doi:10.1152/ajpcell.00643.2008 296:811-820, 2009. First published Feb 4, 2009;Am J Physiol Cell Physiol

Kröncke and Victoria Kolb-Bachofen Miriam M. Cortese-Krott, Christoph V. Suschek, Wiebke Wetzel, Klaus-Dietrichzinc and glutathione from hydrogen peroxide is mediated by intracellular Nitric oxide-mediated protection of endothelial cells

You might find this additional information useful...

71 articles, 43 of which you can access free at: This article cites http://ajpcell.physiology.org/cgi/content/full/296/4/C811#BIBL

including high-resolution figures, can be found at: Updated information and services http://ajpcell.physiology.org/cgi/content/full/296/4/C811

can be found at: AJP - Cell Physiologyabout Additional material and information http://www.the-aps.org/publications/ajpcell

This information is current as of February 10, 2010 .

http://www.the-aps.org/.American Physiological Society. ISSN: 0363-6143, ESSN: 1522-1563. Visit our website at a year (monthly) by the American Physiological Society, 9650 Rockville Pike, Bethesda MD 20814-3991. Copyright © 2005 by the

is dedicated to innovative approaches to the study of cell and molecular physiology. It is published 12 timesAJP - Cell Physiology

on February 10, 2010 ajpcell.physiology.org

Dow

nloaded from

Nitric oxide-mediated protection of endothelial cells from hydrogen peroxideis mediated by intracellular zinc and glutathione

Miriam M. Cortese-Krott,1,2 Christoph V. Suschek,3 Wiebke Wetzel,4 Klaus-Dietrich Kroncke,5

and Victoria Kolb-Bachofen1

1Institute of Molecular Medicine, Research Group Immunobiology, Medical Faculty of Heinrich-Heine-University ofDusseldorf, Dusseldorf; 2Department of Internal Medicine I, Cardio-Bio-Tec Research Group, and 3Department of PlasticSurgery, Hand and Reconstructive Surgery, Burn Center, Medical Faculty, RWTH Aachen University, Aachen; and 4Instituteof Biochemistry and Molecular Biology II and 5Institute of Biochemistry and Molecular Biology I, Medical Facultyof Heinrich-Heine-University of Dusseldorf, Dusseldorf, Germany

Submitted 17 December 2008; accepted in final form 28 January 2009

Cortese-Krott MM, Suschek CV, Wetzel W, Kroncke KD,Kolb-Bachofen V. Nitric oxide-mediated protection of endothelialcells from hydrogen peroxide is mediated by intracellular zinc andglutathione. Am J Physiol Cell Physiol 296: C811–C820, 2009. Firstpublished February 4, 2009; doi:10.1152/ajpcell.00643.2008.—Oxi-dative stress may cause endothelial dysfunction and vascular disease.It has been shown that NO protects endothelial cells (EC) againstH2O2-induced toxicity. In addition, it is known that NO within cellsinduces a zinc release from proteins containing zinc-sulfur complexes.The aim of this study was to investigate whether zinc releasedintracellularly by NO plays a signaling role in the NO-mediatedprotection against H2O2 in rat aortic EC. Our results show that theNO-mediated protection toward H2O2 depends on the activities ofglutathione peroxidase and glutamate cysteine ligase (GCL), therate-limiting enzyme of glutathione (GSH) de novo biosynthesis.Moreover, NO increases the synthesis of the antioxidant GSH byinducing the expression of the catalytic subunit of GCL (GCLC).Chelating intracellular “free” zinc abrogates the NO-mediated in-crease of GCLC and of cellular GSH levels. As a consequence, theNO-mediated protection against H2O2-induced toxicity is impaired.We also show that under proinflammatory conditions, both cellularNO synthesis and intracellular “free” zinc are required to maintain thecellular GSH levels. Using RNA interference and laser scanningmicroscopy, we found that the NO-induced expression of GCLCdepends on the activation of the transcription factor Nrf2 but not onthe activity of the “zinc-sensing” transcription factor MTF-1. Thesefindings show that intracellular “free” zinc plays a signaling role in theprotective activity of NO and could explain why maintenance of anadequate zinc status in the endothelium is important to protect fromoxidative stress and the development of vascular disease.

inducible nitric oxide synthase; zinc signaling; oxidative stress; glu-tamate cysteine ligase; Nrf2

OXIDATIVE STRESS AND CHRONIC INFLAMMATION have been recog-nized to promote endothelial dysfunction and/or damage,which represent the initial steps in the development of chronicvascular diseases, such as atherosclerosis (41). Endothelialcells (EC) are particularly well equipped with defense mech-anisms for the detoxification of radicals and xenobiotics (70).The most abundant endogenously produced antioxidant ineukaryotic cells is the tripeptide glutathione (GSH), which ismainly responsible for the maintenance of the cellular redox

state (20, 44, 73). Selective inhibition of enzymes belonging tothe GSH redox cycle increase the susceptibility of EC toH2O2-mediated cell damage (21). GSH exerts its antioxidantactivity by directly scavenging radical species, as well as byparticipating in reactions catalyzed by antioxidant enzymessuch as glutathione peroxidase (GPx) (20, 44, 73). The totalcellular GSH concentration (0.5–10 mM) is determined by therate of GSH de novo synthesis, the cellular capacity to recycleGSH, and the rate of cellular GSH export (20, 44, 73). Therate-limiting step of the GSH de novo synthesis is catalyzed byglutamate-cysteine ligase (GCL; EC 6.3.2.2), a heterodimercomposed of a catalytic heavy chain (GCLC) and a modifier(or regulatory) light chain (GCLM) (16). Changes in GCLactivity may result from regulation at multiple levels affectingeither the catalytic or the modifying subunit, or both. Heritabledeficiency of antioxidant enzymes and, particularly, of bothGCLC and GCLM imposes a status of oxidative stress on thevasculature and may be associated with an increased risk todevelop vascular diseases (16, 29, 38, 54).

The transcription of human (15, 71) and mouse (7, 8) GCLCis mainly regulated by the transcription factor Nrf2 (nuclearfactor, erythroid-derived 2, like-2) (48). Nrf2 binds to theantioxidant responsive element (ARE; also defined as electro-phile response element), which has been identified in thepromoter regions of human, mouse, and, recently, rat GCLC(7, 40, 48, 51). Adenovirus-mediated overexpression of Nrf2increases intracellular GSH levels in human EC (10). Theantioxidant activity of zinc and nutritional antioxidants isstrongly dependent on the activation of Nrf2 and, particularly,on their ability to regulate the expression of the genes encodingthe subunits of GCL (12, 16, 49).

During inflammation, the inducible form of nitric oxidesynthase (iNOS) is expressed in EC, resulting in high-outputsynthesis of nitric oxide (NO). Whether iNOS-derived NOplays a deleterious or a protective role in endothelial injury hasnot been established yet. iNOS-deficient mice are more proneto atheroma formation (25), and overexpression of iNOS in-hibits experimental vascular lesion formation (69). High-out-put NO synthesis achieved by adding NO donors or by induc-ing the expression of iNOS by treating EC with proinflamma-tory cytokines protects EC against H2O2-induced toxicity byinhibiting lipid peroxidation (58, 63), iron uptake (9, 30), and

Address for reprint requests and other correspondence: M. M. Cortese-Krott,Dept. of Internal Medicine I, Cardio-Bio-Tech Research Group, MedicalFaculty, RWTH Aachen Univ., Pauwelsstrasse 30, D-52074 Aachen, Germany(e-mail: mcortese@ukaachen.de).

The costs of publication of this article were defrayed in part by the paymentof page charges. The article must therefore be hereby marked “advertisement”in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Am J Physiol Cell Physiol 296: C811–C820, 2009.First published February 4, 2009; doi:10.1152/ajpcell.00643.2008.

0363-6143/09 $8.00 Copyright © 2009 the American Physiological Societyhttp://www.ajpcell.org C811

on February 10, 2010 ajpcell.physiology.org

Dow

nloaded from

cytochrome c release and by inducing the expression of pro-tective proteins (50, 63, 65). Application of NO donors also hasbeen shown to induce GSH synthesis in EC by inducing theexpression of GCLC (46, 47) and by controlling the cystineuptake (39).

Both NO donors and iNOS-derived NO induce intracellularzinc release from zinc sulfur clusters as found in proteins suchas metallothionein (5, 33, 45, 55, 59–62). Thus NO may beinvolved in the control of the intracellular zinc homeostasis(31, 32). We recently demonstrated that exogenously addedzinc, similarly to NO, induces the synthesis of GSH by con-trolling the expression of GCLC and in this way protects ECagainst H2O2-induced toxicity (12). In the present study weinvestigated whether “free” zinc released by NO within thecells plays a signaling role in the NO-mediated protectionagainst H2O2 in EC.

MATERIALS AND METHODS

Materials. Unless otherwise specified, chemicals were purchasedfrom Merck (Darmstadt, Germany). Cell culture material was pur-chased from PAA Laboratories (Pashing, Austria); recombinant cy-tokines from Strathmann Biotech (Hannover, Germany); nonfat drymilk from Bio-Rad (Munich, Germany); materials for Western blots(NuPAGE LDS sample buffer, NuPAGE reducing agent, and NuPAGE7% Novex Tris-acetate precast gels) from Invitrogen (Karlsruhe,Germany); Ponceau S from SERVA Electrophoresis (Heidelberg,Germany); Hybond P transfer membrane from Amersham Bio-sciences (Munich, Germany); anti-human GCLC rabbit antiserumfrom Dunn Labortechnik (Asbach, Germany); mouse monoclonalanti-human !-actin antibody, horseradish peroxidase (HRP)-conjugated goat anti-mouse antiserum, and HRP-conjugated goatanti-rabbit antiserum from BD Biosciences (Erembodegem, Belgium);rabbit anti-human Nrf2 antiserum from Santa Cruz Biotechnology(Santa Cruz, CA); and goat anti-rabbit antiserum conjugated withAlexaFluor 594 from Invitrogen. The NO donor (Z)-1-[N-(2-aminoethyl)-N-(2-aminoethyl)amino]diazen-1-ium-1,2-diolate(DETA/NO) was synthesized as described previously (24).

Cells. Aorta was explanted from male Wistar rats "30 days old andof 150-g body weight, obtained from the Heinrich-Heine-Universityof Dusseldorf breeding facility. The procedure was conducted inconformity with the American Physiogical Society’s “Guiding prin-ciples for research involving animals and human beings” and wasapproved by the ethical committee of the Heinrich-Heine-University.Rat aorta EC were isolated by outgrowth from aortic rings, expressedthe phenotype vWFhigh Ox43high eNOShigh (64), and were cultured inRPMI 1640–10% fetal bovine serum (FBS).

Zinc determination. The zinc concentrations of media and bufferswere determined by flame atomic adsorption spectrometry as de-scribed previously (12).

Experimental procedures. Measurements were performed with ECfrom passages 2–8. Viability of the cells was assessed routinely at thebeginning and at the end of every experiment by using the trypan blueexclusion assay. Cells were cultured for 24 h in RPMI 1640–10%FBS in 12-well tissue culture plates (5 # 105 cells/well) in theabsence or presence of the NO donor DETA/NO or with proinflam-matory cytokines inducing endogenous NO synthesis or the respectivecontrols at the concentrations indicated. After washing, oxidativestress-mediated cell death was induced by a 24-h incubation with0.2–1.4 mM H2O2. The effectiveness of the H2O2 treatment wascontrolled by adding 2,000 units of catalase. Cell viability wasdetected by neutral red staining and is expressed as the percentage oflive H2O2-treated vs. untreated cells. Apoptosis or necrosis wasdetected by staining the DNA with H33342 dye and by detectingDNA strand breaks in acetone-fixed cells by using the in situ nick-translation method, as described previously (56). Where indicated, the

catalase inhibitor 3-amino-1,2,3-triazole (ATA; 0.5 mM), the gluta-thione peroxidase inhibitor !-mercaptosuccinic acid (MS; 1 mM), orthe GCL inhibitor L-buthionine sulfoximine (BSO; 0.05 mM) wasadded together with H2O2. Stock solutions of ATA and BSO (100mM) were prepared in double-distilled H2O and then diluted in theculture medium. The MS stock solution (100 mM) was diluted incomplete medium with 20 mM HEPES to equilibrate the pH.

To deplete the cells of zinc during the treatment with the NO donoror proinflammatory cytokines, we added a nontoxic concentration ofthe membrane-permeable zinc chelator N,N,N$,N$-tetrakis(2-pyridyl-methyl)ethylenediamine (TPEN; 3.3 %M) to the culture medium(containing 2.28 & 0.32 %M Zn2'). The effect of the TPEN treatmenton cell viability was determined by neutral red staining and byanalyzing the nuclear morphology after staining the DNA withH33342. The cell viability of TPEN-treated cells compared withuntreated cells was 100% & 9.2% (n ( 8). As a control for zincspecificity, equimolar concentrations of ZnSO4 (3.3 %M) were addedtogether with TPEN. To control whether extracellular zinc plays anyrole, we prepared a zinc-depleted medium by using Chelex 100 asdescribed previously (22, 23, 57). Briefly, FBS (containing 24.98 &2.00 %M Zn2') was treated with Chelex-100 chelating resin accord-ing to the manufacturer’s instructions and then mixed with RPMI1640. The zinc concentration of RPMI 1640 with 10% Chelex-FBSwas found to be below the detection limit (1.0 %M Zn2') of flameatomic adsorption spectrometry. Therefore, it was termed “low-zinc”medium, although Chelex removes other divalent metal ions as well.

Laser scanning microscopy. To determine whether the NO donorDETA/NO induces Nrf2 translocation into the nucleus, we stained EC(1 # 104 cells/well) with rabbit anti-human Nrf2 antiserum (1:100)together with goat anti-rabbit antiserum conjugated with AlexaFluor594 (1:250), as described previously (12). Nuclei were stained withthe DNA stain H33342 (7 %g/ml). Coverslips were analyzed under aZeiss laser scanning microscope 510 with the use of a Zeiss PlanNeofluar #40/1.3 oil differential interference contrast objective.

Real-time RT-PCR and gene-silencing experiments. Gene expres-sion was determined using TaqMan universal PCR master mix in aABI PRISM 7900 system (Applied Biosystems, Foster City, CA). Thefollowing primers and probes were purchased from Applied Biosys-tems: GCLC subunit (Gclc; NM_012815, assay Rn00563101-m1),heme oxygenase (decycling) 1 (Ho1; NM_012580, assay Rn00561387-m1),zinc transporter-1 (Znt1; NM_022853, assay Rn00575737-m1), and18s rRNA, which was chosen as a housekeeping gene. The smallinterference RNA (siRNA) sequences targeting Nrf2 or Mtf1 tran-scription factors were designed and synthesized by Qiagen, as previ-ously described (12). As a control for unspecific effects, a nonsilenc-ing siRNA sequence was used (Qiagen). The positive control was asiRNA sequence directed against Gapdh mRNA (Ambion, Hunting-don, UK). The best silencing efficiency was obtained by transfecting1.5 # 105 cells with 5 nM siRNA and 9 %l of HiPerfect transfectionreagent (Qiagen) following the manufacturer’s instruction. After 48 h,the culture medium was changed and the cells were treated withDETA/NO, proinflammatory cytokines, or the respective controltreatments as indicated. The mRNA levels of Nrf2, Mtf1, and Gapdhwere determined using a QuantiTect SYBR green PCR kit (Qiagen)and 18s rRNA as a housekeeping gene. Primers were designed withPrimer Express software 2.0 (Applied Biosystems) as described pre-viously (12). Results were analyzed following established procedures(52). Forty-eight hours after transfection, a fourfold decrease of theNrf2 mRNA and a threefold decrease of the Mtf1 mRNA weredetected (data not shown).

Determination of the total intracellular glutathione concentration.EC were cultured in 100-mm petri dishes (1 # 107 cells) for 24 h andtreated as indicated. Total cellular glutathione (GSH ' GSSG) wasdetermined photometrically by measuring the formation of 5-thio-2-nitrobenzoic acid from 5,5$-dithionitrobenzoic acid in the presence ofNADPH and glutathione reductase (1) at 405 nm using a FLUOstarOPTIMA plate reader (BMG Labtech, Offenburg, Germany). The

C812 NO PROTECTS EC FROM H2O2 VIA ZINC AND GLUTATHIONE

AJP-Cell Physiol • VOL 296 • APRIL 2009 • www.ajpcell.org

on February 10, 2010 ajpcell.physiology.org

Dow

nloaded from

amount of total glutathione was normalized to the protein content(expressed as %mol GSH/mg protein).

Western blot analysis. EC were cultured in six-well plates (1.5 #106 cells/well) and treated as indicated. Cells were lysed with RIPAlysis buffer (1% Nonidet-P 40, 0.5% sodium deoxycholate, and 0.1%SDS in phosphate-buffered saline) containing Complete mini proteaseinhibitors (Roche Diagnostic). The protein content was determinedusing the DC protein assay (Bio-Rad). Samples (30 %g) were sepa-rated in a 7% NuPAGE Novex Tris-acetate gel and then transferred onHybond P membranes. Membranes were stained with either rabbitanti-GCL antiserum (2.5 %g/ml) or mouse monoclonal anti-human!-actin antibodies (1:5,000) and were then incubated with HRP-conjugated goat anti-rabbit antibodies (1:5,000) or HRP-conjugatedgoat anti-mouse antiserum (1:5,000). The bands were visualized byautoradiography on a Hyperfilm ECL (Amersham Biosciences) usingSuperSignal West Pico chemiluminescent substrate (Pierce).

Statistical analysis. Real-time PCR data were processed as de-scribed previously (52). Data are means & SE. Statistical comparisonswere carried out with one-way or two-way ANOVA, followed by anappropriate post hoc comparison test (Tukey’s, Dunnett’s, or Stu-dent’s t-test).

RESULTS

NO-mediated protection of EC against H2O2 depends on theactivities of GPx and GCL. To analyze the mechanisms in-volved in the NO-induced protection of endothelial cells (EC)against H2O2, we pretreated rat aortic EC with the NO donorDETA/NO (1 mM) for 24 h and then treated cells withincreasing concentrations of H2O2. The half-life of H2O2 in thesupernatant of cultured cells is in the range of 10–20 min (53).Thus H2O2 will be completely destroyed within 2–5 h ascalculated by considering the pseudo-first-order kinetics ofH2O2-decay. The half-maximal lethal dose (LD50) was foundto be 1 mM H2O2 (see Fig. 2A, untreated). A 24-h pretreatmentwith 1 mM DETA/NO resulted in a nearly complete protectionfrom H2O2-induced toxicity (Fig. 1B, lane 2, and Fig. 2A). TheNO-mediated protection cannot be related to a direct scaveng-ing effect of NO on H2O2, because DETA/NO in culturemedium has a half-life of "7–8 h (5). According to first-orderkinetics, 1 mM DETA/NO generates 3.3 nM NO per minute.This is similar to a steady-state concentration of "4–5 %MNO, which has been calculated to be present in the immediatevicinity of a cell monolayer that enzymatically generates NO (37).

To determine whether catalase or GPx, both of whichcatalyze the reduction of H2O2 to H2O, plays a role in theNO-mediated protection toward H2O2, we cultured untreatedcells or cells pretreated with DETA/NO in the presence of 1mM H2O2 together with specific inhibitors of these enzymes,i.e., MS (1 mM) or ATA (0.5 mM). As also recently reportedby our group (9), addition of nontoxic concentrations of theseinhibitors significantly increased the toxicity of H2O2 in oth-erwise untreated cells (Fig. 1A, lanes 3 and 4 vs. lane 2, openbars). In cells pretreated with DETA/NO, inhibition of catalaseby ATA did not exert any effect on the NO-mediated protec-tion (Fig. 1A, lane 3 vs. lane 2, filled bars). In contrast,inhibition of GPx by MS nearly completely abolished theNO-mediated protection against H2O2 (Fig. 1A, lane 4 vs. lane2, filled bars). The role of GCL, the rate-limiting enzyme ofGSH biosynthesis, was investigated by adding the specificinhibitor BSO (50 %M). When GCL was inhibited during the treat-ment with H2O2, the toxicity of the H2O2 treatment was stronglyincreased (Fig. 1A, lane 5 vs. lane 2, open bars). However,

protection conferred by pretreatment with DETA/NO was notaffected (Fig. 1A, lane 5 vs. lane 2, filled bars).

If protein synthesis was inhibited during the pretreatmentwith DETA/NO by addition of the translation inhibitor cyclo-heximide, NO-mediated protection was significantly impaired(Fig. 1B, lane 3 vs. lane 2), suggesting that NO induces theexpression of protective proteins. If the cells were preincubatedwith the GCL inhibitor BSO before the treatment with H2O2,the toxicity of H2O2 was strongly enhanced (Fig. 1B, lane 6 vs.lane 1), and a pretreatment with DETA/NO under these con-ditions did not protect the EC from H2O2-induced toxicity (Fig.1B, lane 5 vs. lane 1). Thus the activities of GPx and GCL,enzymes belonging to the GSH redox cycle, are essential forthe DETA/NO-mediated protection of EC against H2O2.

NO protects EC against H2O2-induced toxicity and in-creases the cellular GSH levels via intracellular “free” zinc.To examine the role of zinc in the NO-mediated protectionfrom H2O2, we depleted the culture medium of zinc (and other

Fig. 1. Nitric oxide (NO) protects endothelial cells (EC) via the activities ofglutathione peroxidase (GPx) and glutamate cysteine ligase (GCL). A: EC werecultured for 24 h in the absence or presence of 1 mM (Z)-1-[N-(2-aminoethyl)-N-(2-aminoethyl)amino]diazen-1-ium-1,2-diolate (DETA/NO). After exten-sive washing, cells were treated for 24 h with 1 mM H2O2 in the presenceof specific inhibitors of catalase [3-amino-1,2,3-triazole (ATA), 0.5 mM] orGPx [!-mercaptosuccinic acid (MS), 1 mM]. Cell viability was determinedby neutral red staining. Values are means & SE (n ( 3– 6). **P ) 0.001vs. untreated cells (lane 1, open bar). #P ) 0.001 vs. cells pretreated withDETA/NO in the absence of inhibitors (lane 1, filled bar). §P ) 0.001 asindicated. B: EC were pretreated for 24 h with 1 mM DETA/NO, 3.5 %Mcycloheximide (cHEX), or 50 %M L-buthionine sulfoximine (BSO) or theircombination as indicated, washed, and then treated with 1 mM H2O2 for24 h. Cell viability was determined by neutral red staining. Values aremeans & SE (n ( 3– 6). *P ) 0.01; **P ) 0.001 vs. cells treated withH2O2 only (lane 1). #P ) 0.01; ##P ) 0.001 vs. cells pretreated withDETA/NO (lane 2).

C813NO PROTECTS EC FROM H2O2 VIA ZINC AND GLUTATHIONE

AJP-Cell Physiol • VOL 296 • APRIL 2009 • www.ajpcell.org

on February 10, 2010 ajpcell.physiology.org

Dow

nloaded from

divalent metals) by treating the serum with the ion exchangeresin Chelex or added the cell-permeable zinc chelator TPEN,which chelates intra- and extracellular zinc (Kd for Zn2'-TPEN: 6.3 # 10*16 M at pH 7.6) (13, 27). The TPENconcentration (3.3 %M) chosen for the experiments in thisstudy overwhelms the zinc concentration found in the culturemedium containing 10% FBS (2.3 & 0.3 %M Zn2' as deter-mined by atomic adsorption spectrometry) and did not result incytotoxicity after a 24-h incubation, as shown previously (12),but increased cytotoxicity following a 24-h incubation withH2O2 (Fig. 2A). High intracellular output of NO was achievedby treating the cells with proinflammatory cytokines (IL-1! 'IFN+), which induces the expression of iNOS (64) after 6 h ofincubation as determined by measuring the accumulation ofnitrite into the supernatant (data not shown). Protection medi-ated by preincubation with DETA/NO or proinflammatorycytokines is not affected by low extracellular zinc concentra-tions ()1 %M), achieved by treating the serum with Chelex.On the contrary, addition of TPEN significantly impaired theprotective effects mediated by pretreatment with DETA/NO orproinflammatory cytokines (Fig. 2A) toward H2O2. The NO-mediated protection was fully restored if TPEN was titratedwith an equimolar concentration of zinc (Fig. 2A). Theseresults show that the NO-mediated protection against the tox-icity of H2O2 is dependent on intracellular TPEN-chelatablezinc.

To verify whether NO increases the GSH synthesis via“free” zinc, we measured intracellular levels of total GSH 24 h

after treating the cells with DETA/NO or proinflammatorycytokines in low-zinc culture medium (prepared with Chelex-treated serum) or in the presence of TPEN. We found thatDETA/NO increased the total GSH levels more than twofoldcompared with untreated cells in normal culture medium aswell as in low-zinc medium (Fig. 2B). Addition of TPENcompletely abolished the NO-induced increase of GSH, andthis inhibition was abrogated by adding an equimolar amountof ZnSO4 (Fig. 2B), demonstrating the zinc specificity of theTPEN effect. Together, these findings demonstrate that theDETA/NO induced GSH-synthesis as well as the protectionagainst H2O2 depend on intracellular “free” zinc.

We also examined the effects of cellular NO production onGSH synthesis. In cytokine-treated EC, the total intracellularGSH levels did not differ from that in untreated cells (Fig. 2B).However, addition of the specific NO synthase inhibitor L-N5-(1-iminoethyl)ornithine hydrochloride (L-NIO) significantlydecreased cellular GSH levels, indicating that endogenous NOsynthesis is needed to maintain the cellular GSH levels underproinflammatory conditions in EC. The same effects wereobserved in low-zinc medium. In contrast, after TPEN wasadded (Fig. 2B), a decrease in the cellular GSH levels wasobserved. If an equimolar amount of ZnSO4 was added to-gether with TPEN, the cytokine-induced response was fullyrestored. Addition of ZnSO4 to cells treated with cytokines '

Fig. 2. NO protects EC from H2O2 and increases total cellular GSH viaintracellular “free” zinc. A: EC were pretreated for 24 h as indicated and, afterextensive washing, were treated for 24 h with 1 mM H2O2. Cell viability wasdetermined by neutral red staining. Values are means & SE (n ( 3–6).Cytokines (Cyt): 200 U/ml IL-1! ' 100 U/ml IFN+. *P ) 0.05; **P ) 0.01;***P ) 0.001 vs. cells treated with H2O2 only (CTRL, open bar). #P ) 0.01as indicated. B: EC were cultured for 24 h as indicated. Total GSH was assayedin cell lysates using the GSH reductase recycling method. Values are means &SE (n ( 5). *P ) 0.05; **P ) 0.01 vs. untreated control (CTRL, open bar).#P ) 0.01 as indicated. n.s., No significant difference.

Fig. 3. Exogenously applied NO increases GCL expression via intracellular“free” zinc. A: EC were cultured in the presence of 1 mM DETA/NO, 3.3 %MN,N,N$,N$-tetrakis(2-pyridylmethyl)ethylenediamine (TPEN), 3.3 %M ZnSO4,or their combination as indicated. At different time points, the expression ofGclc mRNA was measured by real-time RT-PCR and normalized against 18srRNA. Values are means & SE (n ( 3). *P ) 0.05. B: EC were incubated for24 h with 1 mM DETA/NO, 3.3 %M TPEN, 3.3 or 100 %M ZnSO4, or 1 mMDETA as indicated. Subsequently, 30 %g of total protein extract were analyzedby Western blotting using rabbit anti-human GCL antiserum and mouseanti-human !-actin (!Act) antibody as a loading control.

C814 NO PROTECTS EC FROM H2O2 VIA ZINC AND GLUTATHIONE

AJP-Cell Physiol • VOL 296 • APRIL 2009 • www.ajpcell.org

on February 10, 2010 ajpcell.physiology.org

Dow

nloaded from

L-NIO ' TPEN restored the GSH levels to a great extent (Fig.2B) but did not restore cell viability following exposure toH2O2 (Fig. 2A). These results show that under proinflammatoryconditions, cellular NO synthesis via iNOS as well as intracel-lular “free” zinc are mainly responsible for maintaining thecellular GSH levels in EC. However, the NO-mediated protec-tion also may include zinc-independent effects.

Exogenously applied NO increases GCLC expression via“free” zinc. We further analyzed the role of GCL in theNO-induced and zinc-mediated protection from H2O2 by ana-lyzing the expression of the catalytic subunit GCLC. We foundthat DETA/NO increased the expression of Gclc mRNA (Fig.3A) as well as that of GCLC protein (Fig. 3B). In the presenceof TPEN, DETA/NO failed to increase the expression of GclcmRNA and GCLC protein levels (Fig. 3, A and B, lane 2). Ifan equimolar amount of ZnSO4 was added together withTPEN, the NO-induced response was fully restored (Fig. 3, Aand B, lane 3). Thus the NO-induced expression of GCLC isfully dependent on the presence of zinc.

NO-mediated Gclc gene regulation is Nrf2 dependent. Weasked whether the antioxidant transcription factor Nrf2 isinvolved in the NO-mediated protection against H2O2. Usingimmunocytochemistry and laser scanning microscopy, we ob-served nuclear staining corresponding to Nrf2 in cells treatedfor 1 h with DETA/NO (Fig. 4BI) or with the established Nrf2activator tert-butylhydroquinone (tBHQ; 100 %M; Fig. 4CI).Instead, no signal corresponding to Nrf2 was detected in nucleiof untreated cells (Fig. 4AI). In addition, we found that the

silencing of Nrf2 via RNA interference significantly down-regulated the basal expression of Gclc mRNA. Moreover, aftersilencing of Nrf2, neither DETA/NO nor proinflammatorycytokines were effective in inducing the expression of GclcmRNA in EC (Fig. 5A). As a control, we verified whether Nrf2silencing modified the expression of Ho1, which is an estab-lished Nrf2 target (2), and similar results were found comparedwith Gclc (Fig. 5B). To investigate whether the intracellular“zinc-sensing” transcription factor MTF-1 is involved in theregulation of Gclc, we measured the expression of Gclc and ofthe MTF-1 target Znt1 after Mtf1 silencing. The absence ofMTF-1 did not exert any effect on basal or either DETA/NO-or cytokine-induced expression of Gclc (Fig. 5A). On the otherhand, Mtf1 silencing strongly downregulated the basal as wellas the DETA/NO-induced expression of Znt1 (Fig. 5C). Thisfinding shows that Nrf2 but not MTF-1 controls basal as wellas zinc-induced expression of Gclc. These results demonstratethat in rat aortic EC, the NO-mediated induction of GclcmRNA is regulated by Nrf2.

To further control whether DETA/NO induced the expres-sion of Ho1 or Znt1 via intracellular “free” zinc, we measuredtheir mRNA levels after a 24-h incubation with DETA/NO inthe presence or absence of TPEN. In the presence of TPEN,DETA/NO failed to increase the expression of Ho1 and Znt1mRNA (Fig. 6, A and B, lanes 3). If an equimolar amount ofZnSO4 was added together with TPEN, the NO-induced re-sponses were fully restored (Fig. 6, A and B, lanes 4). Thus the

Fig. 4. NO induces the translocation of Nrf2 into thenucleus. EC were seeded on glass coverslips. After24 h, the cells were washed and cultured for 2 h withmedium (A) 1 mM DETA/NO (B), or 100 %Mtert-butylhydroquinone (tBHQ; C), which is an es-tablished Nrf2 activator. Cells were then stainedwith an anti-Nrf2 antibody (I) and the DNA stainH33342 dye (II). Nrf2 staining (I) and DNA staining(II) were merged to produce the images shown in III.

C815NO PROTECTS EC FROM H2O2 VIA ZINC AND GLUTATHIONE

AJP-Cell Physiol • VOL 296 • APRIL 2009 • www.ajpcell.org

on February 10, 2010 ajpcell.physiology.org

Dow

nloaded from

NO-induced expressions of Ho1 as well as Znt1 are fullydependent on the presence of zinc.

DISCUSSION

NO-mediated protection of EC against H2O2-induced toxic-ity. It has been demonstrated that NO donors as well asiNOS-derived NO protect EC from the cytotoxic effects ofH2O2 by various mechanisms including the expression ofprotective proteins such as HO-1 (50) or Bax and Bcl-2 (63,65). To further investigate protective effects of NO, we usedthe following model system: EC were pretreated for 24 h witheither 1 mM DETA/NO or proinflammatory cytokines, whichinduce the expression of iNOS. Afterward, the cells werewashed and then treated with concentrations of H2O2 rangingfrom 0.2 to 1.4 mM. These H2O2 concentrations are 40–300times higher than the 5–15 %M range produced by inflamma-tory cells (42, 67). However, an initial bolus addition of 1 mMH2O2 will be rapidly consumed in the supernatant of culturedcells and completely destroyed within 2–5 h (53). Pretreatmentwith exogenously applied or endogenously synthesized NOprotects EC from the toxic effects of up to 1.4 mM H2O2,although in this report we have presented only data showingthe protective effects toward the LD50 for H2O2 (1 mM).

Others have shown that simultaneous treatment with NOenhances the toxicity of H2O2 by inhibiting the degradation of

H2O2 (56). However, in our cell model system we added theNO donor and the H2O2 at two well-separated time points, thusexcluding a cooperative action of NO and H2O2.

Role of the GSH redox cycle. We found that in rat aortic EC,the NO donor DETA/NO increased the intracellular GSHlevels about twofold by increasing the expression of GCLC,which is the catalytic subunit of the rate-limiting enzyme forthe GSH de novo synthesis. Similar results have previouslybeen published using bovine aortic EC (46, 47). We also foundthat specific inhibition of GCL by BSO completely abolishedthe NO-induced increases of intracellular GSH as well as theNO-induced protection from the toxic effects of H2O2. Inhibi-tion of GPx by MS increased the toxicity of H2O2, as previ-ously shown (12, 21), and strongly impaired the protectioninduced by preincubation with DETA/NO. Catalase does notappear to play a role in the NO-mediated protection, becausespecific inhibition of catalase did not abrogate the protectiveeffect of the DETA/NO pretreatment. These results suggestthat GSH itself as well as the GPx activity are necessary for theprotection against H2O2 by NO released from NO donors.

We also have analyzed the effects of cellular NO synthesison the GSH level as well as on the Gclc mRNA expression.The expression of iNOS was induced by treating the cells withproinflammatory cytokines (IL-1! ' IFN+) (64). An increaseof iNOS mRNA was observed after 2–4 h of incubation withcytokines, and a significant increase of NO production couldalready be detected after 6 h as determined by measuring the

Fig. 6. NO induces Ho1 and Znt1 expression via intracellular “free” zinc. ECwere cultured for 24 h in the presence of 1 mM DETA/NO, 3.3 %M TPEN, 3.3%M ZnSO4, or their combination as indicated. The expression of Ho1 and Znt1mRNA were measured by real-time RT-PCR and normalized against 18srRNA. Values are means & SE (n ( 3). *P ) 0.05 vs. untreated control (lane1). #P ) 0.05 as indicated.

Fig. 5. NO induces Gclc expression via Nrf2. EC were transfected with 5 nMsmall interference RNA (siRNA) and 9 %l of transfection reagent. Forty-eighthours after the transfection, cells were treated for the following 24 h asindicated. The mRNA expression of Gclc, heme oxygenase-1 (Ho1), and zinctransporter-1 (Znt1) were studied by real-time RT-PCR. Ct values from 3independent experiments were averaged, and the expression relative to themock control (cells treated with the transfection reagent only) was normalizedagainst 18s rRNA. Values are means & SE (n ( 3). *P ) 0.05 vs. nonsilencedcontrol. #P ) 0.05 as indicated.

C816 NO PROTECTS EC FROM H2O2 VIA ZINC AND GLUTATHIONE

AJP-Cell Physiol • VOL 296 • APRIL 2009 • www.ajpcell.org

on February 10, 2010 ajpcell.physiology.org

Dow

nloaded from

accumulation of nitrite in the cell culture medium (data notshown). After a 24-h incubation with IL-1! ' IFN+, the GSHlevels were not different from those in untreated cells, as foundby others in a mouse hemoangiotelioma cell line treated withIL-1! (68). In the presence of the specific NOS inhibitorL-NIO, we observed a strong decrease of the Gclc mRNAexpression and of the total cellular GSH levels, which corre-lated with the resistance of the cells toward H2O2. Thusintracellular NO production is required to maintain the cellularGSH levels under proinflammatory conditions. This is in linewith results found with rat hepatocytes, where after stimulationwith IL-1!, an inhibition of NO synthesis by the specific NOSinhibitor N-methyl-L-arginine (L-NMA) resulted in a GSHdepletion and a decrease of the GSH/GSSG ratio and Gclc genetranscription (35). In conclusion, not only NO generated by NOdonors but also high-output cellular NO synthesis protect ECagainst H2O2 by inducing the intracellular de novo synthesisof GSH.

Role of zinc. We recently found that extracellular added zincinduces GSH synthesis by controlling the expression of GCLC,thus protecting EC from H2O2-induced toxicity (12). In thesame setting as used presently, Berendji et al. (5) showed thattreatment with NO donors induces an intracellular zinc releasein rat aortic EC, as monitored using the zinc-dependent fluoro-phore Zinquin. To demonstrate that intracellular zinc releasedby NO is essential for the increase of the GSH levels and theendothelial protection, we used the membrane-permeable zincchelator TPEN at a concentration that overwhelms the zincconcentration of the culture medium. In the presence of TPEN,NO fails to induce the GCLC expression, to increase thecellular GSH levels, and, as a consequence, to protect the ECagainst H2O2 toxicity.

The main limitation of studies about intracellular “free” zincin biological systems is the lack of metal chelators that onlychelate Zn2'. TPEN has very high affinities for Cu2' (1020.5

M*1), Zn2' (1015.6 M*1), Fe2' (1014.6 M*1), and Mn2'

(1010.3 M*1) (3) but a low affinity for Ca2' (104.4 M*1) (4).However, compared with Zn2' and Fe2', the concentrations ofCu2' and Mn2' in cells are quite low (4). In addition, underthe reducing conditions in cells, the oxidation level of copperusually is Cu1'. Nevertheless, to control whether the TPENeffects are indeed due to zinc chelation, we added equimolaramounts of zinc in all experiments with TPEN. Addition ofzinc always completely abrogated the effects exerted by TPEN,indicating that the TPEN effects are mainly due to the chelationof zinc.

To distinguish between effects of intracellular vs. extracel-lular zinc chelation (TPEN chelates both), we measured theeffects exerted by DETA/NO or proinflammatory cytokines ina medium containing low amounts of zinc ()1 %M). Thismedium was prepared with Chelex-treated serum, which is themajor source of zinc within culture media. Chelex is an ionexchange chelating resin and is not specific for zinc. However,it efficiently removes Zn2' from buffers or culture media (22,23, 57), as verified by us using flame atomic adsorptionspectrometry and by others using the more sensitive induc-tively coupled plasma/absorption emission spectrometry (22).The effects of DETA/NO or iNOS-derived NO in low-zincmedia are not different from the effects measured in the normalzinc-containing media. This indicates that intracellular and notextracellular zinc is required for the NO-mediated protection of

EC and the control of the cellular GSH levels. A similarexperimental setting was used by Tang et al. (66), who dem-onstrated that NO decreases the sensitivity of pulmonary EC toLPS-induced apoptosis in a zinc-dependent fashion.

Under physiological conditions, zinc is redox inert, but it isknown to exert antioxidant effects. Our work shows thatNO-mediated intracellular zinc release plays a protective roleagainst oxidative damage. In the brain, a deregulation of theneuronal zinc homeostasis has been shown to be stronglylinked to a mitochondrial dysfunction and to oxidative stress,making zinc a possible contributor to the activation of patho-physiological pathways (17). In EC, Wiseman et al. (72) haveproposed that an alteration of the intracellular zinc homeostasismay be a mechanism for the H2O2-mediated toxicity. They(72) showed that H2O2 induces intracellular zinc release andcell death in serum-starved pulmonary artery EC. Addition of6.5 %M TPEN or overexpression of metallothionein to scav-enge intracellular Zn2' decreased the H2O2-mediated increasesin superoxide anion production within the mitochondria as wellas the activation of caspases. The authors concluded thatintracellular zinc release by H2O2 may result in inappropriatephysiological responses, causing oxidative stress and tissuedamage. However, the TPEN concentration used was quitehigh, and effects were only followed up for 4 h (72). Inaddition, serum starvation depletes cells from zinc, thus ren-dering them more susceptible to apoptotic stimuli (12, 22, 23)and, in particular, to H2O2 (12). Thus, as also shown in thepresent study, preincubation with TPEN increased H2O2-me-

Fig. 7. NO protects EC against H2O2-induced toxicity via zinc-dependentactivation of the GSH redox cycle. In zinc sulfur clusters, zinc ions aretetracoordinated via sulfur (cysteine) and nitrogen (histidine). NO induces zincrelease from zinc sulfur clusters of intracellular proteins. This “free” zincactivates the transcription factor Nrf2, which in turn induces the transcriptionof the Gclc gene. Enhanced expression of GCL, the rate-limiting enzyme ofGSH biosynthesis, results in an increased concentration of intracellular GSH,which as a cofactor of GPx is essential to detoxify H2O2 via reduction to H2O.This protective pathway can be inhibited at various points, e.g., by chelatingintracellular “free” zinc by TPEN, by Nrf2 silencing (SiNrf2), by translationinhibition via cHEX, by GCL inhibition via BSO, or by GPx inhibition via MS.

C817NO PROTECTS EC FROM H2O2 VIA ZINC AND GLUTATHIONE

AJP-Cell Physiol • VOL 296 • APRIL 2009 • www.ajpcell.org

on February 10, 2010 ajpcell.physiology.org

Dow

nloaded from

diated cytotoxicity. It is therefore difficult to compare theseresults with ours. In contrast to other reactive oxygen species(ROS), the NO-mediated disruption of zinc fingers is reversible(34). Exogenous applied NO or endogenous NO releases zincand induces the protective response mediated by GSH. Zincwill then be rebound or sequestrated into intracellular compart-ments. We speculate that the oxidant damage mediated byH2O2 (and therefore H2O2-mediated zinc release) is limited bythe increased antioxidant capacity of the cells.

It has been shown that there is a correlation between theredox potential of a cell and the amount of “free” zinc withincells, which in turn depends on the total amount of cellular zincand the zinc-buffering capacity of the cells (31). Large zincfluctuations are possible at low buffering capacity and highredox potential. In contrast, high buffering capacity and lowredox potential limits zinc fluctuation (31).

Involvement of Nrf2 or of MTF-1? The transcription factorNrf2 mainly regulates the expression of proteins that detoxifyxenobiotics, reduce oxidized proteins, disrupt redox cyclingreactions, maintain the levels of cellular reducing equivalents,inhibit the intracellular generation of ROS, or counteract ROS-mediated effects (for review, see Ref. 28). Data obtained inmouse models as well as in human cells suggest that Nrf2regulates the expression of GCLC (7, 15, 51, 71). In addition,Nrf2 knockout mice exhibit a reduced basal expression of Gclctogether with an increased sensitivity toward oxidative stress.In the presence of activators, Nrf2 translocates into the nucleus,where it activates the transcription of ARE target genes (26).Previously, we observed that in rat EC, Nrf2 under basalconditions is localized mainly in the cytoplasm and, afteraddition of 100 %M Zn2', translocates into the cell nucleus(12). Presently, we have shown that NO released fromDETA/NO also induces a nuclear translocation of Nrf2. Sim-ilar results have been found in rat smooth muscle cells treatedwith other NO donors (43). In addition, we have found thatNrf2 silencing results in an inhibition of Gclc mRNA expres-sion induced by NO released from DETA/NO or synthesizedby iNOS. As a control, we also investigated the expression ofHo1, which is a well-characterized Nrf2 target gene (2), andfound similar results. Thus, in primary rat EC, the expressionof Gclc is induced by NO via Nrf2. Moreover, we found thatsimilarly to Gclc, the NO-induced expression of Ho1 is zincdependent, pointing to a mechanism where NO induces theexpression of Nrf2-dependent genes via zinc and Nrf2.

Studies have indicated that the zinc-sensing transcriptionfactor MTF-1, which binds to metal response elements (MREs), alsomay regulate Gclc gene expression (19, 51). In fact, MREs arepresent in the promoter of both mouse (19) and human Gclc(51), and a strongly reduced basal expression of Gclc has beenobserved in Mtf1 knockout mice (19). Although it is wellknown that zinc may regulate gene transcription via activationof MTF-1, we found that silencing of Mtf1 did not affect eitherthe basal or the NO-induced transcription of rat Gclc, whereasit inhibited the basal as well as the NO-induced expression ofZnt1, a typical MTF-1 target gene (36). Thus Nrf2, and notMTF-1, is the predominant regulator of Gclc expression in ratEC. Using other approaches, Yang et al. (75) found that Nrf2is essential for the regulation of Gclc expression in the rat. Liet al. (40) recently identified a putative ARE sequence 4 kbupstream of the rat GCLC promoter. Of interest, NO-inducedexpression of Znt1 in EC was both MTF-1 dependent and

inhibited by addition of TPEN, as shown previously in HepG2cells (11). This indicates that NO-mediated gene expressionmay act via MTF-1 in genes other than Gclc.

The molecular mechanism underlying the activation of Nrf2has not been completely elucidated so far. Keap1 (Kelch-likeECH-associated protein-1) binds to Nrf2, thus targeting it forubiquitination and proteolytic degradation, and as such isresponsible for the rapid Nrf2 turnover. Compounds that reactwith Keap1 result in subsequent nuclear translocation of Nrf2.Recombinant murine Keap1 in vitro binds zinc via two cys-teine SH residues (14). Electrophiles or ROS have been foundto cause cysteine modifications, which results in zinc releaseand subsequent profound conformational changes of Keap1in vitro (14). When EC are activated with proinflammatorycytokines, they also increase their ROS levels, and ROS areknown to activate the Keap1-Nrf2/ARE pathway. Thus, inaddition to the NO and zinc-dependent pathway, ROS pro-duced by cytokines may directly inactivate Keap1. Recently,NO also has been identified to cause Keap1 thiol modificationin cells (6). Therefore, Keap1 is an excellent candidate as asensor for NO.

Conclusions. Zinc has been named a “second messenger”(74) and the “calcium of the twenty-first century” (18). Ourresults suggest that intracellularly released (“free”) zinc mayindeed exert a signaling function in the NO-mediated protec-tion of EC from oxidative stress-mediated cell death. Zincsignals emerge as integrate members of redox signaling net-works (Fig. 7). In addition, it has been proposed that aninadequate zinc concentration in the plasma or in vasculartissues may be involved in the initiation of cell injury, theamplification of oxidative stress and inflammation, or the lackof protection against apoptosis, all of which are key eventsleading to endothelial dysfunction and vascular disease (3). Wehave shown in the present study that chelation of intracellular“free” zinc limits the protective activity of NO. Thus the datapresented may explain why the maintenance of an adequatezinc status in the endothelium is important to protect fromoxidative stress and the development of vascular diseases.

ACKNOWLEDGMENTS

We thank Dr. M. Falck (Central Institute for Clinical Chemistry andLaboratory Diagnostic, Heinrich-Heine-University of Dusseldorf) for the zincmeasurements by atomic adsorption spectrometry and Marija Lenzen forexpert technical assistance.

GRANTS

This work was supported by Deutsche Forschungsgemeinschaft Grant SFB503, A3.

REFERENCES

1. Abdelmohsen K, Gerber PA, von Montfort C, Sies H, Klotz LO.Epidermal growth factor receptor is a common mediator of quinone-induced signaling leading to phosphorylation of connexin-43: role ofglutathione and tyrosine phosphatases. J Biol Chem 278: 38360–38367,2003.

2. Alam J, Stewart D, Touchard C, Boinapally S, Choi AMK, Cook JL.Nrf2, a cap’n’collar transcription factor, regulates induction of the hemeoxygenase-1 gene. J Biol Chem 274: 26071–26078, 1999.

3. Anderegg G, Hubmann E, Podder NG, Wenk F. Pyridine derivates ascomplexing agents XI. Thermodynamics of metal complex formation withbis-, tris-, and tetrakis[(2-pyridyl)metyl]amines. Helv Chim Acta 60:123–140, 1977.

4. Arslan P, Di VF, Beltrame M, Tsien RY, Pozzan T. Cytosolic Ca2'

homeostasis in Ehrlich and Yoshida carcinomas. A new, membrane-

C818 NO PROTECTS EC FROM H2O2 VIA ZINC AND GLUTATHIONE

AJP-Cell Physiol • VOL 296 • APRIL 2009 • www.ajpcell.org

on February 10, 2010 ajpcell.physiology.org

Dow

nloaded from

permeant chelator of heavy metals reveals that these ascites tumor celllines have normal cytosolic free Ca2'. J Biol Chem 260: 2719–2727,1985.

5. Berendji D, Kolb-Bachofen V, Meyer KL, Grapenthin O, Weber H,Wahn V, Kroncke KD. Nitric oxide mediates intracytoplasmic andintranuclear zinc release. FEBS Lett 405: 37–41, 1997.

6. Buckley BJ, Li S, Whorton AR. Keap1 modification and nuclear accu-mulation in response to S-nitrosocysteine. Free Radic Biol Med 44:692–698, 2008.

7. Chan JY, Kwong M. Impaired expression of glutathione synthetic en-zyme genes in mice with targeted deletion of the Nrf2 basic-leucine zipperprotein. Biochim Biophys Acta 1517: 19–26, 2000.

8. Chan K, Kan YW. Nrf2 is essential for protection against acute pulmo-nary injury in mice. Proc Natl Acad Sci USA 96: 12731–12736, 1999.

9. Chang J, Rao NV, Markewitz BA, Hoidal JR, Michael JR. Nitric oxidedonor prevents hydrogen peroxide-mediated endothelial cell injury. Am JPhysiol Lung Cell Mol Physiol 270: L931–L940, 1996.

10. Chen XL, Dodd G, Thomas S, Zhang X, Wasserman MA, Rovin BH,Kunsch C. Activation of Nrf2/ARE pathway protects endothelial cellsfrom oxidant injury and inhibits inflammatory gene expression. Am JPhysiol Heart Circ Physiol 290: H1862–H1870, 2006.

11. Chung MJ, Hogstrand C, Lee SJ. Cytotoxicity of nitric oxide isalleviated by zinc-mediated expression of antioxidant genes. Exp Biol Med(Maywood ) 231: 1555–1563, 2006.

12. Cortese MM, Suschek CV, Wetzel W, Kroncke KD, Kolb-Bachofen V.Zinc protects endothelial cells from hydrogen peroxide via Nrf2-depen-dent stimulation of glutathione biosynthesis. Free Radic Biol Med 44:2002–2012, 2008.

13. Cousins RJ, Blanchard RK, Popp MP, Liu L, Cao J, Moore JB, GreenCL. A global view of the selectivity of zinc deprivation and excess ongenes expressed in human THP-1 mononuclear cells. Proc Natl Acad SciUSA 100: 6952–6957, 2003.

14. Dinkova-Kostova AT, Holtzclaw WD, Wakabayashi N. Keap1, thesensor for electrophiles and oxidants that regulates the phase 2 response,is a zinc metalloprotein. Biochemistry 44: 6889–6899, 2005.

15. Erickson AM, Nevarea Z, Gipp JJ, Mulcahy RT. Identification of avariant antioxidant response element in the promoter of the humanglutamate-cysteine ligase modifier subunit gene. Revision of the AREconsensus sequence. J Biol Chem 277: 30730–30737, 2002.

16. Franklin CC, Backos DS, Mohar I, White CC, Forman HJ, KavanaghTJ. Structure, function, and post-translational regulation of the catalyticand modifier subunits of glutamate cysteine ligase. Mol Aspects Med. Inpress.

17. Frazzini V, Rockabrand E, Mocchegiani E, Sensi SL. Oxidative stressand brain aging: is zinc the link? Biogerontology 7: 307–314, 2006.

18. Frederickson CJ, Koh JY, Bush AI. The neurobiology of zinc in healthand disease. Nat Rev Neurosci 6: 449–462, 2005.

19. Gunes C, Heuchel R, Georgiev O, Muller KH, Lichtlen P, BluthmannH, Marino S, Aguzzi A, Schaffner W. Embryonic lethality and liverdegeneration in mice lacking the metal-responsive transcriptional activatorMTF-1. EMBO J 17: 2846–2854, 1998.

20. Han D, Hanawa N, Saberi B, Kaplowitz N. Mechanisms of liver injury.III. Role of glutathione redox status in liver injury. Am J Physiol Gastro-intest Liver Physiol 291: G1–G7, 2006.

21. Harlan JM, Levine JD, Callahan KS, Schwartz BR, Harker LA.Glutathione redox cycle protects cultured endothelial cells against lysis byextracellularly generated hydrogen peroxide. J Clin Invest 73: 706–713,1984.

22. Ho E, Ames BN. Low intracellular zinc induces oxidative DNA damage,disrupts p53, NFkappa B, and AP1 DNA binding, and affects DNA repairin a rat glioma cell line. Proc Natl Acad Sci USA 99: 16770–16775, 2002.

23. Ho E, Courtemanche C, Ames BN. Zinc deficiency induces oxidativeDNA damage and increases p53 expression in human lung fibroblasts. JNutr 133: 2543–2548, 2003.

24. Hrabie JA, Arnold EV, Citro ML, George C, Keefer LK. Reaction ofnitric oxide at the beta-carbon of enamines. A new method of preparingcompounds containing the diazeniumdiolate functional group. J Org Chem65: 5745–5751, 2000.

25. Ihrig M, Dangler CA, Fox JG. Mice lacking inducible nitric oxidesynthase develop spontaneous hypercholesterolaemia and aortic athero-mas. Atherosclerosis 156: 103–107, 2001.

26. Kensler TW, Wakabayashi N, Biswal S. Cell survival responses toenvironmental stresses via the keap1-Nrf2-ARE pathway. Annu Rev Phar-macol Toxicol 47: 89–116, 2007.

27. Kimura E, Aoki S, Kikuta E, Koike T. Bioinorganic chemistry specialfeature: a macrocyclic zinc(II) fluorophore as a detector of apoptosis. ProcNatl Acad Sci USA 100: 3731–3736, 2003.

28. Kobayashi M, Yamamoto M. Molecular mechanisms activating theNrf2-Keap1 pathway of antioxidant gene regulation. Antioxid RedoxSignal 7: 385–394, 2005.

29. Koide S, Kugiyama K, Sugiyama S, Nakamura S, Fukushima H,Honda O, Yoshimura M, Ogawa H. Association of polymorphism inglutamate-cysteine ligase catalytic subunit gene with coronary vasomotordysfunction and myocardial infarction. J Am Coll Cardiol 41: 539–545,2003.

30. Kotamraju S, Tampo Y, Keszler A, Chitambar CR, Joseph J, HaasAL, Kalyanaraman B. Nitric oxide inhibits H2O2-induced transferrinreceptor-dependent apoptosis in endothelial cells: role of ubiquitin-protea-some pathway. Proc Natl Acad Sci USA 100: 10653–10658, 2003.

31. Krezel A, Hao Q, Maret W. The zinc/thiolate redox biochemistry ofmetallothionein and the control of zinc ion fluctuations in cell signaling.Arch Biochem Biophys 463: 188–200, 2007.

32. Kroncke KD. Cellular stress and intracellular zinc dyshomeostasis. ArchBiochem Biophys 463: 183–187, 2007.

33. Kroncke KD, Fehsel K, Schmidt T, Zenke FT, Dasting I, Wesener JR,Bettermann H, Breunig KD, Kolb-Bachofen V. Nitric oxide destroyszinc-sulfur clusters inducing zinc release from metallothionein and inhi-bition of the zinc finger-type yeast transcription activator LAC9. BiochemBiophys Res Commun 200: 1994.

34. Kroncke KD, Klotz LO, Suschek CV, Sies H. Comparing nitrosativeversus oxidative stress toward zinc finger-dependent transcription. Uniquerole for NO. J Biol Chem 277: 13294–13301, 2002.

35. Kuo PC, Abe KY, Schroeder RA. Interleukin-1-induced nitric oxideproduction modulates glutathione synthesis in cultured rat hepatocytes.Am J Physiol Cell Physiol 271: C851–C862, 1996.

36. Langmade SJ, Ravindra R, Daniels PJ, Andrews GK. The transcriptionfactor MTF-1 mediates metal regulation of the mouse ZnT1 gene. J BiolChem 275: 34803–34809, 2000.

37. Laurent M, Lepoivre M, Tenu JP. Kinetic modelling of the nitric oxidegradient generated in vitro by adherent cells expressing inducible nitricoxide synthase. Biochem J 314: 109–113, 1996.

38. Leopold JA, Loscalzo J. Oxidative enzymopathies and vascular disease.Arterioscler Thromb Vasc Biol 25: 1332–1340, 2005.

39. Li H, Marshall ZM, Whorton AR. Stimulation of cystine uptake bynitric oxide: regulation of endothelial cell glutathione levels. Am J PhysiolCell Physiol 276: C803–C811, 1999.

40. Li M, Chiu JF, Kelsen AT, Lu SC, Fukagawa NK. Nrf2 and antioxidantresponsive element in the rat glutamate-cysteine ligase catalytic subunit(Abstract). Free Radic Biol Med 45: S34, 2008.

41. Libby P. Inflammation in atherosclerosis. Nature 420: 868–874, 2002.42. Liu X, Zweier JL. A real-time electrochemical technique for measure-

ment of cellular hydrogen peroxide generation and consumption: evalua-tion in human polymorphonuclear leukocytes. Free Radic Biol Med 31:894–901, 2001.

43. Liu X, Peyton KJ, Ensenat D, Wang H, Hannink M, Alam J, DuranteW. Nitric oxide stimulates heme oxygenase-1 gene transcription via theNrf2/ARE complex to promote vascular smooth muscle cell survival.Cardiovasc Res 75: 381–389, 2007.

44. Lu SC. Regulation of glutathione synthesis. Mol Aspects Med. In press.45. Maret W, Vallee BL. Thiolate ligands in metallothionein confer redox

activity on zinc clusters. Proc Natl Acad Sci USA 95: 3478–3482, 1998.46. Moellering D, McAndrew J, Patel RP, Cornwell T, Lincoln T, Cao X,

Messina JL, Forman HJ, Jo H, Darley-Usmar VM. Nitric oxide-dependent induction of glutathione synthesis through increased expressionof gamma-glutamylcysteine synthetase. Arch Biochem Biophys 358: 74–82, 1998.

47. Moellering D, Mc Andrew J, Patel RP, Forman HJ, Mulcahy RT, JoH, Darley-Usmar VM. The induction of GSH synthesis by nanomolarconcentrations of NO in endothelial cells: a role for gamma-glutamylcys-teine synthetase and gamma-glutamyl transpeptidase. FEBS Lett 448:292–296, 1999.

48. Moi P, Chan K, Asunis I, Cao A, Kan Y. Isolation of NF-E2-relatedfactor 2 (Nrf2), a NF-E2-like basic leucine zipper transcriptional activatorthat binds to the tandem NF-E2/AP1 repeat of the beta-globin locuscontrol region. Proc Natl Acad Sci USA 91: 9926–9930, 1994.

49. Moskaug JO, Carlsen H, Myhrstad MC, Blomhoff R. Polyphenols andglutathione synthesis regulation. Am J Clin Nutr 81: 277S–283S, 2005.

C819NO PROTECTS EC FROM H2O2 VIA ZINC AND GLUTATHIONE

AJP-Cell Physiol • VOL 296 • APRIL 2009 • www.ajpcell.org

on February 10, 2010 ajpcell.physiology.org

Dow

nloaded from

50. Motterlini R, Foresti R, Intaglietta M, Winslow RM. NO-mediatedactivation of heme oxygenase: endogenous cytoprotection against oxida-tive stress to endothelium. Am J Physiol Heart Circ Physiol 270: H107–H114, 1996.

51. Mulcahy RT, Gipp JJ. Identification of a putative antioxidant responseelement in the 5$-flanking region of the human gamma-glutamylcysteinesynthetase heavy subunit gene. Biochem Biophys Res Commun 209:227–233, 1995.

52. Muller PY, Janovjak H, Miserez AR, Dobbie Z. Processing of geneexpression data generated by quantitative real-time RT-PCR. Biotech-niques 32: 1372–1379, 2002.

53. Nakagawa H, Hasumi K, Woo JT, Nagai K, Wachi M. Generation ofhydrogen peroxide primarily contributes to the induction of Fe(II)-depen-dent apoptosis in Jurkat cells by (*)-epigallocatechin gallate. Carcino-genesis 25: 1567–1574, 2004.

54. Nakamura S, Kugiyama K, Sugiyama S, Miyamoto S, Koide S,Fukushima H, Honda O, Yoshimura M, Ogawa H. Polymorphism inthe 5$-flanking region of human glutamate-cysteine ligase modifier subunitgene is associated with myocardial infarction. Circulation 105: 2968–2973, 2002.

55. Pearce LL, Gandley RE, Han W, Wasserloos K, Stitt M, Kanai AJ,McLaughlin MK, Pitt BR, Levitan ES. Role of metallothionein in nitricoxide signaling as revealed by a green fluorescent fusion protein. ProcNatl Acad Sci USA 97: 477–482, 2000.

56. Rauen U, Li T, Ioannidis I, de Groot H. Nitric oxide increases toxicityof hydrogen peroxide against rat liver endothelial cells and hepatocytes byinhibition of hydrogen peroxide degradation. Am J Physiol Cell Physiol292: C1440–C1449, 2007.

57. Reaves SK, Fanzo JC, Arima K, Wu JY, Wang YR, Lei KY. Expres-sion of the p53 tumor suppressor gene is up-regulated by depletion ofintracellular zinc in HepG2 cells. J Nutr 130: 1688–1694, 2000.

58. Rubbo H, Parthasarathy S, Barnes S, Kirk M, Kalyanaraman B,Freeman BA. Nitric oxide inhibition of lipoxygenase-dependent liposomeand low-density lipoprotein oxidation: termination of radical chain prop-agation reactions and formation of nitrogen-containing oxidized lipidderivatives. Arch Biochem Biophys 324: 15–25, 1995.

59. Spahl DU, Berendji-Grun D, Suschek CV, Kolb-Bachofen V, KronckeKD. Regulation of zinc homeostasis by inducible NO synthase-derivedNO: nuclear metallothionein translocation and intranuclear Zn2' release.Proc Natl Acad Sci USA 100: 13952–13957, 2003.

60. St Croix CM, Satoh M, Morgan BJ, Skatrud JB, Dempsey JA. Role ofrespiratory motor output in within-breath modulation of muscle sympa-thetic nerve activity in humans. Circ Res 85: 457–469, 1999.

61. St Croix CM, Stitt MS, Leelavanichkul K, Wasserloos KJ, Pitt BR,Watkins SC. Nitric oxide-induced modification of protein thiolate clustersas determined by spectral fluorescence resonance energy transfer in liveendothelial cells. Free Radic Biol Med 37: 785–792, 2004.

62. St Croix CM, Wasserloos KJ, Dineley KE, Reynolds IJ, Levitan ES,Pitt BR. Nitric oxide-induced changes in intracellular zinc homeostasis

are mediated by metallothionein/thionein. Am J Physiol Lung Cell MolPhysiol 282: L185–L192, 2002.

63. Suschek CV, Briviba K, Bruch-Gerharz D, Sies H, Kroncke KD,Kolb-Bachofen V. Even after UVA-exposure will nitric oxide protectcells from reactive oxygen intermediate-mediated apoptosis and necrosis.Cell Death Differ 8: 515–527, 2001.

64. Suschek CV, Fehsel K, Kroncke KD, Sommer A, Kolb-Bachofen V.Primary cultures of rat islet capillary endothelial cells. Constitutive andcytokine-inducible macrophagelike nitric oxide synthases are expressedand activities regulated by glucose concentration. Am J Pathol 145:685–695, 1994.

65. Suschek CV, Schnorr O, Hemmrich K, Aust O, Klotz LO, Sies H,Kolb-Bachofen V. Critical role of L-arginine in endothelial cell survivalduring oxidative stress. Circulation 107: 2607–2614, 2003.

66. Tang ZL, Wasserloos KJ, Liu X, Stitt MS, Reynolds IJ, Pitt BR, StCroix CM. Nitric oxide decreases the sensitivity of pulmonary endothelialcells to LPS-induced apoptosis in a zinc-dependent fashion. Mol CellBiochem 234–235: 211–217, 2002.

67. Test ST, Weiss SJ. Quantitative and temporal characterization of theextracellular H2O2 pool generated by human neutrophils. J Biol Chem 259:399–405, 1984.

68. Urata Y, Yamamoto H, Goto S, Tsushima H, Akazawa S, YamashitaS, Nagataki S, Kondo T. Long exposure to high glucose concentrationimpairs the responsive expression of gamma-glutamylcysteine synthetaseby interleukin-1beta and tumor necrosis factor-alpha in mouse endothelialcells. J Biol Chem 271: 15146–15152, 1996.

69. Von der Leyen HE, Dzau VJ. Therapeutic potential of nitric oxidesynthase gene manipulation. Circulation 103: 2760–2765, 2001.

70. Wassmann S, Wassmann K, Nickenig G. Modulation of oxidant andantioxidant enzyme expression and function in vascular cells. Hyperten-sion 44: 381–386, 2004.

71. Wild AC, Moinova HR, Mulcahy RT. Regulation of gamma-glutamyl-cysteine synthetase subunit gene expression by the transcription factorNrf2. J Biol Chem 274: 33627–33636, 1999.

72. Wiseman DA, Wells SM, Hubbard M, Welker JE, Black SM. Alter-ations in zinc homeostasis underlie endothelial cell death induced byoxidative stress from acute exposure to hydrogen peroxide. Am J PhysiolLung Cell Mol Physiol 292: L165–L177, 2007.

73. Wu G, Fang YZ, Yang S, Lupton JR, Turner ND. Glutathione metab-olism and its implications for health. J Nutr 134: 2004.

74. Yamasaki S, Sakata-Sogawa K, Hasegawa A, Suzuki T, Kabu K, SatoE, Kurosaki T, Yamashita S, Tokunaga M, Nishida K, Hirano T. Zincis a novel intracellular second messenger. J Cell Biol 177: 637–645, 2007.

75. Yang H, Magilnick N, Lee C, Kalmaz D, Ou X, Chan JY, Lu SC. Nrf1and Nrf2 regulate rat glutamate-cysteine ligase catalytic subunit transcrip-tion indirectly via NF-kappaB and AP-1. Mol Cell Biol 25: 5933–5946,2005.

C820 NO PROTECTS EC FROM H2O2 VIA ZINC AND GLUTATHIONE

AJP-Cell Physiol • VOL 296 • APRIL 2009 • www.ajpcell.org

on February 10, 2010 ajpcell.physiology.org

Dow

nloaded from