionylation). Such a mechanism, involving a temporarybinding of cytosolic Glo II to membrane acidic phospholipids,could be important and could be activated in situations requiring ahigh supply of GSH in mitochondria. In addition, data recently (notyet published) obtained by our group indicate an importantregulatory role of Glo II on S-glutathionylation of some proteins,even mitochondrial ones. These results suggest an additional roleof mitochondrial Glo II, which can play a key role in increasing notonly mitochondrial GSH levels but also levels of glutathionylationon mitochondrial proteins. Therefore, in mitochondria, GSH canfunction not only as an antioxidant but also protect cysteinyl

groups of proteins from oxidation by S-glutathionylation. ProteinS-glutathionylation is an important mechanism for dynamic,posttranslational regulation of a variety of regulatory, structural,and metabolic proteins including many mitochondrial proteinssuch as malate dehydrogenase, cytochrome b, oxaloacetate, etc.[47,48]. S-Glutathionylation level depends on GSH availabilitysince it is known that GSH is required for many critical cellprocesses, and plays a particularly important role in the main-tenance and regulation of the thiol-redox status of the cell [49,50].In mitochondria GSH levels may be important for regulating S-Glutathionylation of specific proteins and modifying their activity.

Conclusions

In conclusion, in this work we demonstrate that SLG may entermitochondria by supplying it with GSH. The metabolic pathway isshown in Fig. 5. Mitochondrial Glo II enzyme catalyzes SLGhydrolysis, yielding D-lactate and GSH. D-Lactate can be oxidizedand so it generates mitochondrial membrane potential whilereduced GSH may be utilized as a mitochondrial thiol buffer.Taken together, these data provide a possible mechanism of analternative supply of GSH in mitochondria. This pathway can beconsidered in clinical situations where there is a selective deple-tion of mitochondrial GSH and may be useful in restoring thephysiological condition.

Acknowledgments

We thank Dr Andrea Scirè for scientific debate and liposomespreparation. This work was supported by Università Politecnicadelle Marche.

References

[1] Griffith, O. W.; Meister, A. Origin and turnover of mitochondrial glutathione.Proc. Natl. Acad. Sci. USA 82:4668–4672; 1985.

[2] Kaelin Jr. W. G. ROS: really involved in oxygen sensing. Cell Metab. 1:357–358;2005.

[3] Cadenas, E.; Davies, K. J. Mitochondrial free radical generation, oxidativestress, and aging. Free Radic. Biol Med. 29:222–230; 2000.

[4] Murphy, M. P. Mitochondrial thiols in antioxidant protection and redoxsignaling: distinct roles for glutathionylation and other thiol modifications.Antioxid. Redox Signal. 16:476–495; 2012.

[5] Albrecht, S. C.; Barata, A. G.; Grosshans, J.; Teleman, A. A.; Dick, T. P. In vivomapping of hydrogen peroxide and oxidized glutathione reveals chemical andregional specificity of redox homeostasis. Cell Metab. 14:819–829; 2011.

[6] Yin, F.; Sancheti, H.; Cadenas, E. Mitochondrial thiols in the regulation of celldeath pathways. Antioxid. Redox Signal. 17:1714–1727; 2012.

[7] Reliene, R.; Schiestl, R. H. Glutathione depletion by buthionine sulfoximineinduces DNA deletions in mice. Carcinogenesis 27:240–244; 2006.

[8] Han, D.; Canali, R.; Rettori, D.; Kaplowitz, N. Effect of glutathione depletion onsites and topology of superoxide and hydrogen peroxide production inmitochondria. Mol. Pharmacol. 64:1136–1144; 2003.

[9] Garcia-Ruiz, C.; Fernandez-Checa, J. C. Mitochondrial glutathione: hepatocel-lular survival-death switch. J. Gastroenterol. Hepatol. 21(Suppl. 3):S3–S6;2006.

[10] Fernandez-Checa, J. C.; Kaplowitz, N. Hepatic mitochondrial glutathione:transport and role in disease and toxicity. Toxicol. Appl. Pharmacol. 204:263–-273; 2005.

[11] Mari, M.; Morales, A.; Colell, A.; Garcia-Ruiz, C.; Fernandez-Checa, J. C.Mitochondrial glutathione, a key survival antioxidant. Antioxid. Redox Signal.11:2685–2700; 2009.

[12] Franco, R.; Cidlowski, J. A. Apoptosis and glutathione: beyond an antioxidant.Cell Death Differ. 16:1303–1314; 2009.

[13] Mari, M.; Morales, A.; Colell, A.; Garcia-Ruiz, C.; Kaplowitz, N.; Fernandez-Checa, J. C.Mitochondrial glutathione: features, regulation and role in disease. Biochim. Biophys.Acta 1830:3317–3328; 2013.

[14] Jassem, W.; Battino, M.; Cinti, C.; Norton, S. J.; Saba, V.; Principato, G.Biochemical changes in transplanted rat liver stored in University of Wiscon-sin and Euro-Collins solutions. J. Surg. Res. 94:68–73; 2000.

[15] Armeni, T.; Pieri, C.; Marra, M.; Saccucci, F.; Principato, G. Studies on the lifeprolonging effect of food restriction: glutathione levels and glyoxalaseenzymes in rat liver. Mech. Ageing Dev. 101:101–110; 1998.

T. Armeni et al. / Free Radical Biology and Medicine 67 (2014) 451–459458

Author's personal copy

[16] Bachhawat, A. K.; Thakur, A.; Kaur, J.; Zulkifli, M. Glutathione transporters.Biochim. Biophys. Acta 1830:3154–3164; 2013.

[17] Lash, L. H. Mitochondrial glutathione transport: physiological, pathologicaland toxicological implications. Chem. Biol. Interact. 163:54–67; 2006.

[18] Chen, Z.; Lash, L. H. Evidence for mitochondrial uptake of glutathione bydicarboxylate and 2-oxoglutarate carriers. J. Pharmacol. Exp. Ther. 285:608–-618; 1998.

[19] Chen, Z.; Putt, D. A.; Lash, L. H. Enrichment and functional reconstitution ofglutathione transport activity from rabbit kidney mitochondria: furtherevidence for the role of the dicarboxylate and 2-oxoglutarate carriers inmitochondrial glutathione transport. Arch. Biochem. Biophys. 373:193–202;2000.

[20] Zhong, Q.; Putt, D. A.; Xu, F.; Lash, L. H. Hepatic mitochondrial transport ofglutathione: studies in isolated rat liver mitochondria and H4IIE rat hepatomacells. Arch. Biochem. Biophys. 474:119–127; 2008.

[21] Wilkins, H. M.; Kirchhof, D.; Manning, E.; Joseph, J. W.; Linseman, D. A.Mitochondrial glutathione transport is a key determinant of neuronal suscept-ibility to oxidative and nitrosative stress. J. Biol. Chem. 288:5091–5101; 2013.

[22] Fraternale, A.; Paoletti, M. F.; Casabianca, A.; Orlandi, C.; Schiavano, G. F.;Chiarantini, L.; Clayette, P.; Oiry, J.; Vogel, J. U.; Cinatl Jr J.; Magnani, M.Inhibition of murine AIDS by pro-glutathione (GSH) molecules. Antiviral Res.77:120–127; 2008.

[23] Fraternale, A.; Paoletti, M. F.; Casabianca, A.; Nencioni, L.; Garaci, E.; Palamara, A. T.;Magnani, M. GSH and analogs in antiviral therapy. Mol. Aspects Med. 30:99–110;2009.

[24] Wu, J. H.; Batist, G. Glutathione and glutathione analogues; therapeuticpotentials. Biochim. Biophys. Acta 1830:3350–3353; 2013.

[25] Mannervik, B. Molecular enzymology of the glyoxalase system. Drug Metab.Drug Interact. 23:13–27; 2008.

[26] Thornalley, P. J. Glyoxalase I—structure, function and a critical role in theenzymatic defence against glycation. Biochem. Soc. Trans. 31:1343–1348; 2003.

[27] Ridderstrom, M.; Saccucci, F.; Hellman, U.; Bergman, T.; Principato, G.;Mannervik, B. Molecular cloning, heterologous expression, and characteriza-tion of human glyoxalase II. J. Biol. Chem. 271:319–323; 1996.

[28] Uotila, L. Thioesters of glutathione. Methods Enzymol. 77:424–430; 1981.[29] Lowry, O. H.; Rosebrough, N. J.; Farr, A. L.; Randall, R. J. Protein measurement

with the Folin phenol reagent. J. Biol. Chem. 193:265–275; 1951.[30] Guide for the care and use laboratory animals. Washingthon DC; 1985.[31] Akerboom, T. P.; Sies, H. Assay of glutathione, glutathione disulfide, and

glutathione mixed disulfides in biological samples.Methods Enzymol. 77:373–-382; 1981.

[32] Principato, G. B.; Rosi, G.; Talesa, V.; Giovannini, E.; Uotila, L. Purification andcharacterization of two forms of glyoxalase II from the liver and brain ofWistar rats. Biochim. Biophys. Acta 911:349–355; 1987.

[33] Ekwall, K.; Mannervik, B. Inhibition of yeast S-lactylglutathione lyase (glyox-alase I) by sulfhydryl reagents. Arch. Biochem. Biophys. 137:128–132; 1970.

[34] Vander Jagt, D. L.; Daub, E.; Krohn, J. A.; Han, L. P. Effects of pH and thiols onthe kinetics of yeast glyoxalase I. An evaluation of the random pathwaymechanism. Biochemistry 14:3669–3675; 1975.

[35] Chelstowska, A.; Liu, Z.; Jia, Y.; Amberg, D.; Butow, R. A. Signalling betweenmitochondria and the nucleus regulates the expression of a new D-lactatedehydrogenase activity in yeast. Yeast 15:1377–1391; 1999.

[36] Broekemeier, K. M.; Pfeiffer, D. R. Cyclosporin A-sensitive and insensitivemechanisms produce the permeability transition in mitochondria. Biochem.Biophys. Res. Commun. 163:561–566; 1989.

[37] Figueira, T. R.; Melo, D. R.; Vercesi, A. E.; Castilho, R. F. Safranine as afluorescent probe for the evaluation of mitochondrial membrane potentialin isolated organelles and permeabilized cells. Methods Mol. Biol. 810:103–117;2012.

[38] Yang, H.; Paromov, V.; Smith, M.; Stone, W. L. Preparation, characterization,and use of antioxidant-liposomes. Methods Mol. Biol. 477:277–292; 2008.

[39] Rabbani, N.; Thornalley, P. J. Glyoxalase in diabetes, obesity and relateddisorders. Semin. Cell Dev. Biol. 22:309–317; 2011.

[40] Thornalley, P. J.; Rabbani, N. Glyoxalase in tumourigenesis and multidrugresistance. Semin. Cell Dev. Biol. 22:318–325; 2011.

[41] Thornalley, P. J. Glutathione-dependent detoxification of alpha-oxoaldehydesby the glyoxalase system: involvement in disease mechanisms and antipro-liferative activity of glyoxalase I inhibitors. Chem. Biol. Interact. 111–112:137–-151; 1998.

[42] Talesa, V.; Uotila, L.; Koivusalo, M.; Principato, G.; Giovannini, E.; Rosi, G.Demonstration of glyoxalase II in rat liver mitochondria. Partial purificationand occurrence in multiple forms. Biochim. Biophys. Acta 955:103–110; 1988.

[43] Melino, S.; Capo, C.; Dragani, B.; Aceto, A.; Petruzzelli, R. A zinc-binding motifconserved in glyoxalase II, beta-lactamase and arylsulfatases. Trends Biochem.Sci. 23:381–382; 1998.

[44] Limphong, P.; McKinney, R. M.; Adams, N. E.; Bennett, B.; Makaroff, C. A.;Gunasekera, T.; Crowder, M. W. Human glyoxalase II contains an Fe(II)Zn(II)center but is active as a mononuclear Zn(II) enzyme. Biochemistry 48:5426–-5434; 2009.

[45] de Bari, L.; Atlante, A.; Guaragnella, N.; Principato, G.; Passarella, S. D-Lactatetransport and metabolism in rat liver mitochondria. Biochem. J. 365:391–403;2002.

[46] Scirè, A.; Tanfani, F.; Saccucci, F.; Bertoli, E.; Principato, G. Specific interactionof cytosolic and mitochondrial glyoxalase II with acidic phospholipids in formof liposomes results in the inhibition of the cytosolic enzyme only. Proteins41:33–39; 2000.

[47] Mieyal, J. J.; Gallogly, M. M.; Qanungo, S.; Sabens, E. A.; Shelton, M. D.Molecular mechanisms and clinical implications of reversible proteinS-glutathionylation. Antioxid. Redox Signal. 10:1941–1988; 2008.

[48] Ghezzi, P. Protein glutathionylation in health and disease. Biochim. Biophys.Acta 1830:3165–3172; 2013.

[49] Armeni, T.; Ercolani, L.; Urbanelli, L.; Magini, A.; Magherini, F.; Pugnaloni, A.;Piva, F.; Modesti, A.; Emiliani, C.; Principato, G. Cellular redox imbalance andchanges of protein S-glutathionylation patterns are associated with senes-cence induced by oncogenic H-ras. PLoS One 7:e52151; 2012.

[50] Schafer, F. Q.; Buettner, G. R. Redox environment of the cell as viewed throughthe redox state of the glutathione disulfide/glutathione couple. Free Radic. Biol.Med. 30:1191–1212; 2001.

T. Armeni et al. / Free Radical Biology and Medicine 67 (2014) 451–459 459