Toxicology and Applied Pharmacology 241 (2009) 81–89

Contents lists available at ScienceDirect

Toxicology and Applied Pharmacology

j ourna l homepage: www.e lsev ie r.com/ locate /ytaap

Activation of Nrf2 by cadmium and its role in protection against cadmium-inducedapoptosis in rat kidney cells

Jun Chen 1, Zahir A. Shaikh ⁎Department of Biomedical and Pharmaceutical Science, and Center for Molecular Toxicology, College of Pharmacy, University of Rhode Island, Kingston, RI 02881, USA

Abbreviations: Cd, cadmium; DCFH-DA, 2,7-dichloroKeap1, Kelch-like ECH-associated protein 1; ARE, antioxfluorescence activated cell sorting; FITC, fluorescein isocysteine ligase catalytic subunit; HO-1, heme oxygeerythroid 2 related factor 2; PI, propidium iodide; RsiRNA-Nrf2, small interfering RNA for Nrf2; siRNA-cnegative control.⁎ Corresponding author. Fax: +1 401 874 2646.

E-mail address: ZShaikh@uri.edu (Z.A. Shaikh).1 Present address: Department of Occupational and E

of Education Key Laboratory of Environment and Health,Medical College, Huazhong University of Science andPeople’s Republic of China.

0041-008X/$ – see front matter © 2009 Elsevier Inc. Adoi:10.1016/j.taap.2009.07.038

a b s t r a c t

a r t i c l e i n f o

Article history:Received 29 June 2009Accepted 25 July 2009Available online 12 August 2009

Keywords:CadmiumNRK-52E cellsNrf2Reactive oxygen speciesApoptosis

Kidney is the primary target organ in chronic cadmium (Cd) toxicity, and oxidative stress plays an importantrole in this process. The nuclear transcription factor Nrf2 binds to antioxidant response elements (AREs) andregulates genes involved in protecting cells from oxidative damage. Whether kidney cells respond to Cd byactivating Nrf2 is unknown. This study was designed to examine the Cd-induced activation of Nrf2transcriptional activity in a stable rat kidney cell line, NRK-52E, and to investigate the protection this mightoffer against apoptosis. The cells were treated with 5–20 μM CdCl2 for 5 h, followed by a recovery period ofup to 24 h. A concentration-dependent increase (up to 2.9-fold) in the level of reactive oxygen species wasnoted upon termination of 5-h Cd treatment. The Nrf2-ARE binding activity also increased and peaked (6.1-fold) at 10 μM Cd concentration. Time-course study revealed that the binding activity increased at 1 h of Cdtreatment and peaked 2 h post Cd treatment. Apoptosis was detected 6 h post treatment with Cd and aconcentration- and time-dependent increase in the apoptotic cell population occurred during the next 18 h.Over-expression of Nrf2 by transient transfection conferred resistance against Cd-induced apoptosis.Conversely, suppression of Nrf2 expression by specific siRNA resulted in greater sensitivity of the cells to Cdby decreasing the levels of two antioxidant enzymes, hemeoxygenase-1 and glutamate-cysteine ligase.Taken together, these results suggest that in kidney cells the activation of Nrf2 is an adaptive intracellularresponse to Cd-induced oxidative stress, and that Nrf2 is protective against Cd-induced apoptosis.

© 2009 Elsevier Inc. All rights reserved.

Introduction

Cd is used mainly in metallurgy, battery manufacturing, andpigment production. Environmental exposure occurs mainly via thediet, drinking water, and cigarette smoke (ATSDR, 1989). Cd toxicitydue to dietary exposure has been widely recognized in Japan as Itai-itai disease, a chronic ailment with renal tubular dysfunction anddebilitating osteomalacia and osteoporesis (Takebayashi et al., 2000).The release of metallothionein-bound Cd from the initial depot, liver,and subsequent accumulation in the proximal tubular cells of thekidney is believed to be responsible for Cd-induced nephrotoxicity(Vestergaard and Shaikh, 1994; Klaassen and Liu, 1997).

-dihydrofluorescein diacetate;idant response element; FACS,thiocyanate; GCLC, glutamate-nase-1; Nrf2, nuclear factorOS, reactive oxygen species;ontrol, small interfering RNA

nvironmental Health, MinistrySchool of Public Health, TongjiTechnology, Wuhan, 430030,

ll rights reserved.

Cell death via apoptosis plays an important role not only inphysiological processes such as embryogenesis and normal cellturnover, but also in pathological conditions such as tumor regression,degenerative diseases, and in chemical-induced cytotoxicity (Elmore,2007). Indeed, apoptosis constitutes a major mode of cell death in Cdtoxicity and has been demonstrated in a variety of cell types, such asHeLa cells (Szuster-Ciesielska et al., 2000), mouse thymocytes (Donget al., 2001), rat C6 glioma cells (Wätjen and Beyersmann, 2004),HepG2 cells (Oh and Lim, 2006), NRK-52E cells (Xie and Shaikh, 2006;Lee et al., 2007), pig kidney cells (Liu et al., 2007), T lymphocytes(Pathak and Khandelwal, 2008), primary rat Sertoli cells (Yu et al.,2008), and mouse mesangial cells (Liu and Templeton, 2008), andmouse embryonic fibroblasts (He et al., 2008). However, in certainother cell types, such as rat pheochromocytoma cells, L929 mousefibroblasts, 3T3L1 mouse preadipocytes, VH16 human fibroblasts, andA549 human lung adenocarcinoma cells, Wätjen et al. (2002) failed toobserve Cd-induced apoptotic DNA fragmentation. Furthermore, inCHO cells, Cd blocked apoptosis induced by a variety of agents byinhibiting caspase-3 (Yuan et al., 2000). Similarly, in rat mesangialcells, Cd (10 μM) was anti-apoptotic by a mechanism that may alsoinvolve caspase inhibition (Gunawardana et al., 2005). These reportssuggest a cell type dependence of Cd-induced cytotoxicity.

Nuclear factor erythroid 2 related factor 2 (Nrf2) is a basic leucinezipper transcription factor that binds to antioxidant responsive element

82 J. Chen, Z.A. Shaikh / Toxicology and Applied Pharmacology 241 (2009) 81–89

(ARE) and is a chief regulator of a battery of cytoprotective genes(Zhang, 2006; Kensler et al., 2007). For instance, Nrf2-regulatedantioxidant genes are temporally altered in global gene expressionarrays during the progression of pulmonary emphysema in A/J mice;there is reduced expression of Nrf2-regulated antioxidant genes in thelungs of A/J mice in cigarette-smoke induced emphysema (Rangasamyet al., 2009). The products of these cytoprotective genes, either a subsetof drug-metabolizing enzymes such as glutathione S-transferases andNAD(P)H: quinine oxidoreductase 1 (Korashy and El-Kadi, 2006) or asubset of antioxidant-related enzymes such as heme oxygenase 1 (HO-1; Chen et al., 2005) and glutamate-cysteine ligase (GCL; also known asγ-glutamylcysteine synthetase; Yang et al., 2005), elicit protectionagainst chemical toxicity, oxidative/electrophilic stress, and certainchronic disorders. This is considered to be one of the most importantmechanisms bywhich cells neutralize the effects of various stresses andsurvive (Jaiswal, 2004). Under normal condition, Nrf2 is sequestered inthe cytoplasm by Kelch-like ECH-associated protein 1 (Keap1), whichfunctions as a negative regulator and drives Nrf2 to rapid degradationthrough the ubiquitin proteasome system (Motohashi and Yamamoto,2004; Kobayashi et al., 2006). Upon exposure of cells to chemopreven-tive agents and in response oxidative stress, Nrf2dissociates fromKeap1and translocates into the nucleus where it forms a heterodimer with itsobligatory partner, Maf, and ultimately activates ARE-dependent geneexpression (Zhang 2006; Kimura et al., 2007). Cd stabilizes Nrf2 andprevents it from degradation via the ubiquitin proteasome in mousehepatoma cells (Stewart et al., 2003) and mouse embryonic fibroblasts(He et al., 2008). Nrf2 is also activated by Cd in vivo (Abu-Bakar et al.,2004; Casalino et al., 2007) and in cultured cells such as vascularendothelial cells (Sakurai et al., 2005) andmouse embryonic fibroblasts(He et al., 2008). It remains unknown, however, whether Cd causesNrf2activation in kidney-derived cells.

While it appears that Nrf2-dependent ARE-driven detoxificationand antioxidant proteins are major contributing factors for Nrf2-conferred cellular protection, emerging evidence shows that Nrf2might also be involved in regulating apoptosis signaling pathways(Lee and Johnson, 2004). For example, Kotlo et al. (2003) found thatover-expression of Nrf2 protected cells from Fas-induced apoptosis.Also, mouse embryonic fibroblast cells lacking Nrf2 exhibited elevatedROS production and apoptosis which was markedly elevated bychromium (VI), suggesting a protective role of Nrf2 in its toxicity (Heet al., 2007). Similar results were reported by these investigators forCd (He et al., 2008). Moreover, the products of downstream genes thatare regulated by Nrf2, such as GCL catalytic subunit (GCLC) and HO-1,possess anti-apoptotic properties (Fan et al., 2005; Okouchi et al.,2006; Olszanecki et al., 2007; Tan et al., 2007), which presumably isthe basis of the anti-apoptotic effects of Nrf2.

The purpose of the present study was to examine the effect of Cdtreatment on Nrf2 transcriptional activity and to evaluate whether itoffered protection against Cd-induced apoptosis.

Materials and methods

Materials. Cd chloride (CdCl2), Dulbecco's modified Eagle's medium(DMEM), calf serum, trypsin, ethylenediaminetetraacetic acid (EDTA),penicillin/streptomycin, protease inhibitor cocktail, dithiothreitol(DTT), phenylmethylsulfonyl fluoride (PMSF), bovine serum albumin(BSA), glycine, 4-(2-hydroxyethyl) piperazine-1-ethanesulfonic acid(HEPES), 2,7-dichloro-dihydrofluorescein diacetate (DCFH-DA) wereobtained from Sigma (St. Louis, MO). IRDye 680-labeled goat anti-rabbit secondary antibody, electrophoretic mobility shift assay(EMSA) buffer kit, and IRDye 700 infrared dye end-labeled double-stranded oligonucleotide probe containing a Nrf2-consensussequence or ARE-consensus sequence were purchased from LI-CORBiosciences (Lincoln, NE). Lipofectamine 2000 transfection reagent,pCMV-SPORT6 vector, TrypLE Express cell dissociation enzyme andOpti-MEM I Reduced Serum Media were from Invitrogen (Carlsbad,

CA). The Nrf2 expression plasmid (pCMV-SPORT6-Nrf2) was fromOpen Biosystems (Huntsville, AL). Silencer Select siRNA for Nrf2 andnegative control siRNA were obtained from Applied Biosystems(Foster City, CA). Annexin V-Fluorescein-5-isothiocyanate (AnnexinV-FITC), apoptosis detection kit and flow cytometer (FACSCalibur)were from BD Biosciences (San Diego, CA). Rabbit polyclonal anti-Nrf2and heme oxygenase 1 were from Santa Cruz Biotechnology, Inc.(Santa Cruz, CA). Rabbit polyclonal anti-GCL antibody was fromThermo Fisher Scientific Anatomical Pathology (Fremont, CA). Micro-BCA protein assay kit was obtained from Pierce (Rockford, IL).

Cell culture and treatment. NRK-52E cells, a stable cell line derivedfrom rat kidney proximal tubules, were purchased from the AmericanType Culture Collection (Manassas, VA). The cells were cultured inDMEM containing 1.5 g/L sodium bicarbonate, 5% newborn calfserum, 100 IU/mL penicillin and 100 μg/mL streptomycin in anatmosphere of 5% CO2–95% air at 37 °C. Concentrated stock solutionfor Cd (20 mM) was prepared in phosphate buffered saline (PBS)followed by filter sterilization. Just prior to the cell treatment, freshdilutions of Cd stock solution, at the final concentrations needed, wereprepared in the culture medium containing no serum. The cells (60–70% confluent) were treated with Cd for up to 5 h. In someexperiments, the cells werewashed to remove Cd-containingmediumand incubated in DMEM containing 5% calf serum for up to 24 h post-treatment.

Determination of intracellular reactive oxygen species (ROS)production. DCFH-DA was used for ROS detection. DCFH-DA enterscells by simple diffusion and is trapped within the cells asdichlorofluorescein free base (DCFH), which is further oxidized byROS to form the fluorescent compound DCF by deacetylation via anesterase. Although DCFH-DA is generally thought to react with varioustypes of ROS and used to study oxidative stress in cells, H2O2 is themajor ROS that oxidizes DCFH-DA (LeBel et al., 1992). The cells wereseeded in a 96-well microplate at a density of 5000 cells per well. Afterpreloading with 10 μM DCFH-DA at 37 °C for 30 min, the cells werewashed with PBS and treated with different concentrations of Cd asdescribed above. Fluorescence intensity was determined bySpectraMax M2 Microplate Readers (Molecular Devices, Sunnyvale,CA) using 485 nm excitation and 538 nm emission filters.

Preparation of cytoplasmic and nuclear extracts. The cells werecultured in the 6-well plates, washed with ice-cold PBS andsuspended in 150 μL/well of ice-cold lysis buffer containing 10 mMHEPES–NaOH (pH 7.9), 10 mM KCl, 1 mM EDTA, 1 mM dithiothreitol(DTT), 0.5 mM phenylmethylsulfonyl fluoride (PMSF), and 1% (v/v)protease inhibitor cocktail. The cells were lysed in the presence of 0.2%Nonidet P-40 (NP-40) on ice for 15minwith occasional vortexing. Celllysates were centrifuged for 1 min at 14,000×g at 4 °C and thesupernatant was saved as cytoplasmic extract. The nuclear pellet wasresuspended in nuclear extraction buffer containing 20 mM HEPES(pH 7.9), 420 mM NaCl, 1 mM EDTA, 1 mM DTT, 1 mM PMSF and 1%(v/v) protease inhibitor cocktail on ice, with gentle shaking for 30min, and then centrifuged at 14,000×g for 15 min at 4 °C. Thesupernatant (nuclear extract) was either stored at −80 °C or wasquantified freshly using Micro BCA protein assay kit.

Electrophoretic mobility shift assay (EMSA). Nrf2 binding to the AREwas determined by non-radioisotopic EMSA. The oligonucleotides (5′-TGG GGA ACC TGT GCT GAG TCA CTG GAG-3′ and 3′-ACC CCT TGG ACACGA CTC AGT GAC CTC) were labeled with an infrared dye (IRD700; LI-COR Biosciences, Lincoln, NE) and annealed to generate double-stranded probes for EMSA. For DNA-binding reaction, the nuclearextract (5 μg protein) was incubated in the dark for 30 min in a finalvolume of 20 μL binding reaction buffer containing 10mMTris–HCl (pH7.5), 50 mM KCl, 2.5 mM DTT, 2.5% glycerol, 1 μg poly(dI-dC), 0.5 μg

Fig. 1. Concentration dependence of ROS generation in cells treated with Cd. The cellswere cultured in 96-well plates and preloaded with the fluorescent probe DCFH-DA(10 μM) for 30 min at 37 °C. After washing with PBS, the cells were treated with up to20 μM Cd in serum-free DMEM for 5 h. The cells were washed with PBS and detachedimmediately with trypsin/EDTA. The harvested cells were suspended in 100 μL PBS, andthe fluorescence intensity was determined at excitation wavelength of 480 nm andemission wavelength of 520 nm. Data plotted are mean±SE of three independentexperiments. ⁎Significantly greater than the cells not treated with Cd (Pb0.05).

Fig. 2. Concentration dependence of Nrf2 DNA binding activity in Cd-treated cells.Nuclear proteins were extracted from cells treated with up to 20 μM Cd in serum-freeDMEM for 5 h. After electrophoresis on 6% non-denaturing polyacrylamide gel, Nrf2DNA binding was visualized and quantified using the LI-COR Odyssey Imaging System.Data plotted are mean±SD of three independent experiments. ⁎Significantly greaterthan the cells not treated with Cd (Pb0.05).

83J. Chen, Z.A. Shaikh / Toxicology and Applied Pharmacology 241 (2009) 81–89

sheared salmon sperm DNA, and 50 fmol infrared dye end-labeledprobes. Orange loading dye (2 μL, 10×) was added to each sample priorto gel electrophoresis. The samples were loaded in 6% non-denaturingpolyacrylamide gels and run for 45min at 120 V. The gels were scannedusing the Odyssey Imaging System (LI-COR Biosciences, Lincoln, NE).The optical density of the bands was background-corrected andquantified using LI-COR Odyssey Application Software (Version1.2.15). All experiments were repeated at least three times.

Transient transfection with Nrf2 plasmid and siRNA. NRK-52E cellswere plated in six-well plates at a density of 2×105 cells per well in2 mL DMEM.When the cells reached 60–70% confluence the next day,transient transfection was carried out using Lipofectamine 2000according to themanufacturer's instructions. pCMV-SPORT6-Nrf2 andpCMV-SPORT6 were used for over-expression of Nrf2 and controlvectors, respectively. For each transfection, 2 μg plasmid and 3 μLLipofectamine 2000 were diluted in 100 μL Opti-MEM I ReducedSerum Medium, respectively. After incubation for 5 min at roomtemperature, the diluted plasmids and Lipofectamine 2000 werecombined and further incubated for 20 min at room temperature. Thecell culture medium was replaced with 800 μL DMEM withoutantibiotics and the DNA-Lipofectamine 2000 complexes were added.After 5 h, the medium was changed to 2 mL DMEM containingantibiotics and the cells were further incubated for 24 h at 37 °C.

For Nrf2 knockdown, Silencer Select Pre-designed siRNA for Nrf2(sense 5′-GCACGGUGGAGUUCAUGAtt-3′, antisense 5′-UCAUUGAA-CUCCACCGUGCct-3′) was used. Silencer negative control 1 siRNA(ABI) with no known mammalian homology was used as negativecontrols (siRNA-control). Cell cultures of 50–60% confluence wereprepared in six-well plates and siRNA was introduced using Lipofec-tamine 2000 according to themanufacturer's instructions. In brief, 1 μLsiRNA (5 μM solution) and 3 μL Lipofectamine 2000 were dilutedseparately with 100 μL Opti-MEM I Reduced SerumMedium and keptat room temperature for 5 min. The diluted siRNA and Lipofectamine2000 were mixed gently followed by incubation for 20 min at roomtemperature. 200 μL of siRNA-Lipofectamine complex was added toeach well containing 800 μL DMEM without antibiotics. After a 5-hincubation of cells with siRNA-Lipofectamine 2000 complex, themedium was replaced with normal DMEM, and the cells weremaintained for an additional 19 h. Cells were either harvested forprotein extraction or treated with CdCl2. Nrf2 over-expression andknockdown were determined by Western blot analysis.

Western blot analysis. A total of 20 μg cytoplasmic protein from eachsample wasmixed with loading buffer (50mM Tris, 10% glycerol, 0.1%SDS, 3 mM 2-mercaptoethanol, and 0.005% bromophenol blue),heated for 4 min at 95 °C, and loaded onto a 7.5% sodium dodecylsulfate-polyacrylamide gel (SDS-PAGE) for electrophoresis. Afterelectrophoresis at 160 V for 50 min, the proteins were transferredto nitrocellulose membrane at 20 V for 20 min using Trans-Blot SDSemi-dry Transfer Cell (Bio-Rad, Hercules, CA). Membranes were thenblocked with Odyssey blocking buffer (LI-COR Biosciences, Lincoln,NE) for 60 min and incubated overnight at 4 °C with the appropriateprimary antibodies diluted with the blocking buffer. Each membranewas washed with PBS containing 0.1% Tween 20 four times andincubated with IRDye680-labeled goat anti-rabbit secondary antibodydiluted (1:10,000) with the blocking buffer containing 0.1% Tween 20for 1 h. The bands on the membranes were scanned and quantifiedwith β-actin normalization using Odyssey Imaging System andApplication Software (LI-COR Biosciences, Lincoln, NE).

Apoptosis analysis. Apoptotic cells were determined by fluorescenceactivated cell sorter (FACS) analysis using Annexin-V-FITC andpropidium iodide (PI) double staining kit, according to themanufacturer's instructions. In brief, after Cd treatment the cells insix-well plates were washed twice with cold PBS, dislodged by TrypLE

Express cell dissociation enzyme (Invitrogen, Carlsbad, CA), and re-suspended in 1× binding buffer at a concentration of 1×105 cells/mL.Next, 5 μL Annexin V-FITC and 5 μL PI were added to the cell suspension(100 μL), followedby incubation for 15min at room temperature. To thestained cell suspension, 300 μL 1× binding buffer was added and thecells were subjected to FACS analysis. Data for 1×104 cells wereacquired within 1 h after staining and analyzed using the CELLQuestsoftware program (BD Biosciences, San Diego, CA). The cell populationwas sorted into viable cells (both Annexin V-FITC and PI negative),apoptotic cells (Annexin V-FITC positive, but PI negative), and necroticcells (both Annexin V-FITC and PI positive).

Statistical analyses. Data were analyzed by t-test or by one-wayanalysis of variance (ANOVA) followed by Tukey's post hoc test,

84 J. Chen, Z.A. Shaikh / Toxicology and Applied Pharmacology 241 (2009) 81–89

where appropriate, using SPSS 13.0 for Windows (SPSS, Inc., Chicago,IL). Pb0.05 was considered statistically significant.

Results

Reactive oxygen species (ROS) generation

High levels of ROS within a cell have a number of direct andindirect consequences on cell signaling pathways and may lead toapoptosis (Miyaguchi et al., 2004; England and Cotter, 2005). Toestablish that the elevation of ROS levels occurred prior to apoptosis inCd-treated cells, DCFH-DA was used to measure ROS production. Asshown in Fig. 1, treatment of NRK-52E cells with 5–20 μM Cd for 5 hunder serum-free condition significantly increased the level of ROS ascompared to the untreated cells. The effect of Cd on ROS generationwas concentration-dependent, with 20 μM Cd resulting in the highestincrease (2.9-fold).

Nrf2-ARE binding activity

Cd treatment is reported to delay the rate of Nrf2 degradation,resulting in its translocation into the nucleus and transcription ofdownstream genes (Stewart et al., 2003; Casalino et al., 2007). Todetermine whether Cd activated Nrf2 transcriptional activity inNRK-52E cells, EMSA was performed to measure the Nrf2-AREbinding activity. Fig. 2A and B depicts that treatment with 5-

Fig. 3. Time dependence of Nrf2 DNA binding activity in Cd-treated cells. Nuclear proteins wincubation with serum-containing medium for up to 12 h. The control cells were culturedpolyacrylamide gel, Nrf2 DNA binding was visualized and quantified by the LI-COR Odys⁎Significantly greater than the control cells (Pb0.05).

20 μM Cd for 5 h significantly increased Nrf2-ARE binding activity,with the peak (6-fold increase over control) at the 10-μMconcentration.

The time course of Nrf2 DNA binding at 20 μMCd is summarized inFig. 3A and B. It shows that the Nrf2-ARE binding activity started toincrease from 1 h of Cd treatment and peaked (3.4-fold increase overcontrol) at 2 h following the 5-h Cd treatment. Return to basal levelwas achieved by 12 h post-treatment.

Apoptosis

Fig. 4 illustrates representative FACS plots of Annexin-V-FITC andPI double-stained cells at 12 h post-treatment with up to 20 μMCd for5 h. Based on the plot for the positive control, Camptothesin, theapoptotic cells were identified in the lower right quadrant of the plotsand show a concentration-dependent shift in cells undergoingapoptosis.

The FACS data for 6, 12, and 24 h post-treatment wereanalyzed with the aid of CELLQuest software and plotted (Fig. 5).The results show that Cd treatment resulted in apoptotic cell deathnot only in a concentration-dependent but also in a time-dependent manner. A small but significant increase in apoptosis,compared to the cells not treated with Cd, was evident at thelowest concentration of Cd (5 μM) used, as early as 6 h post-treatment. The highest apoptotic cell death (about 60%) occurredat 20 μM Cd, 24 h post-treatment.

ere extracted from cells treated with 20 μM Cd in serum-free DMEM for 5 h, followed bywithout Cd in serum-free DMEM for 5 h. After electrophoresis on 6% non-denaturingsey Imaging System. Data plotted are mean±SD of three independent experiments.

Fig. 4. Representative scatter plots of cells treated with various concentrations of Cd. The cells were seeded in six-well plates and treatedwith up to 20 μMCd in serum-free DMEM for5 h, washed with PBS, and incubated in serum-containing medium. After 12 h, the cells were harvested and single cell suspensions were stained with Annexin V-FITC and PI andanalyzed by FACS. Camptothesin was used as a positive control. The scatter plots were divided into four quadrants: lower left quadrant (viable cells: both Annexin V-FITC and PInegative), lower right quadrant (apoptotic cells: Annexin V-FITC positive, but PI negative), and upper left and right quadrants (necrotic cells: both Annexin V-FITC and PI positive).

85J. Chen, Z.A. Shaikh / Toxicology and Applied Pharmacology 241 (2009) 81–89

Nrf2 levels following gene transfection or knockdown

Western blot analysis was used to evaluate the success of thetransient gene transfection and knockdown procedures. Transienttransfection with pCMV-SPORT6-Nrf2 plasmid increased the cyto-plasmic Nrf2 level about 2.7-fold, as compared to the cells transfectedwith the control vector (Fig. 6A and B), indicating the success of thetransfection. In comparison, when the Nrf2 gene was knocked downby siRNA-Nrf2, the basal Nrf2 expressionwas only about 60% of that ofthe control (Fig. 7A and B).

Fig. 5. Concentration and time dependence of Cd-induced apoptosis in NRK-52E cells.The cells were seeded in six-well plates and treated with up to 20 μM Cd in serum-freeDMEM for 5 h, washed with PBS, and incubated in serum-containing medium for 6, 12or 24 h. After harvesting the cells, the single-cell suspensions were stained withAnnexin V-FITC and PI and analyzed by FACS. Camptothesin was used as a positivecontrol. The scatter plot data were analyzed using the CELLQuest software program andplotted as mean±SD of three independent experiments. ⁎Significantly greater than thecells not treated with Cd and harvested at the same time point (Pb0.05).

The cells in which the Nrf2 gene was knocked down were alsotreated with 10 μM Cd to see if Cd merely stabilized the Nrf2 proteinlevels or caused an induction. As shown in Fig. 6A and B, whereas the

Fig. 6. Increase of Nrf2 expression in NRK-52E cells by transient transfection with Nrf2plasmid. The cells were grown in six-well plates overnight (60–70% confluent) and,using Lipofectamine 2000, transfected with pCMV-SPORT vector or pCMV-SPORT-Nrf2plasmid. The cytoplasmic proteins were extracted and Nrf2 expression was determinedby Western blot analysis. β-Actin was used as a reference protein. The protein bandswere visualized and quantified with the aid of LI-COR Odyssey Imaging System. (A)Representative Western blots from one experiment. (B) Mean±SD of data from threeindependent experiments. ⁎Significantly greater than the vector control cells (Pb0.05).

Fig. 8. Effect of modulation of Nrf2 expression on apoptosis in Cd-treated cells. Asdescribed in legends for Figs. 6 and 7, the Nrf2 expression was altered by transienttransfection of cells with either: (A) Nrf2 plasmid or (B) SiRNA. The transfected cellswere treated with up to 10 μM Cd for 5 h, washed with PBS, and incubated in serum-containing medium for 24 h. Upon harvesting, single cell suspensions were preparedand double-stained with Annexin V-FITC and PI to detect the apoptotic cells by FACS.The scatter plot data were analyzed by the CELLQuest program to quantify relative

86 J. Chen, Z.A. Shaikh / Toxicology and Applied Pharmacology 241 (2009) 81–89

Nrf2 level increased 3.7-fold in the siRNA-control cells, only a modestincrease (1.4-fold over the siRNA-control) was evident in the Cd-treated siRNA-Nrf2 cells, suggesting that Cd somehow induced Nrf2synthesis.

Protection against Cd-induced apoptosis by Nrf2

In order to explore whether over-expression of Nrf2 would protectagainst Cd-induced apoptosis, 24 h prior to a 5-h treatment with Cdthe cells were transfected with pCMV-SPORT6-Nrf2. The cells wereharvested 24 h after the Cd treatment and single cell suspensionswere prepared for Annexin V-FITC and PI double-staining anddetection of apoptotic cells. As shown in Fig. 8A, at both 5 and10 μM Cd the cells transfected with pCMV-SPORT6-Nrf2 hadsignificantly fewer apoptotic cells than the cells transfected withpCMV-SPORT6 vector, suggesting a protective role of Nrf2 against Cd-induced apoptosis. The reduction in apoptotic cells was about 50% at5 μM and about 30% at 10 μM Cd.

In a parallel experiment inwhich the Nrf2 genewas knocked downby siRNA-Nrf2 transfection, the cells exhibited greater sensitivity toCd-induced apoptosis, as compared to the cells transfected withsiRNA-control (Fig. 8B). While the control cells showed no effect of2.5 μM Cd on apoptosis 24 h following treatment, the cells transfectedwith siRNA-Nrf2 showed more than two-fold increase in apoptosisover the basal level. A significant increase in apoptosis was alsoobserved in the siRNA-Nrf2 transfected cells at the higher concentra-tion of Cd (5 μM).

Role of HO-1 and GCLC in Cd-induced apoptosis

To determine whether two downstream genes that are regulatedby Nrf2 might be responsible for protection against Cd-inducedapoptosis, changes in the expression of HO-1 and GCLC wereinvestigated following Nrf2 gene knock down. Treatment with

Fig. 7. Suppression of Nrf2 expression in NRK-52E cells by transient transfection withsiRNA and its effect on Cd-mediated increase in Nrf2 levels. The cells were grown in six-well plates overnight (60–70% confluent) and transfectedwith Silencer negative control1 siRNA (siRNA-control) or Silencer Select Pre-designed siRNA for Nrf2 (siRNA-Nrf2).The cells were treated 24 h later with 10 μM Cd for 5 h, washed with PBS, and incubatedin serum-containing medium for 24 h. Upon harvesting the cells, the cytoplasmicproteins were extracted and Nrf2 expression was determined by Western blot analysis.β-Actin was used as a reference protein. The protein bands were visualized andquantified with the aid of LI-COR Odyssey Imaging System. (A) Representative Westernblots from one experiment. (B)Mean±SD of data from three independent experiments.⁎Significantly greater than the cells transfected with siRNA-control (Pb0.05).

number of apoptotic cells. Mean±SD of results from three independent experimentswere plotted. ⁎Significantly greater than the corresponding control cells not treatedwith Cd (Pb0.05). #Significantly different from the corresponding (A) vector-transfected or (B) siRNA-control cells treated with the same concentration of Cd(Pb0.05).

10 μM Cd for 5 h followed by 6 h incubation in serum-containingmedium resulted in an 18.6-fold increase in HO-1 expression in thesiRNA-control cells (Fig. 9A and B). The HO-1 level was significantlyattenuated (66%) by siRNA-Nrf2 transfection prior to the Cdtreatment. Additionally, while treatment with 10 μM Cd had nosignificant effect on the GCLC expression in the siRNA-control cells,the expression was reduced significantly (about 30%) in the siRNA-transfected cells (Fig. 9C). These results suggest that both GCLC andHO-1 are associated with Nrf2 protection against Cd-inducedapoptosis in the NRK-52E cells.

Discussion

Even though Cd is not a Fenton metal, it has ROS-inducingproperties (Shaikh et al., 1999). ROS are generated following acute Cdoverload and play an important role in tissue damage (Liu et al.,2009). It is thought that Cd displaces other Fenton metals from theirbinding sites, which in turn induce ROS formation (Casalino et al.,1997). Another mechanism by which Cd can generate free radicals isthrough decreasing the activities of antioxidant enzymes such ascatalase, glutathione peroxidase and superoxide dismutase, orintracellular levels of antioxidants such as glutathione (Sen et al.,2004). Our results showed that treatment of NRK-52E cells with 5–20 μM Cd for 5 h increased the intracellular ROS production in aconcentration-dependent manner. This event occurred before Cd-induced apoptosis became evident. Since ROS act as secondmessengers in intracellular signaling cascades (Valko et al., 2006),

Fig. 9. Effect of suppression of Nrf2 expression on HO-1 and GCLC expression in Cd-treated cells. The cells were cultured in six-well plates to reach 50–60% confluence andtransfected with siRNA-Nrf2 to knockdown Nrf2 expression. The transfected cells weretreated with 10 μM Cd in serum-free DMEM for 5 h, washed, and incubated in DMEMwith serum for 6 h. Control cells received no Cd treatment. The cytoplasmic proteinswere extracted from the harvested cells. After electrophoresis on 7.5% polyacrylamidegel, the bands were visualized by Odyssey Imaging System. β-Actin was used as areference protein. (A) Representative Western blot analysis of HO-1 and GCLCexpression in control and Cd-treated cells. The bands were quantified using LI-COROdyssey Imaging System. Expression data for (B) HO-1 and (C) GCLC from threeindependent experiments were plotted as mean±SD. #Significantly greater than therespective siRNA-control or siRNA-Nrf2 cells not treated with Cd (Pb0.05). ⁎Sig-nificantly lower than the siRNA-control cells treated with Cd (Pb0.05).

87J. Chen, Z.A. Shaikh / Toxicology and Applied Pharmacology 241 (2009) 81–89

the increase in ROS by Cd may play an important role in themodulation of gene expression and resultant apoptosis.

Nrf2 is a critical regulator of the cellular antioxidant response andxenobiotic metabolism. Nuclear accumulation of Nrf2 is an essentialsignaling step for its function as a transcription factor (Nguyen et al.,2004). In the present study, treatment of NRK-52E cells with Cdresulted in a significant increase in Nrf2-ARE binding activity in thenuclear fraction; the maximum activity occurring 2 h followingtreatment with 10 μM Cd for 5 h. The expression of a wide array ofantioxidant and detoxification genes is positively regulated by theARE-driven sequence, a cis-acting regulatory element that Nrf2 binds.Thus, Nrf2 may serve as a master regulator of ARE-driven cellulardefense systems against the oxidative stress (Lee et al., 2005). Itappears that the increase in Nrf2-ARE binding activity may be acellular adaptive response against Cd toxicity (He et al., 2008). Themaximum tolerated concentration of Cd for the activation of Nrf2under our experimental conditions was 10 µM. While the cellulardefense system, including the Nrf2 pathway, might alleviate or delaythe deleterious effects of Cd, it is unable to completely overcome thetoxicity of Cd concentrations greater than 2.5 μM, as discussed below.

A possible mechanism for the Nrf2 activation by Cd might be thatexisting Nrf2 escaped ubiquitination and proteasomal degradation,resulting in increased stability of this transcription factor in thecytoplasm and its subsequent nuclear translocation (Stewart et al.,2003; He et al., 2008). Whether Cd also interacts with Nrf2 and/orKeap 1, its negative regulator, requires further study.

While a majority of studies reported that Cd triggered apoptoticcell death in diverse cell types, negative results and even anti-apoptotic action of Cd were also reported in other cells (Wätjen et al.,2002; Gunawardana et al., 2005; Yuan et al., 2000). Our resultsshowed that 5 μM Cd initiated apoptosis in NRK-52E cells as early as6 h following a 5-h treatment. The cell death was probably mediatedthrough ROS generation, which is a well-established early signal ofapoptosis (Simon et al., 2000). In our study, ROS generation promotedapoptosis during the period when the cells were no longer exposed toCd. Our finding that 20 μM Cd induced approximately 60% apoptoticcell death 24 h post-treatment suggests that apoptosis constitutes amajor mode of cell death in NRK-52E cells following Cd treatment.

In order to explore the possible role of Nrf2 in protection againstCd-induced apoptosis, prior to Cd treatment, the cells were transientlytransfected with pCMV-SPORT6-Nrf2 plasmid or siRNA-Nrf2 to induceNrf2 over-expression and gene knockdown, respectively. The over-expression of Nrf2 significantly reduced the apoptotic cell death.Conversely, Nrf2 gene knockdown sensitized the cells to Cd andsignificantly increased the apoptotic cell death in the siRNA-Nrf2group such that 2.5 μM Cd became apoptotic. Based on theseobservations, it seems reasonable to conclude that Nrf2 activationconfers protection against Cd-induced apoptosis in the NRK-52E cells.Similar results were reported in mouse embryonic fibrobasts (Heet al., 2008). Several studies reported Nrf2 protection against thecytotoxicity of other agents that also produce ROS, such as chromium(He et al., 2007), arsenic (Wang et al., 2007), ultraviolet radiation(Hirota et al., 2005) and hydrogen peroxide (Li et al., 2005). Thesereports are also consistent with the temporal gene expression changesobserved during progression of pulmonary emphysema in A/J mice,where there is reduced expression of various cytoprotective genesconstituting ubiquitin-proteosome complex, cell survival pathways,solute carriers and transporters, transcription factors, and Nrf2-regulated antioxidant and phase II-responsive genes (Rangasamyet al., 2009).

To further explore the possible downstream genes through whichNrf2 might elicit protection against apoptosis, the expression of HO-1and GCLC, whose promoters contain an ARE, were determined in Cd-treated cells transfected with siRNA-Nrf2. HO-1 is a rate-limitingenzyme in the degradation of heme to produce equimolar amounts ofCO, iron, and biliverdin that is further converted by biliverdinreductase to the antioxidant, bilirubin (Maines, 1988). Similarly,GCL is the rate-limiting enzyme in the two-step biosynthesis ofglutathione. It is a heterodimeric enzyme composed of two subunits.The catalytic subunit, GCLC, possesses all of the catalytic functions ofthe enzyme (Griffith and Mulcahy, 1999). Consistent with the resultsreported in hepa1c1c7 cells (He et al., 2008), our data showed that Cdinduced a marked increase in HO-1 level and transfection with siRNA-Nrf2 markedly attenuated this increase. This suggests that HO-1 maybe partly responsible for the protective effect of Nrf2 against Cd-induced apoptosis in the NRK-52E cells. The results of our study are inaccord with a recent report showing that the genetic suppression ofHO-1 exacerbates gentamicin-induced apoptosis in NRK-52E cells(Sue et al., 2009). Our results showed that transfection with siRNA-Nrf2 reduced the GCLC expression upon treatment with Cd, suggest-ing that, besides HO-1, GCLC also contributes to the protective effect ofNrf2 against the Cd-induced apoptosis.

In summary, from the results obtained in this study, it is concludedthat (a) treatment of NRK-52E cells with Cd generated oxidative stressthat led to apoptosis at concentrations of Cd≥5 μM, (b) as a defensiveresponse against the oxidative stress, the level of the antioxidant

88 J. Chen, Z.A. Shaikh / Toxicology and Applied Pharmacology 241 (2009) 81–89

transcription factor Nrf2 was elevated and resulted in increasedexpression of antioxidant enzymes such as HO-1 and GCL, and (c) thisprovided a mechanism by which the cells could protect themselvesagainst the toxicity of 2.5 μM Cd.

Conflict of interest statementThere is no conflict of interest.

Acknowledgments

This research was made possible by the use of RI-INBRE CentralizedResearch Core Facility, supported by grant P20RR016457 from theNational Center for Research Resources, National Institutes of Health.The authors thank Nathan Nous for assistance with the FACS analysisand Joanna Fueyo for assistance in editing themanuscript. Jun Chenwasa visiting scholar fromHuazhong University of Science and Technology,Wuhan, China, and was supported through research fellowships fromthe China Scholarship Council and the RI-INBRE program.

References

Abu-Bakar, A., Satarug, S., Marks, G.C., Lang, M.A., Moore, M.R., 2004. Acute cadmiumchloride administration induces hepatic and renal CYP2A5 mRNA, protein andactivity in the mouse: involvement of transcription factor NRF2. Toxicol. Lett. 148,199–210.

ATSDR (Agency for Toxic Substances and Disease Registry), 1989. Toxicological profilefor cadmium. ATSDR/U.S. Public Health Service, ATSDR/TP-88/08.

Casalino, E., Sblano, C., Landriscina, C., 1997. Enzyme activity alteration by cadmiumadministration to rats: the possibility of iron involvement in lipid peroxidation.Arch. Biochem. Biophys. 346, 171–179.

Casalino, E., Calzaretti, G., Landriscina, M., Sblano, C., Fabiano, A., Landriscina, C., 2007.The Nrf2 transcription factor contributes to the induction of alpha-class GSTisoenzymes in liver of acute cadmium or manganese intoxicated rats: comparisonwith the toxic effect on NAD(P)H:quinone reductase. Toxicology 237, 24–34.

Chen, C., Jang, J., Li, M., Surh, Y., 2005. Resveratrol upregulates heme oxygenase-1expression via activation of NF-E2-related factor 2 in PC12 cells. Biochem. Biophys.Res. Commun. 331, 993–1000.

Dong, S., Shen, H.M., Ong, C.N., 2001. Cadmium-induced apoptosis and phenotypicchanges in mouse thymocytes. Mol. Cell. Biochem. 222, 11–20.

Elmore, S., 2007. Apoptosis: a review of programmed cell death. Toxicol. Pathol. 35,495–516.

England, K., Cotter, T.G., 2005. Direct oxidative modifications of signaling proteins inmammalian cells and their effects on apoptosis. Redox Rep. 10, 237–245.

Fan, Y., Wu, D., Jin, L., Yin, Z., 2005. Human glutamylcysteine synthetase protectsHEK293 cells against UV-induced cell death through inhibition of c-Jun NH2-terminal kinase. Cell Biol. Int. 29, 695–702.

Griffith, O.W., Mulcahy, R.T., 1999. The enzymes of glutathione synthesis: γ-glutamylcysteine synthetase. Adv. Enzymol. Relat. Areas Mol. Biol. 73, 209–267.

Gunawardana, C.G., Martinez, R.E., Xiao, W., Templeton, D.M., 2005. Cadmium inhibitsboth intrinsic and extrinsic apoptotic pathways in renal mesangial cells. Am. J.Physiol. Renal Physiol. 290, 1074–1082.

He, X., Lin, G., Chen, M., Zhang, J., Ma, Q., 2007. Protection against chromium (VI)-induced oxidative stress and apoptosis by Nrf2. Recruiting Nrf2 into the nucleusand disrupting the nuclear Nrf2/Keap1 association. Toxicol. Sci. 98, 298–309.

He, X., Chen, M.G., Ma, Q., 2008. Activation of Nrf2 in defense against cadmium-inducedoxidative stress. Chem. Res. Toxicol. 21, 1375–1383.

Hirota, A., Kawachi, Y., Itoh, K., Nakamura, Y., Xu, X., Banno, T., Takahashi, T., Yamamoto,M., Otsuka, F., 2005. Ultraviolet A irradiation induces NF-E2-related factor 2activation in dermal fibroblasts: protective role in UVA-induced apoptosis. J. Invest.Dermatol. 124, 825–832.

Jaiswal, A.K., 2004. Nrf2 signaling in coordinated activation of antioxidant geneexpression. Free Radic. Biol. Med. 36, 1199–1207.

Kensler, T.W.,Wakabayashi, N., Biswal, S., 2007. Cell survival response to environmentalstress via the Keap1-Nrf2-ARE pathway. Annu. Rev. Pharmacol. Toxicol. 47, 89–116.

Kimura, M., Yamamoto, T., Zhang, J., Itoh, K., Kyo, M., Kamiya, T., Aburatani, H., Katsuoka,F., Kurokawa, H., Tanaka, T., Motohashi, H., Yamamoto, M., 2007. Molecular basisdistinguishing the DNA binding profile of Nrf2-Maf heterodimer from that of Mafhomodimer. J. Biol. Chem. 282, 33681–33690.

Klaassen, C.D., Liu, J., 1997. Role of metallothionein in cadmium-induced hepatotoxicityand nephrotoxicity. Drug Metab. Rev. 29, 79–102.

Kobayashi, A., Kang, M.I., Watai, Y., Tong, K.I., Shibata, T., Uchida, K., Yamamoto, M.,2006. Oxidative and electrophilic stresses activate Nrf2 through inhibition ofubiquitination activity of Keap1. Mol. Cell. Biol. 26, 221–229.

Korashy, H.M., El-Kadi, A.O., 2006. Transcriptional regulation of the NAD(P)H:quinoneoxidoreductase 1 and glutathione S-transferase ya genes by mercury, lead, andcopper. Drug Metab. Dispos. 34, 152–165.

Kotlo, K.U., Yehiely, F., Efimova, E., Harasty, H., Hesabi, B., Shchors, K., Einat, P., Rozen, A.,Berent, E., Deiss, L.P., 2003. Nrf2 is an inhibitor of the Fas pathway as identified byAchilles' Heel Method, a new function-based approach to gene identification inhuman cells. Oncogene 22, 797–806.

LeBel, C.P., Ischiropoulos, H., Bondy, S.C., 1992. Evaluation of the probe 29,79-dichlorofluorescein as an indicator of reactive oxygen species formation andoxidative stress. Chem. Res. Toxicol. 5, 227–231.

Lee, J., Johnson, J.A., 2004. An important role of Nrf2-ARE pathway in the cellulardefense mechanism. J. Biochem. Mol. Biol. 37, 139–143.

Lee, J.M., Li, J., Johnson, D.A., Stein, T.D., Kraft, A.D., Calkins, M.J., Jakel, R.J., Johnson, J.A.,2005. Nrf2, a multi-organ protector? FASEB J. 19, 1061–1066.

Lee, W.K., Torchalski, B., Thevenod, F., 2007. Cadmium-induced ceramide formationtriggers calpain-dependent apoptosis in cultured kidney proximal tubule cells. Am.J. Physiol. Cell. Physiol. 293, 839–847.

Li, J., Johnson, D., Calkins, M., Wrigh, L., Svendsen, C., Johnson, J., 2005. Stabilization ofNrf2 by tBHQ confers protection against oxidative stress-induced cell death inhuman neural stem cells. Toxicol. Sci. 83, 313–328.

Liu, J., Qu, W., Kadiiska, M.B., 2009. Role of oxidative stress in cadmium toxicity andcarcinogenesis. Toxicol. Appl. Pharmacol. PMID: 19236887.

Liu, Y., Templeton, D.M., 2008. Initiation of caspase-independent death in mousemesangial cells by Cd2+: involvement of p38 kinase and CaMK-II. J. Cell Physiol.217, 307–318.

Liu, Y., Zhang, S.P., Cai, Y.Q., 2007. Cytoprotective effects of selenium on cadmium-induced LLC-PK1 cells apoptosis by activating JNK pathway. Toxicol. In Vitro 21,677–684.

Maines, M.D., 1988. Heme oxygenase: function, multiplicity, regulatory mechanisms,and clinical applications. FASEB J. 2, 2557–2568.

Miyaguchi, C., Muranaka, S., Kanno, T., Fujita, H., Akiyama, J., Yoshioka, T., Yasuda, T.,2004. 17beta-estradiol suppresses ROS-induced apoptosis of CHO cells throughinhibition of lipid peroxidation-coupled membrane permeability transition.Physiol. Chem. Phys. Med. NMR 36, 21–35.

Motohashi, H., Yamamoto, M., 2004. Nrf2-Keap1 defines a physiologically importantstress response mechanism. Trends Mol. Med. 11, 549–557.

Nguyen, T., Yang, C.S., Pickett, C.B., 2004. The pathways and molecular mechanismsregulating Nrf2 activation in response to chemical stress. Free Radic. Biol. Med. 37,433–441.

Oh, S.H., Lim, S.C., 2006. A rapid and transient ROS generation by cadmium triggersapoptosis via caspase-dependent pathway in HepG2 cells and this is inhibitedthrough N-acetylcysteine-mediated catalase upregulation. Toxicol. Appl. Pharma-col. 212, 212–223.

Okouchi, M., Okayama, N., Alexander, J.S., Aw, T.Y., 2006. NRF2-dependent glutamate-L-cysteine ligase catalytic subunit expression mediates insulin protection againsthyperglycemia-induced brain endothelial cell apoptosis. Curr. Neurovasc. Res. 3,249–261.

Olszanecki, R., Rezzani, R., Omura, S., Stec, D.E., Rodella, L., Botros, F.T., Goodman, A.I.,Drummond, G., Abraham, N.G., 2007. Genetic suppression of HO-1 exacerbatesrenal damage: reversed by an increase in the antiapoptotic signaling pathway. Am.J. Physiol. Renal Physiol. 292, F148–F157.

Pathak, N., Khandelwal, S., 2008. Impact of cadmium in T lymphocyte subsets andcytokine expression: differential regulation by oxidative stress and apoptosis.Biometals 21, 179–187.

Rangasamy, T., Misra, V., Zhen, L., Tankersley, C.G., Tuder, R.M., Biswal, S., 2009.Cigarette smoke-induced emphysema in A/J mice is associated with pulmonaryoxidative stress, apoptosis of lung cells, and global alterations in gene expression.Am. J. Physiol. Lung. Cell. Mol. Physiol. 296, L888–900.

Sakurai, A., Nishimoto, M., Himeno, S., Imura, N., Tsujimoto, M., Kunimoto, M., Hara, S.,2005. Transcriptional regulation of thioredoxin reductase 1 expression by cadmiumin vascular endothelial cells: role of NF-E2-related factor-2. J. Cell Physiol. 203,529–537.

Sen, G.R., Sen, G.E., Dhakal, B.K., Thakur, A.R., Ahnn, J., 2004. Vitamin C and vitamin Eprotect the rat testes from cadmium-induced reactive oxygen species. Mol. Cells 17,132–139.

Shaikh, Z.A., Vu, T.T., Zaman, K., 1999. Oxidative stress as a mechanism of chroniccadmium-induced hepatotoxicity and renal toxicity and protection by antiox-idants. Toxicol. Appl. Pharmacol. 154, 256–263.

Simon, H.U., Haj-Yehia, A., Levi-Schaffer, F., 2000. Role of reactive oxygen species (ROS)in apoptosis induction. Apoptosis 5, 415–418.

Stewart, D., Killeen, E., Naquin, R., Alam, S., Alam, J., 2003. Degradation of transcriptionfactor Nrf2 via the ubiquitin-proteasome pathway and stabilization by cadmium.J. Biol. Chem. 278, 2396–2402.

Sue, Y.M., Cheng, C.F., Chang, C.C., Chou, Y., Chen, C.H., Juan, S.H., 2009. Antioxidationand anti-inflammation by haem oxygenase-1 contribute to protection by tetra-methylpyrazine against gentamicin-induced apoptosis in murine renal tubularcells. Nephrol. Dial. Transplant. 24, 769–777.

Szuster-Ciesielska, A., Stachura, A., Słotwinska, M., Kaminska, T., Sniezko, R., Paduch, R.,Abramczyk, D., Filar, J., Kandefer-Szerszen, M., 2000. The inhibitory effect of zinc oncadmium-induced cell apoptosis and reactive oxygen species (ROS) production incell cultures. Toxicology 145, 159–171.

Takebayashi, S., Jimi, S., Segawa, M., Kiyoshi, Y., 2000. Cadmium induces osteomalaciamediated by proximal tubular atrophy and disturbances of phosphate reabsorp-tion. A study of 11 autopsies. Pathol. Res. Pract. 196, 653–663.

Tan, K.P., Yang, M., Ito, S., 2007. Activation of nuclear factor (erythroid-2 like) factor 2 bytoxic bile acids provokes adaptive defense responses to enhance cell survival at theemergence of oxidative stress. Mol. Pharmacol. 72, 1380–1390.

Vestergaard, P., Shaikh, Z.A., 1994. The nephrotoxicity of intravenously administeredcadmium-metallothionein: effect of dose, mode of administration, and preexistingrenal cadmium burden. Toxicol. Appl. Pharmacol. 126, 240–247.

Valko, M., Rhodes, C.J., Moncol, J., Izakovic, M., Mazur, M., 2006. Free radicals, metalsand antioxidants in oxidative stress-induced cancer. Chem. Biol. Interact. 160,1–40.

89J. Chen, Z.A. Shaikh / Toxicology and Applied Pharmacology 241 (2009) 81–89

Wang, X.J., Sun, Z., Chen, W., Eblin, K.E., Gandolfi, J.A., Zhang, D.D., 2007. Nrf2 protectshuman bladder urothelial cells from arsenite and monomethylarsonous acidtoxicity. Toxicol. Appl. Pharmacol. 225, 206–213.

Wätjen, W., Beyersmann, D., 2004. Cadmium-induced apoptosis in C6 glioma cells:influence of oxidative stress. Biometals 17, 65–78.

Wätjen, W., Haase, H., Biagioli, M., Beyersmann, D., 2002. Induction of apoptosis inmammalian cells by cadmium and zinc. Environ. Health Perspect. 110, 865–867.

Xie, J., Shaikh, Z.A., 2006. Cadmium-induced apoptosis in rat kidney epithelial cellsinvolves decrease in nuclear factor-kappa B activity. Toxicol. Sci. 91, 299–308.

Yang, H., Magilnick, N., Lee, C., Kalmaz, D., Ou, X., Chan, J., Lu, S., 2005. Nrf1 and Nrf2

regulate rat glutamate-cysteine ligase catalytic subunit transcription indirectly viaNF-kappaB and AP-1. Mol. Cell Biol. 25, 5933–5946.

Yuan, C., Kadiiska, M., Achanzar, W.E., Mason, R.P., Waalkes, M.P., 2000. Possible role ofcaspase-3 inhibition in cadmium-induced blockage of apoptosis. Toxicol. Appl.Pharmacol. 164, 321–329.

Yu, X., Hong, S., Faustman, E.M., 2008. Cadmium-induced activation of stress signalingpathways, disruption of ubiquitin-dependent protein degradation and apoptosis inprimary rat Sertoli cell-gonocyte cocultures. Toxicol. Sci. 104, 385–396.

Zhang, D.D., 2006. Mechanistic studies of the Nrf2-Keap1 signaling pathway. DrugMetab. Rev. 38, 769–789.