Nrf2 Activation Involves an Oxidative-Stress Independent Pathway in Tetrafluoroethylcysteine-Induced Cytotoxicity
http://www.100md.com
《毒物学科学杂志》
Departments of Medicinal Chemistry, Environmental and Occupational Health Sciences
Pathology, University of Washington, Seattle, Washington 98195
ABSTRACT
Tetrafluoroethylcysteine (TFEC), a metabolite of the industrial gas tetrafluoroethylene, can cause both nephrotoxicity and limited hepatotoxicity in animal models, and this is associated with the covalent modification of specific intramitochondrial proteins including heat shock protein 60 (HSP60), mitochondrial HSP70 (mtHSP70), aspartate aminotransferase (AST), aconitase, and -ketoglutarate dehydrogenase (KGDH). Using the murine TAMH cell line as a useful in vitro model for TFEC toxicity, we demonstrate a rapid and sustained induction of Nrf2, a member of the "cap-and-collar" transcription factor family, following exposure to cytotoxic concentrations of TFEC. A functional correlate was also established with the rapid translocation of cytosolic Nrf2 into the nucleus. In addition, transcriptional and translational upregulation of known Nrf2 regulated genes including glutamate cysteine ligase (GCL), both catalytic and modulatory subunits, heme oxygenase-1, and glutathione S-transferase (GST) isoforms were detected. While Nrf2 activation is often linked to perturbation of cellular thiol status and/or oxidative stress, we were unable to detect any significant depletion of cellular glutathione or oxidation of mitochondrial membrane cardiolipin or increases in reactive oxygen species (ROS). These data suggest Nrf2 activation is likely independent of classical oxidative stress or, at best, a result of a transient, low-level redox stress. Moreover, supporting evidence indicates an early endoplasmic reticular (ER) stress response after TFEC treatment, with a time-dependent upregulation of the ER responsive genes gadd34, gadd45, gadd153, and ndr1 . These findings suggest an alternative pathway for Nrf2 activation, i.e., Nrf2 phosphorylation through ER-mediated protein kinases such as PKR-like endoplasmic reticular kinase (PERK). Overall, the results implicate a role for Nrf2 in the cellular response to TFEC toxicity and suggest a previously unrecognized role for the ER in this model of mitochondrially initiated cytotoxicity.
Key Words: tetrafluoroethylcysteine; mitochondrial dysfunction; Nrf2; oxidative stress; ER stress.
INTRODUCTION
S-(1,1,2,2-tetrafluoroethyl)-L-cysteine (TFEC) is a metabolite of the industrial gas, tetrafluoroethylene, which is used commercially as a precursor in TeflonTM production. Severe nephrotoxicity and slight hepatotoxicity have been observed previously in experimental animals as well as cell culture systems, signifying a potential risk to humans through environmental exposure (Cooper et al., 2002; Lock and Ishmael, 1998). The current consensus on the mechanism of toxicity involves bioactivation of TFEC by cysteine S-conjugate -lyase, to an unstable difluorothioacetyl fluoride (DFTAL) intermediate (Cooper et al., 2002; James et al., 2002). This reactive electrophilic species covalently modifies a very small, but well-defined group of intramitochondrial proteins including aconitase, -ketoglutarate dehydrogenase (KGDH) subunits, aspartate aminotransferase (AST), heat shock protein 60 (HSP60), and HSP70 (Bruschi et al., 1993, 1998; James et al., 2002). It is generally accepted that these binding events are the initiating lesions in the mitochondrial dysfunction, necrotic cell death, and tissue damage produced by TFEC (Ho et al., 2005).
Previous investigations have characterized a predominant mitochondrial role in progression of injury involving translocation of pro-apoptotic cytosolic BAX to the mitochondrial outer membrane as a relatively early event in TFEC toxicity (Ho et al., 2005). Subsequent membrane permeability transition (MPT) and cytochrome c release have also been shown to be reversible with either BCL-xL overexpression or bongkekric acid administration (Ho et al., 2005; James et al., 2002). Collectively, these studies indicate that TFEC is a unique intramitochondrial toxicant, which acts to disrupt bioenergetics at precisely defined sites. Nonetheless, clear evidence linking the binding of intramitochondrial proteins to the final outcome of TFEC-induced cytotoxicity is still lacking. We have previously reported extensive biochemical characterizations of TFEC-induced damage in vitro utilizing the differentiated TAMH mouse cell line (Ho et al., 2005; James et al., 2002; Pierce et al., 2002; Wu et al., 1994). For example, the covalent modification of aconitase and KGDH protein targets in vitro attenuates their enzymatic activities and adequately models the deficits observed in vivo (Ho et al., 2005; James et al., 2002; Pierce et al., 2002; Wu et al., 1994). The relationship(s) between reduced protein function and subsequent mitochondrial damage and cell death have yet to be elucidated. The signaling pathways to and from the mitochondria leading to progression of cellular injury also lacks adequate characterization. Therefore, the specific aim of this study was to use global gene expression-profiling to identify unknown cellular and biochemical responses to TFEC treatment in the TAMH cell line that may be involved in pathways progressing to cytotoxicity.
MATERIALS AND METHODS
Cell culture.
Serum-free cell culture of the TAMH line between passages 21 and 35 was undertaken as previously described (Wu et al., 1994). All chemicals were obtained from Sigma unless otherwise stated. Briefly, cells were grown in serum free Dulbecco's modified Eagle's medium/Ham's F12 (Gibco, Rockville, MD) supplemented with 5 μg/ml insulin, 5 μg/ml transferrin, 5 ng/ml selenium (Collaborative Biomedical Products, Bedford, MA), 100 nM dexamethasone, 10 mM nicotinamide, and 0.1% v/v gentamicin (Gibco). Cultures were maintained in a humidified incubator with 5% carbon dioxide/95% air at 37°C and passaged at 70–90% confluence. For glucose-free experiment, cells were grown and subsequently treated in Dulbecco's modified Eagle's medium, with or without glucose. Hepa-1 cells overexpressing glutamate cysteine ligase (GCL) were developed and maintained as previously described (Botta et al., 2004).
RNA isolation.
Cells were grown to confluence in 150 cm2 tissue culture dishes in quadruplicate for each sample and treated with 200 μM TFEC for 2, 4, and 6 h. At the end of respective treatments, cells were harvested by scraping with rubber policemen. The resultant cell pellets were spun down and washed once with Dulbecco's PBS (Gibco). Immediately, 1 ml of Trizol reagent (Gibco) per 107 cells was added for cell lysis. After vortexing, the lysate was passed through 22G needles 10 times to ensure complete lysis. Quickly, 0.2 ml of chloroform was added to every 1 ml of cell lysate and vortexed vigorously for 15 s (1-ml aliquots of lysates were measured into microcentrifuge tubes). The tubes were left to stand for 2–3 min before spinning at 10,000 rpm for 15 min at 4°C. Gently, 0.5 ml of aqueous phase was transferred to a fresh tube, and an equivolume of 70% ethanol was added. This resulting mix was loaded onto an RNeasy column (Qiagen, Valencia, CA), and purified total RNAs were extracted according to the manufacturer's protocol.
Microarray analysis procedures.
Gene expression analyses were performed using the Amersham, "Codelink" 10K mouse array (Amersham Biosciences, Piscataway, NJ) according to manufacturer's protocols. Briefly, total RNA from each sample was quantified before first and second strand cDNA synthesis. The resulting double-stranded cDNA was purified with a QIAquick spin column (Qiagen). After drying the cDNA in a SpeedVac concentrator, cRNA was synthesized by in vitro transcription and purified using the RNeasy kit. The quality of the cRNA was evaluated using an Agilent, 2100 Bioanalyzer (Agilent, Palo Alto, CA), and only those with A260:A280 ratio of 1.8–2.1 were used for subsequent microarray analysis. Each 10 μg of cRNA sample was hybridized onto Codelink microarray slides and incubated for 18 h at 37°C. At the end of incubation, the arrays were washed with 0.75x TNT buffer (0.1 M Tris–HCl pH 7.6, 0.15 M NaCl, 0.05% Tween-20) at 46°C for 1 h and incubated with streptavidin-Alexa 647 (Molecular Probes, Eugene, OR) working solution at 25°C for 30 min to label the fluorogenic probe. The arrays were scanned with an Axon GenePix 4000B fluorescent scanner and the GenePix Pro imaging software (Axon Instruments, Foster City, CA). Fluorescent intensity of each spot in the image was determined using ImaGeneTM 5 (Biodiscovery, Marina del Rey, CA) for spot finding and analysis.
Real time RT-PCR.
Fluorogenic 5' nuclease assays (TaqMan) were carried out using an ABI Prism 7700 Sequence detection system (Applied Biosystems, Foster City, CA). The thermal cycling condition comprised an initial denaturation step at 95°C for 10 min, followed by 40 cycles at 95°C for 20 s and 62°C for 60 s. The gene-specific sequences of the primer pairs and probes used in the assays are as follows: GCLc (U85498 [GenBank] ): forward primer, ATGTGGACACCCGATGCAGTATT; reverse primer, TGTCTTGCTTGTAGTCAGGATGGTTT; probe, CCTAAAGCTAATTAAGAAGAGAGC. GCLm (NM_008129 [GenBank] ): forward primer, GCCACCAGATTTGACTGCCTTT; reverse primer, CAGGGATGCTTTCTTGAAGAGCTT; probe, TCTGAGGCAAGTTTCCA. GSTA3 (NM_010356 [GenBank] ): forward primer, AGGAACAAACCAGGAACCGTTACTT; reverse primer, CAGCGCTCCTCAGCCTGTT; probe, TCTTCAACACCTTTTCAAAGG. GSTA2 (NM_008182 [GenBank] ): forward primer, GTATTATGTCCCCCAGACCAAAGAG; reverse primer, CTGTTGCCCACAAGGTAGTCTTGT. GAPDH (NM_008084 [GenBank] ): forward primer, TCCTGCACCACCAACTGCTT; reverse primer, GAGGGGCCATCCACAGTCTT; probe, CACTCATGACCACAGTCCATGCCATCAC. GSTA2 was analyzed by SYBR green instead of TaqMan, and no probe was needed.
Isolation of cytosolic/nuclear fractions.
Nuclear and cytosolic fractions were isolated with slight modifications to the protocol described previously (Buckley et al., 2003). Briefly, the harvested cells were pelleted and resuspended in 475 μl of cytosolic extraction buffer (10 mM Tris-base, 60 mM KCl, 1 mM EDTA, 1 mM DTT, protease inhibitor cocktail) and kept on ice for 10 min. Subsequently, 25 μl of 10% v/v Igepal CA-630 was added to the cell suspension and mixed with gentle pipetting for 10–15 s. This mixture was spun at 12,000 x g for 5 min at 4°C. The resultant supernatant was removed as the cytosolic fraction. The pellet, which contained the nuclear-enriched fraction, was then resuspended again in nuclear extraction buffer (20 mM Tris-base, 400 mM NaCl, 1.5 mM MgCl2, 1.5 mM EDTA, 1 mM DTT, 25% v/v glycerol, protease inhibitor cocktail). Independent verification of the relative purity of subcellular fractions was by immunoblot (as described below).
Immunoblot procedures.
All the fractions collected were assayed for protein concentration using the BCA protein assay kit (Pierce Chemical Co., Rockford, IL). Each 30–50 μg of sample proteins were resolved by denaturing electrophoresis, SDS–PAGE (Mini-PROTEAN II; Bio-Rad Laboratories, Hercules, CA) and transferred to nitrocellulose membrane for 1 h at 15 V using Trans-Blot SD Semi-Dry Transfer Cell (Bio-Rad). Immunodetection was by chemiluminescence (SuperSignal ULTRA; Pierce, Rockford, IL) using specific antibodies diluted in PBS with 0.05% v/v Tween 20 and 5% w/v powdered milk. Anti-Nrf2 (1:1,000), anti-Nrf1, anti-Gadd153, anti-Gadd34, and anti-Histone-H1 were from Santa Cruz Biotechnology (San Diego CA), anti-HO-1 from Stressgen (Victoria, BC, Canada), and anti-GAPDH was developed in-house (Dietze et al., 1997). Secondary anti-mouse and anti-rabbit horseradish peroxidase conjugated secondary antibodies (Pierce) were used at 1:20,000 dilution. All primary antibodies were used at 1:2,000 dilution unless otherwise stated. Densitometric analyses were performed on selected immunoblots using Bio-Rad ChemiDoc and the Quantity One Version 4.3.0 program (Bio-Rad).
Immunocytochemistry.
Cells were grown on 4-well chambered slides (Labtek II, Nalgen, Naperville, IL). Cultures were dosed with 200 μM TFEC for 0–4 h. After treatment, medium was aspirated, and cells were washed twice with Hank's BSS and fixed with 3.7% (v/v) paraformaldehyde (EMS, Ft. Washington, PA) in Hank's BSS for 20 min at room temperature. Ice-cold acetone was added for 5 min, and nonspecific binding was blocked by soaking the chambers overnight in PBS with 10% FBS at 4°C. Immunostaining was with anti-Nrf2 and anti-rabbit fluorescein isothiocyanate (FITC) conjugated IgG (Molecular Probes) in the presence of saponin (0.2% w/v) to enhance antibody accessibility. Nuclear staining was performed by incubating 4',6-diamidine-2'-phenylindole dihydrochloride (DAPI) at 0.5 μg/ml in PBS for 5 min. Cells were washed extensively with PBS before being mounted with Fluoromount G (Southern Biotechnologies, Birmingham, AL) and examined using Nikon Eclipse fluorescence microscope (Nikon, Melville, NY) with 40x lenses. Images were then processed with Q-Imaging software (Burnaby, BC, Canada).
Spectrofluorometric analyses of hydrogen peroxide formation and free cytosolic calcium.
Cells were grown on 6-well dishes and treated with TFEC as indicated. Cells were then incubated with either 10 μM dihydro-dichlorofluorescein-diacetate (H2DCFDA) for measuring intracellular hydrogen peroxide, or 2 μM Indo-1 AM for measuring intracellular calcium. Cells were rinsed, harvested, and resuspended in 1 ml Hanks' BSS. Fluorescence detection was by SLM-Aminco 8100 spectrofluorometer (Spectronic Instruments, Rochester, NY), monitoring excitation 468 nm, emission 528 nm for H2DCFDA. Indo-1 detection and calcium quantification required a ratiometric analysis of [(excitation 320 nm, emission 405 nm) / (excitation 355 nm, emission 475 nm)].
Viability assay by MTT.
Cell viability was determined by 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl-tetrazolium bromide (MTT) viability assay according to protocol as previously described (Ho et al., 2005). Experiments investigating the impact of antioxidants on TFEC toxicity were with Trolox (1 mM) and t-butylhydroperoxide as a positive control.
ATP depletion assay.
Intracellular ATP level was measured by its activity using the CellTiter-Glo Luminescent Cell Viability Assay (Promega, Madison, WI) according to the manufacturer's protocol. Briefly, cells were treated with 250 μM TFEC for 0–8 h on 96-well plates. After incubation, an equivolume of the luminescent substrate and lysis buffer mix from the assay kit was added. The mixture was transferred to an opaque 96-well plate, and luminescence was read and analyzed with PlateLumino (Phenix, Hayward CA).
Flow cytometry.
Flow cytometry methods were adopted as previously described (Botta et al., 2004). Briefly, cells were grown on 12-well dishes and treated for the time periods indicated. All cells were harvested, resuspended, and incubated in nonylacridine orange (NAO, Molecular Probes; 2 μM; excitation 488 nm, emission 530 nm), as a measure of mitochondrial cardiolipin content (Molecular Probes), with monochlorobimane (4 μM; excitation 351–362 nm, emission 450 nm) for reduced glutatione (GSH) content, or with hydroethidine for superoxide (Molecular Probes; 5 μM; excitation 488 nm, emission 590 nm). Diethylmaleate (DEM) was used as a positive control for GSH depletion (125 μM; 4 h). After staining for 30 min at 37°C in the dark, the cells were examined by flow cytometry (Epics Elite, Beckman-Coulter Corp, Miami, FL) for the intensity of green fluorescence. The PMTs were gated for live cells using propidium iodide (2 μM), added just before acquisition. NAD(P)H redox status was also monitored concurrently with a UV-excited blue autofluorescence (excitation 351–362 nm, emission 450 nm). Data from at least 5,000 cells were collected in list mode and processed with MPlus Software (Phoenix Flow Systems, San Diego, CA).
RESULTS
Induction of Oxidative Stress-Related Genes
Microarray analyses were performed on TAMH cells treated with 200 μM TFEC for 2, 4, and 6 h using a 10K mouse oligonucleotide array (Codelink). A comprehensive analysis of genomic responses is currently underway, and a separate manuscript is in preparation. Here, we highlight a group of oxidative stress–related genes that were highly upregulated in a time-dependent fashion, with the expression level increasing from 2 to 6 h treatment with TFEC. These genes included heme oxygenase-1 (HO-1), GST-alpha isoforms, GCL subunits, glutathione reductase, and thioredoxin reductase (Table 1). Heme oxygenase-1 was the most upregulated gene in the entire array as measured at 4 h of TFEC treatment (29.4-fold higher than control cells). The mRNA expression of GCL subunits GSTA2 and GSTA3 was validated by quantitative RT-PCR (Table 2). These confirmatory data demonstrate a good correlation with the microarray analyses. For example, GSTA2, the most highly upregulated of the four genes validated (40.8-fold induction at 6 h with microarray), was also the most highly expressed by RT-PCR (111.5-fold induction at 6 h).
Nrf2 Induction and Translocation Is an Early Event in TFEC Toxicity
Many of the oxidative stress–related genes identified in these studies share an antioxidant response element (ARE) in their promoter sequences. Consequently, Nrf2 expression was assessed using Western blot analyses following treatment with 200 μM of TFEC for 0–6 h. An almost negligible level of Nrf2 protein was observed in control cells, but the level increased markedly after only 2 h of TFEC treatment (Fig. 1A). Higher doses of TFEC (250 μM) produced similar induction at 6 h, but with a lower absolute amount of Nrf2 induced (Fig. 1A). Furthermore, Western blot analyses have demonstrated a dramatic protein induction at 6 h treatment, which is completely reversed by the specific protein synthesis inhibitor, cycloheximide (Fig. 1B). Similar cycloheximide-dependent inhibition of expression was also observed with HO-1 (Fig. 1C).
Determination of Nrf2 translocation from the cytosol to the nucleus was used as a functional correlate for Nrf2 induction. As shown in Figure 2, cytosolic Nrf2 levels increased in a time-dependent manner from 0 to 2 h and, thereafter, decreased. Meanwhile, nuclear Nrf2 reached a maximum at about 1 h and remained at that level. This translocation was also specific for Nrf2, since the highly homologous Nrf1 did not display any change in expression level or subcellular distribution following TFEC treatment (Fig. 2). Likewise, complementary immunocytochemical staining of Nrf2 and fluorescence microscopy demonstrated the translocation with a distinct nuclear staining pattern after similar treatment (Fig. 3). Thus, these data clearly indicate that Nrf2 is rapidly mobilized from the cytosol to the nucleus in response to TFEC treatment, consistent with the induction of ARE-controlled genes.
Lack of an Early Phase Oxidative Stress in TFEC Toxicity
Nrf2 activation and the upregulation of effector ARE genes suggest that TFEC might act by an oxidative stress mechanism. GSH depletion, oxidation of membrane cardiolipin, levels of reduced pyridine nucleotides (NAD(P)H), and intracellular hydrogen peroxide/superoxide formation were examined as indicators of cellular redox perturbations. Using flow cytometric measurements with a thiol-reactive dye, monochlorobimane, no significant change was observed in cellular concentrations of GSH from 0–6 h treatment with 250 μM TFEC (Fig. 4A). A supratoxicological dose of TFEC (400 μM) caused a decrease (approx. 20%) in reduced GSH that was not as pronounced as the decrease (approx. 80%) caused by diethylmaleate (Fig. 4A). Likewise, there was little change in membrane cardiolipin oxidation as measured by NAO staining and flow cytometry, nor any significant increase in intracellular hydrogen peroxide or superoxide formation within the first 2 h (Figs. 4B, 4C, 4D). NAD(P)H levels, as detected by UV-excited autofluorescence, were significantly reduced at 4 and 6 h (Fig. 4E). The reduction is time dependent, such that cells after 6 h of treatment had approximately 50% lower levels of reduced pyridine dinucleotides in comparison to controls. Arguably, it is still possible that oxidative stress was transient and/or below our detection limits. Hence, the significance of any low-level oxidative stress was investigated by looking at the cytoprotective action of an antioxidant, as well as the transgenic overexpression of GCL subunits. An effective dose of Trolox (1 mM) failed to reverse the toxicity of TFEC (250 μM) as demonstrated with the MTT viability assay (Fig. 4F). Likewise, Hepa-1 cells overexpressing GCL subunits (i.e., CR17) did not alter the dose-dependent profile of TFEC toxicity (Fig. 4G).
Cellular Calcium Dysregulation
Endoplasmic reticular (ER) stress is known to result in nonphysiological mobilization of calcium from this compartment. Therefore, intracellular calcium was measured spectrofluorometrically using the ratiometric UV-excitable dye, Indo-1 AM, which fluoresces strongly at 405 nm with high cytosolic calcium concentrations, and at 475 nm with low concentrations. Ratiometric analysis also corrects for differential dye-loading between samples. The results show an increasing intracellular calcium concentration with time following TFEC treatment (Fig. 5A). Significant increases were observed after 4 and 6 h with ratios of 0.57 and 0.72, respectively. These values were comparable to those found in cells treated with 100 nM calcium ionophore, A23187 [GenBank] , (0.85) as a positive control indicative of calcium dysregulation from more generalized ER stress following TFEC exposure.
Depletion of Intracellular ATP
Changes to cellular ATP levels are expected as a consequence of TFEC-mediated inhibition of the TCA cycle and were investigated using a commercially available luminescent procedure (CellTiter-Glo Luminescent Cell Viability Assay, Promega). There was a rapid depletion of intracellular ATP within the first 2 h of TFEC treatment (to approx. 50% of starting levels). By 8 h, the levels were reduced even further to less than 20% of control levels (Fig. 5A).
To further determine the importance of intracellular ATP to subsequent cell signaling, TAMH cells were preincubated in glucose-free or high-glucose (4,500 mg/l) medium before subjecting to cytotoxic concentrations of TFEC (250 μM) for an additional 24 h. Our results show that cells survived significantly better in the presence of high-glucose medium (Fig. 5B), and this was associated with the maintenance of cellular ATP content (approx. 4-fold higher) by glucose supplementation (Fig. 5C).
Induction of ER Stress Response Genes
Activation of Nrf2 in the absence of oxidative stress suggests involvement of another signaling pathway known to occur at the level of the ER (Cullinan et al., 2003; Liu et al., 2005). Our microarray analyses have indicated a strong link to ER stress with an early and pronounced upregulation of a number of ER-stress response genes (Table 3). This list includes four ER resident proteins (Gadd153, Gadd45, Gadd34, Atf3) as well as cytosolic proteins that have been well-established to be induced in response to ER stress (i.e., Ndr1). At least one of these genes was shown here to be upregulated at the protein level (Gadd153), while Gadd34 did not show any increase in protein level despite significant transcriptional changes (Fig. 6 and Table 3). Western blot analyses for both Hsp70i and Atf3 showed that Atf3 is strongly upregulated temporally from 4 to 8 h after the initiation of TFEC treatment, whereas Hsp70i levels remained high even at 20 h (Fig. 7).
DISCUSSION
The work presented here has shown that, despite a highly focused origin of subcellular damage, signal transduction of the initial chemical insult caused by TFEC rapidly expands to affect other organelles such as the ER. We previously demonstrated that the initial damage to mitochondrial proteins results in the translocation of cytosolic BAX to the outer mitochondrial membrane (Ho et al., 2005). This translocation is pivotal to the activation of subsequent mitochondrial effects, including a mitochondrial permeability transition (MPT) and the release of cytochrome c into the cytosol where it mediates downstream proteolysis (Ho et al., 2005; James et al., 2002). Gene expression profiling has now revealed a previously undetected early induction of ARE-responsive genes including heme oxygenase-1, GCL, thioredoxin reductase, and GSTs during TFEC cytotoxicity. Subsequent RT-PCR has confirmed this high level of HO-1 induction (data not shown) as well as the induction of GCL subunits and GST-alpha isoforms. Further independent microarray studies using the same samples but run on a different platform (NIA 15K mouse cDNA array) have also shown that many of the same genes were upregulated as well as other oxidative stress related genes, including various ferritin subunits and GST-pi (Hu et al., manuscript in preparation). This was a surprising observation, since previous reports had indicated that the level of oxidative stress in TFEC-mediated cell death was absent or minimal (Groves et al., 1991). To establish a link from the initial mitochondrial protein binding events to the upregulation of these genes, we considered the possible role of a predominant and well-characterized transcription factor, Nrf2.
Nrf2 is an important stress-responsive transcription factor of the "cap-and-collar" -leucine zipper family, especially during oxidative stress (Nguyen et al., 2003). It regulates the expression of a number of Phase II enzymes (e.g., NQO1, GSTs) and antioxidant proteins (e.g., GCL, HO-1, thioredoxin). This process is driven by the association of Nrf2 to the ARE consensus sequence (5'-TGACnnnGCA-3') on the promoter region of these genes (Itoh et al., 2004; Jaiswal, 2004; Lee and Johnson, 2004; Numazawa and Yoshida, 2004). Recent reports continue to identify new downstream effector genes for Nrf2, including thioredoxin reductase and MafG (Katsuoka et al., 2005; Sakurai et al., 2005).
At present, the mechanisms of Nrf2 upstream activation are not completely understood. A few independent, yet not mutually exclusive, theories have emerged in this respect. Nrf2 has been shown to be constitutively expressed and localized in the cytosol and maintained in a repressed state by complexing with the actin-associating protein, Keap1. This heterodimerization confines most of Nrf2 to the cytoskeleton and away from the nucleus. Keap1 has a cysteine-rich surface which is subject to oxidation in cases of oxidative and nitrosative stress. This apparently results in global conformational changes to Keap1 thereby, leading to the liberation of Nrf2. The monomeric Nrf2 is then available to translocate to the nucleus. In this manner, Keap1 acts as a redox-sensor that upregulates ARE antioxidant responses through Nrf2 (Itoh et al., 2003, 2004; Kang et al., 2004; Levonen et al., 2004; Zhang and Hannink, 2003).
Nrf2 activation also has been shown to be mediated through phosphorylation of Nrf2 by mitogen-activated protein kinases (MAPKs), protein kinase C (atypical isoform), and phosphoinositol-3-kinase (PI3K) (Nakaso et al., 2003; Nguyen et al., 2003; Numazawa et al., 2003; Yu et al., 2000). Further upstream kinases may also play a role in these events. In addition, it has been proposed that Nrf2 can be activated through a redox-independent pathway. This involves a prior ER stress that induces an ER-specific protein kinase, termed PKR-like endoplasmic reticular kinase (PERK), which can directly phosphorylate Nrf2 (Cullinan and Diehl, 2004; Cullinan et al., 2003).
From the studies reported here, TFEC can be seen to induce some Phase II enzymes and antioxidant-responsive genes in TAMH cells, and this is likely a consequence of early Nrf2 induction. This phenomenon has not been previously reported for TFEC or other halogenated aliphatics. To establish mechanistic links to the activation of Nrf2, we examined numerous indicators of cellular oxidative stress, but the overall lack of any significant changes in these parameters suggests that classical oxidative stress does not contribute significantly to TFEC-mediated cytotoxicity in TAMH cells. These findings are consistent with previously reported in vivo findings (Groves et al., 1991). While this does not rule out a role for Keap1 in the regulation of Nrf2 translocation, it does imply that redox changes related to GSH status or reactive oxygen species (ROS) production are not pivotal in the activation pathway. In future studies it will be of interest to examine the possibility of Keap1 alterations following transient, low-level oxidative stress using highly sensitive mass spectrometric techniques.
One of the earliest effects of TFEC observed was a rapid depletion of intracellular ATP. TFEC-mediated modification and inhibition of mitochondrial aconitase and KGDH activities (both important enzymes in the TCA cycle) might result in a localized intramitochondrial oxidative stress leading to the rapid inhibition of ATP production—an important determinant of the commitment to necrosis or apoptosis (Leist et al., 1997, 1999). Intracellular ATP concentrations are critical for cell viability, and marked ATP depletion (15–25% of control) has been hypothesized to switch the cell death mechanism from apoptosis to necrosis (Lieberthal et al., 1998). In fact, our previous studies with TFEC have shown that, despite mitochondrial changes supporting apoptosis (e.g., cytochrome c release), activation of pro-apoptotic caspases does not occur. Rather, energy-independent cysteine proteases like calpains replace caspases as the major enzymes catalyzing proteolysis (Ho et al., 2005). The present work provides direct evidence for an immediate loss of ATP that is consistent with the decay of early apoptotic signals into a secondary necrosis. More importantly, we have also demonstrated that replenishment of ATP by glucose supplementation significantly restored cell viability. Even though depletion of ATP in glucose supplemented cells was still significant, it appears that a higher basal level was sufficient to "cushion" some of the damage produced by TFEC. These data confirm the importance of low intracellular ATP levels on critical downstream energy-dependent processes.
The lack of clear evidence for a generalized cellular oxidative stress, coupled to significant ATP depletion, implicates an alternative pathway for the activation of Nrf2 in this cell death model. One possibility is an ER stress–mediated Nrf2 induction. An ER resident protein kinase, PERK, is known to directly phosphorylate Nrf2 and trigger dissociation from Keap1 without the involvement of ROS (Cullinan et al., 2003). PERK itself is strongly activated as part of the unfolded protein response (UPR) and has been shown to be critical for cell survival during ER stress (Cullinan and Diehl, 2004). A recent study has also reported that ER stress–stimulated HO-1 induction occurs through Nrf2 binding to ARE, consistent with our findings (Liu et al., 2005). Furthermore, regulation of calcium homeostasis appears to be important during Nrf2 activation (Lee et al., 2003), and recently, a selective calmodulin/CaMK inhibitor, KN93, was shown to block Nrf2 activation of ARE gene induction in HepG2 cells treated with diallyl trisulfide (Chen et al., 2004).
The role of ER stress in TFEC-induced cytotoxicity is not well studied, but there are scattered reports which suggest that this might be a significant and under-appreciated phenomenon. For example, a role for calpains in renal cell death following TFEC treatment has been demonstrated (Schnellmann and Williams, 1998), in agreement with our previous work using TAMH cells (Ho et al., 2005). Of particular significance, calpain activation was blocked by overexpression of anti-apoptotic BCL-xL (Ho et al., 2005). This study also provides additional evidence for ER stress coupled with induction of specific ER stress proteins in TFEC-induced cytotoxicity. Particularly, Atf3 is recognized as an important stress-responsive ER-bound transcription factor that has been shown to induce Gadd153 (Wolfgang et al., 1997). Since covalent binding precedes all other events, these findings suggest that ER stress occurs as a result of some prior mitochondrial dysfunction. Because a strong link has already been made between ATP depletion and calcium release from the ER (Harriman et al., 2002), we have hypothesized that interruption of mitochondrial function causes a rapid depletion of intracellular ATP by TFEC, which leads to ER calcium release and an unfolded protein response (UPR), which is an energy-dependent process (Fig. 8). It is also possible that ATP depletion directly inhibits Ca-ATPase such that calcium is released from ER stores. Our ongoing efforts will include detecting activation of PERK, as well as other calcium-dependent signaling pathways, in Nrf2 phosphorylation and activation.
Overall, to the best of our knowledge, this is the first evidence of Nrf2 activation in TFEC-induced cytotoxicity. The apparent lack of oxidative stress despite clear transcriptional upregulation of classical antioxidant response genes suggests that a pathway independent of classical oxidative stress might be involved in this mitochondrially-initiated pathway of toxicity. A strong ER response following ATP deprivation provides important clues to support further investigations of ER stress–mediated Nrf2 activation.
ACKNOWLEDGMENTS
This work was supported by NIH grants GM51916 (SAB), GM32165 (SDN), CA74131 (NF), UW NIEHS sponsored Center for Ecogenetics and Environmental Health: NIEHS P30ES07033, 5 R01 ES010849 (TJK) and Pfizer Inc. Authors wish to thank the technical help of Yankai Jia, Fred Farin, Theo Bammler, Dick Beyer, and Jasmine Wilkerson for microarray and RT-PCR support, and Keck Center for fluorescence microscopy. Conflict of interest: none declared.
REFERENCES
Botta, D., Franklin, C. C., White, C. C., Krejsa, C. M., Dabrowski, M. J., Pierce, R. H., Fausto, N., and Kavanagh, T. J. (2004). Glutamate-cysteine ligase attenuates TNF-induced mitochondrial injury and apoptosis. Free Radic. Biol. Med. 37, 632–642.
Bruschi, S. A., Lindsay, J. G., and Crabb, J. W. (1998). Mitochondrial stress protein recognition of inactivated dehydrogenases during mammalian cell death. Proc. Natl. Acad. Sci. U.S.A. 95, 13413–13418.
Bruschi, S. A., West, K. A., Crabb, J. W., Gupta, R. S., and Stevens, J. L. (1993). Mitochondrial HSP60 (P1 protein) and a HSP70-like protein (mortalin) are major targets for modification during S-(1,1,2,2-tetrafluoroethyl)-L-cysteine-induced nephrotoxicity. J. Biol. Chem. 268, 23157–23161.
Buckley, B. J., Marshall, Z. M., and Whorton, A. R. (2003). Nitric oxide stimulates Nrf2 nuclear translocation in vascular endothelium. Biochem. Biophys. Res. Commun. 307, 973–939.
Chen, C., Pung, D., Leong, V., Hebbar, V., Shen, G., Nair, S., Li, W., and Tony Kong, A. N. (2004). Induction of detoxifying enzymes by garlic organosulfur compounds through transcription factor Nrf2: Effect of chemical structure and stress signals. Free Radic. Biol. Med. 37, 1578–1590.
Cooper, A. J., Bruschi, S. A., and Anders, M. W. (2002). Toxic, halogenated cysteine S-conjugates and targeting of mitochondrial enzymes of energy metabolism. Biochem. Pharmacol. 64, 553–564.
Cullinan, S. B., and Diehl, J. A. (2004). PERK-dependent activation of Nrf2 contributes to redox homeostasis and cell survival following endoplasmic reticulum stress. J. Biol. Chem. 279, 20108–20117.
Cullinan, S. B., Zhang, D., Hannink, M., Arvisais, E., Kaufman, R. J., and Diehl, J. A. (2003). Nrf2 is a direct PERK substrate and effector of PERK-dependent cell survival. Mol. Cell. Biol. 23, 7198–7209.
Dietze, E. C., Schafer, A., Omichinski, J. G., and Nelson, S. D. (1997). Inactivation of glyceraldehyde-3-phosphate dehydrogenase by a reactive metabolite of acetaminophen and mass spectral characterization of an arylated active site peptide. Chem. Res. Toxicol. 10, 1097–1103.
Groves, C. E., Lock, E. A., and Schnellmann, R. G. (1991). Role of lipid peroxidation in renal proximal tubule cell death induced by haloalkene cysteine conjugates. Toxicol. Appl. Pharmacol. 107, 54–62.
Harriman, J. F., Liu, X. L., Aleo, M. D., Machaca, K., and Schnellmann, R. G. (2002). Endoplasmic reticulum Ca(2+) signaling and calpains mediate renal cell death. Cell Death Differ. 9, 734–741.
Ho, H. K., Hu, Z. H., Tzung, S. P., Hockenbery, D. M., Fausto, N., Nelson, S. D., and Bruschi, S. A. (2005). BCL-xL overexpression effectively protects against tetrafluoroethylcysteine-induced intramitochondrial damage and cell death. Biochem. Pharmacol. 69, 147–157.
Itoh, K., Tong, K. I., and Yamamoto, M. (2004). Molecular mechanism activating Nrf2-Keap1 pathway in regulation of adaptive response to electrophiles. Free Radic. Biol. Med. 36, 1208–1213.
Itoh, K., Wakabayashi, N., Katoh, Y., Ishii, T., O'Connor, T., and Yamamoto, M. (2003). Keap1 regulates both cytoplasmic-nuclear shuttling and degradation of Nrf2 in response to electrophiles. Genes Cells 8, 379–391.
Jaiswal, A. K. (2004). Nrf2 signaling in coordinated activation of antioxidant gene expression. Free Radic. Biol. Med. 36, 1199–1207.
James, E. A., Gygi, S. P., Adams, M. L., Pierce, R. H., Fausto, N., Aebersold, R. H., Nelson, S. D., and Bruschi, S. A. (2002). Mitochondrial aconitase modification, functional inhibition, and evidence for a supramolecular complex of the TCA cycle by the renal toxicant S-(1,1,2,2-tetrafluoroethyl)-L-cysteine. Biochemistry 41, 6789–6797.
Kang, M. I., Kobayashi, A., Wakabayashi, N., Kim, S. G., and Yamamoto, M. (2004). Scaffolding of Keap1 to the actin cytoskeleton controls the function of Nrf2 as key regulator of cytoprotective phase 2 genes. Proc. Natl. Acad. Sci. U.S.A. 101, 2046–2051.
Katsuoka, F., Motohashi, H., Engel, J. D., and Yamamoto, M. (2005). Nrf2 transcriptionally activates the mafG gene through an antioxidant response element. J. Biol. Chem. 280, 4483–4490.
Lee, J. M., and Johnson, J. A. (2004). An important role of Nrf2-ARE pathway in the cellular defense mechanism. J. Biochem. Mol. Biol. 37, 139–143.
Lee, J. M., Shih, A. Y., Murphy, T. H., and Johnson, J. A. (2003). NF-E2-related factor-2 mediates neuroprotection against mitochondrial complex I inhibitors and increased concentrations of intracellular calcium in primary cortical neurons. J. Biol. Chem. 278, 37948–37956.
Leist, M., Single, B., Castoldi, A. F., Kuhnle, S., and Nicotera, P. (1997). Intracellular adenosine triphosphate (ATP) concentration: A switch in the decision between apoptosis and necrosis. J. Exp. Med. 185, 1481–1486.
Leist, M., Single, B., Naumann, H., Fava, E., Simon, B., Kuhnle, S., and Nicotera, P. (1999). Inhibition of mitochondrial ATP generation by nitric oxide switches apoptosis to necrosis. Exp. Cell Res. 249, 396–403.
Levonen, A. L., Landar, A., Ramachandran, A., Ceaser, E. K., Dickinson, D. A., Zanoni, G., Morrow, J. D., and Darley-Usmar, V. M. (2004). Cellular mechanisms of redox cell signalling: Role of cysteine modification in controlling antioxidant defences in response to electrophilic lipid oxidation products. Biochem. J. 378, 373–382.
Lieberthal, W., Menza, S. A., and Levine, J. S. (1998). Graded ATP depletion can cause necrosis or apoptosis of cultured mouse proximal tubular cells. Am. J. Physiol. 274, F315–F327.
Liu, X. M., Peyton, K. J., Ensenat, D., Wang, H., Schafer, A. I., Alam, J., and Durante, W. (2005). Endoplasmic reticulum stress stimulates heme oxygenase-1 gene expression in vascular smooth muscle. Role in cell survival. J. Biol. Chem. 280, 872–877.
Lock, E. A., and Ishmael, J. (1998). The nephrotoxicity and hepatotoxicity of 1,1,2,2-tetrafluoroethyl-L-cysteine in the rat. Arch. Toxicol. 72, 347–354.
Nakaso, K., Yano, H., Fukuhara, Y., Takeshima, T., Wada-Isoe, K., and Nakashima, K. (2003). PI3K is a key molecule in the Nrf2-mediated regulation of antioxidative proteins by hemin in human neuroblastoma cells. FEBS Lett. 546, 181–184.
Nguyen, T., Sherratt, P. J., and Pickett, C. B. (2003). Regulatory mechanisms controlling gene expression mediated by the antioxidant response element. Annu. Rev. Pharmacol. Toxicol. 43, 233–260.
Numazawa, S., Ishikawa, M., Yoshida, A., Tanaka, S., and Yoshida, T. (2003). Atypical protein kinase C mediates activation of NF-E2-related factor 2 in response to oxidative stress. Am. J. Physiol. Cell Physiol. 285, C334–C342.
Numazawa, S., and Yoshida, T. (2004). Nrf2-dependent gene expressions: A molecular toxicological aspect. J. Toxicol. Sci. 29, 81–89.
Pierce, R. H., Franklin, C. C., Campbell, J. S., Tonge, R. P., Chen, W., Fausto, N., Nelson, S. D., and Bruschi, S. A. (2002). Cell culture model for acetaminophen-induced hepatocyte death in vivo. Biochem. Pharmacol. 64, 413–424.
Sakurai, A., Nishimoto, M., Himeno, S., Imura, N., Tsujimoto, M., Kunimoto, M., and Hara, S. (2005). Transcriptional regulation of thioredoxin reductase 1 expression by cadmium in vascular endothelial cells: Role of NF-E2-related factor-2. J. Cell Physiol. 203, 529–537.
Schnellmann, R. G., and Williams, S. W. (1998). Proteases in renal cell death: Calpains mediate cell death produced by diverse toxicants. Ren. Fail. 20, 679–686.
Wolfgang, C. D., Chen, B. P., Martindale, J. L., Holbrook, N. J., and Hai, T. (1997). gadd153/Chop10, a potential target gene of the transcriptional repressor ATF3. Mol. Cell. Biol. 17, 6700–6707.
Wu, J. C., Merlino, G., Cveklova, K., Mosinger, B., Jr., and Fausto, N. (1994). Autonomous growth in serum-free medium and production of hepatocellular carcinomas by differentiated hepatocyte lines that overexpress transforming growth factor alpha 1. Cancer Res. 54, 5964–5973.
Yu, R., Chen, C., Mo, Y. Y., Hebbar, V., Owuor, E. D., Tan, T. H., and Kong, A. N. (2000). Activation of mitogen-activated protein kinase pathways induces antioxidant response element-mediated gene expression via a Nrf2-dependent mechanism. J. Biol. Chem. 275, 39907–39913.
Zhang, D. D., and Hannink, M. (2003). Distinct cysteine residues in Keap1 are required for Keap1-dependent ubiquitination of Nrf2 and for stabilization of Nrf2 by chemopreventive agents and oxidative stress. Mol. Cell. Biol. 23, 8137–8151.(Han K. Ho, Collin C. Whit)
Pathology, University of Washington, Seattle, Washington 98195
ABSTRACT
Tetrafluoroethylcysteine (TFEC), a metabolite of the industrial gas tetrafluoroethylene, can cause both nephrotoxicity and limited hepatotoxicity in animal models, and this is associated with the covalent modification of specific intramitochondrial proteins including heat shock protein 60 (HSP60), mitochondrial HSP70 (mtHSP70), aspartate aminotransferase (AST), aconitase, and -ketoglutarate dehydrogenase (KGDH). Using the murine TAMH cell line as a useful in vitro model for TFEC toxicity, we demonstrate a rapid and sustained induction of Nrf2, a member of the "cap-and-collar" transcription factor family, following exposure to cytotoxic concentrations of TFEC. A functional correlate was also established with the rapid translocation of cytosolic Nrf2 into the nucleus. In addition, transcriptional and translational upregulation of known Nrf2 regulated genes including glutamate cysteine ligase (GCL), both catalytic and modulatory subunits, heme oxygenase-1, and glutathione S-transferase (GST) isoforms were detected. While Nrf2 activation is often linked to perturbation of cellular thiol status and/or oxidative stress, we were unable to detect any significant depletion of cellular glutathione or oxidation of mitochondrial membrane cardiolipin or increases in reactive oxygen species (ROS). These data suggest Nrf2 activation is likely independent of classical oxidative stress or, at best, a result of a transient, low-level redox stress. Moreover, supporting evidence indicates an early endoplasmic reticular (ER) stress response after TFEC treatment, with a time-dependent upregulation of the ER responsive genes gadd34, gadd45, gadd153, and ndr1 . These findings suggest an alternative pathway for Nrf2 activation, i.e., Nrf2 phosphorylation through ER-mediated protein kinases such as PKR-like endoplasmic reticular kinase (PERK). Overall, the results implicate a role for Nrf2 in the cellular response to TFEC toxicity and suggest a previously unrecognized role for the ER in this model of mitochondrially initiated cytotoxicity.
Key Words: tetrafluoroethylcysteine; mitochondrial dysfunction; Nrf2; oxidative stress; ER stress.
INTRODUCTION
S-(1,1,2,2-tetrafluoroethyl)-L-cysteine (TFEC) is a metabolite of the industrial gas, tetrafluoroethylene, which is used commercially as a precursor in TeflonTM production. Severe nephrotoxicity and slight hepatotoxicity have been observed previously in experimental animals as well as cell culture systems, signifying a potential risk to humans through environmental exposure (Cooper et al., 2002; Lock and Ishmael, 1998). The current consensus on the mechanism of toxicity involves bioactivation of TFEC by cysteine S-conjugate -lyase, to an unstable difluorothioacetyl fluoride (DFTAL) intermediate (Cooper et al., 2002; James et al., 2002). This reactive electrophilic species covalently modifies a very small, but well-defined group of intramitochondrial proteins including aconitase, -ketoglutarate dehydrogenase (KGDH) subunits, aspartate aminotransferase (AST), heat shock protein 60 (HSP60), and HSP70 (Bruschi et al., 1993, 1998; James et al., 2002). It is generally accepted that these binding events are the initiating lesions in the mitochondrial dysfunction, necrotic cell death, and tissue damage produced by TFEC (Ho et al., 2005).
Previous investigations have characterized a predominant mitochondrial role in progression of injury involving translocation of pro-apoptotic cytosolic BAX to the mitochondrial outer membrane as a relatively early event in TFEC toxicity (Ho et al., 2005). Subsequent membrane permeability transition (MPT) and cytochrome c release have also been shown to be reversible with either BCL-xL overexpression or bongkekric acid administration (Ho et al., 2005; James et al., 2002). Collectively, these studies indicate that TFEC is a unique intramitochondrial toxicant, which acts to disrupt bioenergetics at precisely defined sites. Nonetheless, clear evidence linking the binding of intramitochondrial proteins to the final outcome of TFEC-induced cytotoxicity is still lacking. We have previously reported extensive biochemical characterizations of TFEC-induced damage in vitro utilizing the differentiated TAMH mouse cell line (Ho et al., 2005; James et al., 2002; Pierce et al., 2002; Wu et al., 1994). For example, the covalent modification of aconitase and KGDH protein targets in vitro attenuates their enzymatic activities and adequately models the deficits observed in vivo (Ho et al., 2005; James et al., 2002; Pierce et al., 2002; Wu et al., 1994). The relationship(s) between reduced protein function and subsequent mitochondrial damage and cell death have yet to be elucidated. The signaling pathways to and from the mitochondria leading to progression of cellular injury also lacks adequate characterization. Therefore, the specific aim of this study was to use global gene expression-profiling to identify unknown cellular and biochemical responses to TFEC treatment in the TAMH cell line that may be involved in pathways progressing to cytotoxicity.
MATERIALS AND METHODS
Cell culture.
Serum-free cell culture of the TAMH line between passages 21 and 35 was undertaken as previously described (Wu et al., 1994). All chemicals were obtained from Sigma unless otherwise stated. Briefly, cells were grown in serum free Dulbecco's modified Eagle's medium/Ham's F12 (Gibco, Rockville, MD) supplemented with 5 μg/ml insulin, 5 μg/ml transferrin, 5 ng/ml selenium (Collaborative Biomedical Products, Bedford, MA), 100 nM dexamethasone, 10 mM nicotinamide, and 0.1% v/v gentamicin (Gibco). Cultures were maintained in a humidified incubator with 5% carbon dioxide/95% air at 37°C and passaged at 70–90% confluence. For glucose-free experiment, cells were grown and subsequently treated in Dulbecco's modified Eagle's medium, with or without glucose. Hepa-1 cells overexpressing glutamate cysteine ligase (GCL) were developed and maintained as previously described (Botta et al., 2004).
RNA isolation.
Cells were grown to confluence in 150 cm2 tissue culture dishes in quadruplicate for each sample and treated with 200 μM TFEC for 2, 4, and 6 h. At the end of respective treatments, cells were harvested by scraping with rubber policemen. The resultant cell pellets were spun down and washed once with Dulbecco's PBS (Gibco). Immediately, 1 ml of Trizol reagent (Gibco) per 107 cells was added for cell lysis. After vortexing, the lysate was passed through 22G needles 10 times to ensure complete lysis. Quickly, 0.2 ml of chloroform was added to every 1 ml of cell lysate and vortexed vigorously for 15 s (1-ml aliquots of lysates were measured into microcentrifuge tubes). The tubes were left to stand for 2–3 min before spinning at 10,000 rpm for 15 min at 4°C. Gently, 0.5 ml of aqueous phase was transferred to a fresh tube, and an equivolume of 70% ethanol was added. This resulting mix was loaded onto an RNeasy column (Qiagen, Valencia, CA), and purified total RNAs were extracted according to the manufacturer's protocol.
Microarray analysis procedures.
Gene expression analyses were performed using the Amersham, "Codelink" 10K mouse array (Amersham Biosciences, Piscataway, NJ) according to manufacturer's protocols. Briefly, total RNA from each sample was quantified before first and second strand cDNA synthesis. The resulting double-stranded cDNA was purified with a QIAquick spin column (Qiagen). After drying the cDNA in a SpeedVac concentrator, cRNA was synthesized by in vitro transcription and purified using the RNeasy kit. The quality of the cRNA was evaluated using an Agilent, 2100 Bioanalyzer (Agilent, Palo Alto, CA), and only those with A260:A280 ratio of 1.8–2.1 were used for subsequent microarray analysis. Each 10 μg of cRNA sample was hybridized onto Codelink microarray slides and incubated for 18 h at 37°C. At the end of incubation, the arrays were washed with 0.75x TNT buffer (0.1 M Tris–HCl pH 7.6, 0.15 M NaCl, 0.05% Tween-20) at 46°C for 1 h and incubated with streptavidin-Alexa 647 (Molecular Probes, Eugene, OR) working solution at 25°C for 30 min to label the fluorogenic probe. The arrays were scanned with an Axon GenePix 4000B fluorescent scanner and the GenePix Pro imaging software (Axon Instruments, Foster City, CA). Fluorescent intensity of each spot in the image was determined using ImaGeneTM 5 (Biodiscovery, Marina del Rey, CA) for spot finding and analysis.
Real time RT-PCR.
Fluorogenic 5' nuclease assays (TaqMan) were carried out using an ABI Prism 7700 Sequence detection system (Applied Biosystems, Foster City, CA). The thermal cycling condition comprised an initial denaturation step at 95°C for 10 min, followed by 40 cycles at 95°C for 20 s and 62°C for 60 s. The gene-specific sequences of the primer pairs and probes used in the assays are as follows: GCLc (U85498 [GenBank] ): forward primer, ATGTGGACACCCGATGCAGTATT; reverse primer, TGTCTTGCTTGTAGTCAGGATGGTTT; probe, CCTAAAGCTAATTAAGAAGAGAGC. GCLm (NM_008129 [GenBank] ): forward primer, GCCACCAGATTTGACTGCCTTT; reverse primer, CAGGGATGCTTTCTTGAAGAGCTT; probe, TCTGAGGCAAGTTTCCA. GSTA3 (NM_010356 [GenBank] ): forward primer, AGGAACAAACCAGGAACCGTTACTT; reverse primer, CAGCGCTCCTCAGCCTGTT; probe, TCTTCAACACCTTTTCAAAGG. GSTA2 (NM_008182 [GenBank] ): forward primer, GTATTATGTCCCCCAGACCAAAGAG; reverse primer, CTGTTGCCCACAAGGTAGTCTTGT. GAPDH (NM_008084 [GenBank] ): forward primer, TCCTGCACCACCAACTGCTT; reverse primer, GAGGGGCCATCCACAGTCTT; probe, CACTCATGACCACAGTCCATGCCATCAC. GSTA2 was analyzed by SYBR green instead of TaqMan, and no probe was needed.
Isolation of cytosolic/nuclear fractions.
Nuclear and cytosolic fractions were isolated with slight modifications to the protocol described previously (Buckley et al., 2003). Briefly, the harvested cells were pelleted and resuspended in 475 μl of cytosolic extraction buffer (10 mM Tris-base, 60 mM KCl, 1 mM EDTA, 1 mM DTT, protease inhibitor cocktail) and kept on ice for 10 min. Subsequently, 25 μl of 10% v/v Igepal CA-630 was added to the cell suspension and mixed with gentle pipetting for 10–15 s. This mixture was spun at 12,000 x g for 5 min at 4°C. The resultant supernatant was removed as the cytosolic fraction. The pellet, which contained the nuclear-enriched fraction, was then resuspended again in nuclear extraction buffer (20 mM Tris-base, 400 mM NaCl, 1.5 mM MgCl2, 1.5 mM EDTA, 1 mM DTT, 25% v/v glycerol, protease inhibitor cocktail). Independent verification of the relative purity of subcellular fractions was by immunoblot (as described below).
Immunoblot procedures.
All the fractions collected were assayed for protein concentration using the BCA protein assay kit (Pierce Chemical Co., Rockford, IL). Each 30–50 μg of sample proteins were resolved by denaturing electrophoresis, SDS–PAGE (Mini-PROTEAN II; Bio-Rad Laboratories, Hercules, CA) and transferred to nitrocellulose membrane for 1 h at 15 V using Trans-Blot SD Semi-Dry Transfer Cell (Bio-Rad). Immunodetection was by chemiluminescence (SuperSignal ULTRA; Pierce, Rockford, IL) using specific antibodies diluted in PBS with 0.05% v/v Tween 20 and 5% w/v powdered milk. Anti-Nrf2 (1:1,000), anti-Nrf1, anti-Gadd153, anti-Gadd34, and anti-Histone-H1 were from Santa Cruz Biotechnology (San Diego CA), anti-HO-1 from Stressgen (Victoria, BC, Canada), and anti-GAPDH was developed in-house (Dietze et al., 1997). Secondary anti-mouse and anti-rabbit horseradish peroxidase conjugated secondary antibodies (Pierce) were used at 1:20,000 dilution. All primary antibodies were used at 1:2,000 dilution unless otherwise stated. Densitometric analyses were performed on selected immunoblots using Bio-Rad ChemiDoc and the Quantity One Version 4.3.0 program (Bio-Rad).
Immunocytochemistry.
Cells were grown on 4-well chambered slides (Labtek II, Nalgen, Naperville, IL). Cultures were dosed with 200 μM TFEC for 0–4 h. After treatment, medium was aspirated, and cells were washed twice with Hank's BSS and fixed with 3.7% (v/v) paraformaldehyde (EMS, Ft. Washington, PA) in Hank's BSS for 20 min at room temperature. Ice-cold acetone was added for 5 min, and nonspecific binding was blocked by soaking the chambers overnight in PBS with 10% FBS at 4°C. Immunostaining was with anti-Nrf2 and anti-rabbit fluorescein isothiocyanate (FITC) conjugated IgG (Molecular Probes) in the presence of saponin (0.2% w/v) to enhance antibody accessibility. Nuclear staining was performed by incubating 4',6-diamidine-2'-phenylindole dihydrochloride (DAPI) at 0.5 μg/ml in PBS for 5 min. Cells were washed extensively with PBS before being mounted with Fluoromount G (Southern Biotechnologies, Birmingham, AL) and examined using Nikon Eclipse fluorescence microscope (Nikon, Melville, NY) with 40x lenses. Images were then processed with Q-Imaging software (Burnaby, BC, Canada).
Spectrofluorometric analyses of hydrogen peroxide formation and free cytosolic calcium.
Cells were grown on 6-well dishes and treated with TFEC as indicated. Cells were then incubated with either 10 μM dihydro-dichlorofluorescein-diacetate (H2DCFDA) for measuring intracellular hydrogen peroxide, or 2 μM Indo-1 AM for measuring intracellular calcium. Cells were rinsed, harvested, and resuspended in 1 ml Hanks' BSS. Fluorescence detection was by SLM-Aminco 8100 spectrofluorometer (Spectronic Instruments, Rochester, NY), monitoring excitation 468 nm, emission 528 nm for H2DCFDA. Indo-1 detection and calcium quantification required a ratiometric analysis of [(excitation 320 nm, emission 405 nm) / (excitation 355 nm, emission 475 nm)].
Viability assay by MTT.
Cell viability was determined by 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl-tetrazolium bromide (MTT) viability assay according to protocol as previously described (Ho et al., 2005). Experiments investigating the impact of antioxidants on TFEC toxicity were with Trolox (1 mM) and t-butylhydroperoxide as a positive control.
ATP depletion assay.
Intracellular ATP level was measured by its activity using the CellTiter-Glo Luminescent Cell Viability Assay (Promega, Madison, WI) according to the manufacturer's protocol. Briefly, cells were treated with 250 μM TFEC for 0–8 h on 96-well plates. After incubation, an equivolume of the luminescent substrate and lysis buffer mix from the assay kit was added. The mixture was transferred to an opaque 96-well plate, and luminescence was read and analyzed with PlateLumino (Phenix, Hayward CA).
Flow cytometry.
Flow cytometry methods were adopted as previously described (Botta et al., 2004). Briefly, cells were grown on 12-well dishes and treated for the time periods indicated. All cells were harvested, resuspended, and incubated in nonylacridine orange (NAO, Molecular Probes; 2 μM; excitation 488 nm, emission 530 nm), as a measure of mitochondrial cardiolipin content (Molecular Probes), with monochlorobimane (4 μM; excitation 351–362 nm, emission 450 nm) for reduced glutatione (GSH) content, or with hydroethidine for superoxide (Molecular Probes; 5 μM; excitation 488 nm, emission 590 nm). Diethylmaleate (DEM) was used as a positive control for GSH depletion (125 μM; 4 h). After staining for 30 min at 37°C in the dark, the cells were examined by flow cytometry (Epics Elite, Beckman-Coulter Corp, Miami, FL) for the intensity of green fluorescence. The PMTs were gated for live cells using propidium iodide (2 μM), added just before acquisition. NAD(P)H redox status was also monitored concurrently with a UV-excited blue autofluorescence (excitation 351–362 nm, emission 450 nm). Data from at least 5,000 cells were collected in list mode and processed with MPlus Software (Phoenix Flow Systems, San Diego, CA).
RESULTS
Induction of Oxidative Stress-Related Genes
Microarray analyses were performed on TAMH cells treated with 200 μM TFEC for 2, 4, and 6 h using a 10K mouse oligonucleotide array (Codelink). A comprehensive analysis of genomic responses is currently underway, and a separate manuscript is in preparation. Here, we highlight a group of oxidative stress–related genes that were highly upregulated in a time-dependent fashion, with the expression level increasing from 2 to 6 h treatment with TFEC. These genes included heme oxygenase-1 (HO-1), GST-alpha isoforms, GCL subunits, glutathione reductase, and thioredoxin reductase (Table 1). Heme oxygenase-1 was the most upregulated gene in the entire array as measured at 4 h of TFEC treatment (29.4-fold higher than control cells). The mRNA expression of GCL subunits GSTA2 and GSTA3 was validated by quantitative RT-PCR (Table 2). These confirmatory data demonstrate a good correlation with the microarray analyses. For example, GSTA2, the most highly upregulated of the four genes validated (40.8-fold induction at 6 h with microarray), was also the most highly expressed by RT-PCR (111.5-fold induction at 6 h).
Nrf2 Induction and Translocation Is an Early Event in TFEC Toxicity
Many of the oxidative stress–related genes identified in these studies share an antioxidant response element (ARE) in their promoter sequences. Consequently, Nrf2 expression was assessed using Western blot analyses following treatment with 200 μM of TFEC for 0–6 h. An almost negligible level of Nrf2 protein was observed in control cells, but the level increased markedly after only 2 h of TFEC treatment (Fig. 1A). Higher doses of TFEC (250 μM) produced similar induction at 6 h, but with a lower absolute amount of Nrf2 induced (Fig. 1A). Furthermore, Western blot analyses have demonstrated a dramatic protein induction at 6 h treatment, which is completely reversed by the specific protein synthesis inhibitor, cycloheximide (Fig. 1B). Similar cycloheximide-dependent inhibition of expression was also observed with HO-1 (Fig. 1C).
Determination of Nrf2 translocation from the cytosol to the nucleus was used as a functional correlate for Nrf2 induction. As shown in Figure 2, cytosolic Nrf2 levels increased in a time-dependent manner from 0 to 2 h and, thereafter, decreased. Meanwhile, nuclear Nrf2 reached a maximum at about 1 h and remained at that level. This translocation was also specific for Nrf2, since the highly homologous Nrf1 did not display any change in expression level or subcellular distribution following TFEC treatment (Fig. 2). Likewise, complementary immunocytochemical staining of Nrf2 and fluorescence microscopy demonstrated the translocation with a distinct nuclear staining pattern after similar treatment (Fig. 3). Thus, these data clearly indicate that Nrf2 is rapidly mobilized from the cytosol to the nucleus in response to TFEC treatment, consistent with the induction of ARE-controlled genes.
Lack of an Early Phase Oxidative Stress in TFEC Toxicity
Nrf2 activation and the upregulation of effector ARE genes suggest that TFEC might act by an oxidative stress mechanism. GSH depletion, oxidation of membrane cardiolipin, levels of reduced pyridine nucleotides (NAD(P)H), and intracellular hydrogen peroxide/superoxide formation were examined as indicators of cellular redox perturbations. Using flow cytometric measurements with a thiol-reactive dye, monochlorobimane, no significant change was observed in cellular concentrations of GSH from 0–6 h treatment with 250 μM TFEC (Fig. 4A). A supratoxicological dose of TFEC (400 μM) caused a decrease (approx. 20%) in reduced GSH that was not as pronounced as the decrease (approx. 80%) caused by diethylmaleate (Fig. 4A). Likewise, there was little change in membrane cardiolipin oxidation as measured by NAO staining and flow cytometry, nor any significant increase in intracellular hydrogen peroxide or superoxide formation within the first 2 h (Figs. 4B, 4C, 4D). NAD(P)H levels, as detected by UV-excited autofluorescence, were significantly reduced at 4 and 6 h (Fig. 4E). The reduction is time dependent, such that cells after 6 h of treatment had approximately 50% lower levels of reduced pyridine dinucleotides in comparison to controls. Arguably, it is still possible that oxidative stress was transient and/or below our detection limits. Hence, the significance of any low-level oxidative stress was investigated by looking at the cytoprotective action of an antioxidant, as well as the transgenic overexpression of GCL subunits. An effective dose of Trolox (1 mM) failed to reverse the toxicity of TFEC (250 μM) as demonstrated with the MTT viability assay (Fig. 4F). Likewise, Hepa-1 cells overexpressing GCL subunits (i.e., CR17) did not alter the dose-dependent profile of TFEC toxicity (Fig. 4G).
Cellular Calcium Dysregulation
Endoplasmic reticular (ER) stress is known to result in nonphysiological mobilization of calcium from this compartment. Therefore, intracellular calcium was measured spectrofluorometrically using the ratiometric UV-excitable dye, Indo-1 AM, which fluoresces strongly at 405 nm with high cytosolic calcium concentrations, and at 475 nm with low concentrations. Ratiometric analysis also corrects for differential dye-loading between samples. The results show an increasing intracellular calcium concentration with time following TFEC treatment (Fig. 5A). Significant increases were observed after 4 and 6 h with ratios of 0.57 and 0.72, respectively. These values were comparable to those found in cells treated with 100 nM calcium ionophore, A23187 [GenBank] , (0.85) as a positive control indicative of calcium dysregulation from more generalized ER stress following TFEC exposure.
Depletion of Intracellular ATP
Changes to cellular ATP levels are expected as a consequence of TFEC-mediated inhibition of the TCA cycle and were investigated using a commercially available luminescent procedure (CellTiter-Glo Luminescent Cell Viability Assay, Promega). There was a rapid depletion of intracellular ATP within the first 2 h of TFEC treatment (to approx. 50% of starting levels). By 8 h, the levels were reduced even further to less than 20% of control levels (Fig. 5A).
To further determine the importance of intracellular ATP to subsequent cell signaling, TAMH cells were preincubated in glucose-free or high-glucose (4,500 mg/l) medium before subjecting to cytotoxic concentrations of TFEC (250 μM) for an additional 24 h. Our results show that cells survived significantly better in the presence of high-glucose medium (Fig. 5B), and this was associated with the maintenance of cellular ATP content (approx. 4-fold higher) by glucose supplementation (Fig. 5C).
Induction of ER Stress Response Genes
Activation of Nrf2 in the absence of oxidative stress suggests involvement of another signaling pathway known to occur at the level of the ER (Cullinan et al., 2003; Liu et al., 2005). Our microarray analyses have indicated a strong link to ER stress with an early and pronounced upregulation of a number of ER-stress response genes (Table 3). This list includes four ER resident proteins (Gadd153, Gadd45, Gadd34, Atf3) as well as cytosolic proteins that have been well-established to be induced in response to ER stress (i.e., Ndr1). At least one of these genes was shown here to be upregulated at the protein level (Gadd153), while Gadd34 did not show any increase in protein level despite significant transcriptional changes (Fig. 6 and Table 3). Western blot analyses for both Hsp70i and Atf3 showed that Atf3 is strongly upregulated temporally from 4 to 8 h after the initiation of TFEC treatment, whereas Hsp70i levels remained high even at 20 h (Fig. 7).
DISCUSSION
The work presented here has shown that, despite a highly focused origin of subcellular damage, signal transduction of the initial chemical insult caused by TFEC rapidly expands to affect other organelles such as the ER. We previously demonstrated that the initial damage to mitochondrial proteins results in the translocation of cytosolic BAX to the outer mitochondrial membrane (Ho et al., 2005). This translocation is pivotal to the activation of subsequent mitochondrial effects, including a mitochondrial permeability transition (MPT) and the release of cytochrome c into the cytosol where it mediates downstream proteolysis (Ho et al., 2005; James et al., 2002). Gene expression profiling has now revealed a previously undetected early induction of ARE-responsive genes including heme oxygenase-1, GCL, thioredoxin reductase, and GSTs during TFEC cytotoxicity. Subsequent RT-PCR has confirmed this high level of HO-1 induction (data not shown) as well as the induction of GCL subunits and GST-alpha isoforms. Further independent microarray studies using the same samples but run on a different platform (NIA 15K mouse cDNA array) have also shown that many of the same genes were upregulated as well as other oxidative stress related genes, including various ferritin subunits and GST-pi (Hu et al., manuscript in preparation). This was a surprising observation, since previous reports had indicated that the level of oxidative stress in TFEC-mediated cell death was absent or minimal (Groves et al., 1991). To establish a link from the initial mitochondrial protein binding events to the upregulation of these genes, we considered the possible role of a predominant and well-characterized transcription factor, Nrf2.
Nrf2 is an important stress-responsive transcription factor of the "cap-and-collar" -leucine zipper family, especially during oxidative stress (Nguyen et al., 2003). It regulates the expression of a number of Phase II enzymes (e.g., NQO1, GSTs) and antioxidant proteins (e.g., GCL, HO-1, thioredoxin). This process is driven by the association of Nrf2 to the ARE consensus sequence (5'-TGACnnnGCA-3') on the promoter region of these genes (Itoh et al., 2004; Jaiswal, 2004; Lee and Johnson, 2004; Numazawa and Yoshida, 2004). Recent reports continue to identify new downstream effector genes for Nrf2, including thioredoxin reductase and MafG (Katsuoka et al., 2005; Sakurai et al., 2005).
At present, the mechanisms of Nrf2 upstream activation are not completely understood. A few independent, yet not mutually exclusive, theories have emerged in this respect. Nrf2 has been shown to be constitutively expressed and localized in the cytosol and maintained in a repressed state by complexing with the actin-associating protein, Keap1. This heterodimerization confines most of Nrf2 to the cytoskeleton and away from the nucleus. Keap1 has a cysteine-rich surface which is subject to oxidation in cases of oxidative and nitrosative stress. This apparently results in global conformational changes to Keap1 thereby, leading to the liberation of Nrf2. The monomeric Nrf2 is then available to translocate to the nucleus. In this manner, Keap1 acts as a redox-sensor that upregulates ARE antioxidant responses through Nrf2 (Itoh et al., 2003, 2004; Kang et al., 2004; Levonen et al., 2004; Zhang and Hannink, 2003).
Nrf2 activation also has been shown to be mediated through phosphorylation of Nrf2 by mitogen-activated protein kinases (MAPKs), protein kinase C (atypical isoform), and phosphoinositol-3-kinase (PI3K) (Nakaso et al., 2003; Nguyen et al., 2003; Numazawa et al., 2003; Yu et al., 2000). Further upstream kinases may also play a role in these events. In addition, it has been proposed that Nrf2 can be activated through a redox-independent pathway. This involves a prior ER stress that induces an ER-specific protein kinase, termed PKR-like endoplasmic reticular kinase (PERK), which can directly phosphorylate Nrf2 (Cullinan and Diehl, 2004; Cullinan et al., 2003).
From the studies reported here, TFEC can be seen to induce some Phase II enzymes and antioxidant-responsive genes in TAMH cells, and this is likely a consequence of early Nrf2 induction. This phenomenon has not been previously reported for TFEC or other halogenated aliphatics. To establish mechanistic links to the activation of Nrf2, we examined numerous indicators of cellular oxidative stress, but the overall lack of any significant changes in these parameters suggests that classical oxidative stress does not contribute significantly to TFEC-mediated cytotoxicity in TAMH cells. These findings are consistent with previously reported in vivo findings (Groves et al., 1991). While this does not rule out a role for Keap1 in the regulation of Nrf2 translocation, it does imply that redox changes related to GSH status or reactive oxygen species (ROS) production are not pivotal in the activation pathway. In future studies it will be of interest to examine the possibility of Keap1 alterations following transient, low-level oxidative stress using highly sensitive mass spectrometric techniques.
One of the earliest effects of TFEC observed was a rapid depletion of intracellular ATP. TFEC-mediated modification and inhibition of mitochondrial aconitase and KGDH activities (both important enzymes in the TCA cycle) might result in a localized intramitochondrial oxidative stress leading to the rapid inhibition of ATP production—an important determinant of the commitment to necrosis or apoptosis (Leist et al., 1997, 1999). Intracellular ATP concentrations are critical for cell viability, and marked ATP depletion (15–25% of control) has been hypothesized to switch the cell death mechanism from apoptosis to necrosis (Lieberthal et al., 1998). In fact, our previous studies with TFEC have shown that, despite mitochondrial changes supporting apoptosis (e.g., cytochrome c release), activation of pro-apoptotic caspases does not occur. Rather, energy-independent cysteine proteases like calpains replace caspases as the major enzymes catalyzing proteolysis (Ho et al., 2005). The present work provides direct evidence for an immediate loss of ATP that is consistent with the decay of early apoptotic signals into a secondary necrosis. More importantly, we have also demonstrated that replenishment of ATP by glucose supplementation significantly restored cell viability. Even though depletion of ATP in glucose supplemented cells was still significant, it appears that a higher basal level was sufficient to "cushion" some of the damage produced by TFEC. These data confirm the importance of low intracellular ATP levels on critical downstream energy-dependent processes.
The lack of clear evidence for a generalized cellular oxidative stress, coupled to significant ATP depletion, implicates an alternative pathway for the activation of Nrf2 in this cell death model. One possibility is an ER stress–mediated Nrf2 induction. An ER resident protein kinase, PERK, is known to directly phosphorylate Nrf2 and trigger dissociation from Keap1 without the involvement of ROS (Cullinan et al., 2003). PERK itself is strongly activated as part of the unfolded protein response (UPR) and has been shown to be critical for cell survival during ER stress (Cullinan and Diehl, 2004). A recent study has also reported that ER stress–stimulated HO-1 induction occurs through Nrf2 binding to ARE, consistent with our findings (Liu et al., 2005). Furthermore, regulation of calcium homeostasis appears to be important during Nrf2 activation (Lee et al., 2003), and recently, a selective calmodulin/CaMK inhibitor, KN93, was shown to block Nrf2 activation of ARE gene induction in HepG2 cells treated with diallyl trisulfide (Chen et al., 2004).
The role of ER stress in TFEC-induced cytotoxicity is not well studied, but there are scattered reports which suggest that this might be a significant and under-appreciated phenomenon. For example, a role for calpains in renal cell death following TFEC treatment has been demonstrated (Schnellmann and Williams, 1998), in agreement with our previous work using TAMH cells (Ho et al., 2005). Of particular significance, calpain activation was blocked by overexpression of anti-apoptotic BCL-xL (Ho et al., 2005). This study also provides additional evidence for ER stress coupled with induction of specific ER stress proteins in TFEC-induced cytotoxicity. Particularly, Atf3 is recognized as an important stress-responsive ER-bound transcription factor that has been shown to induce Gadd153 (Wolfgang et al., 1997). Since covalent binding precedes all other events, these findings suggest that ER stress occurs as a result of some prior mitochondrial dysfunction. Because a strong link has already been made between ATP depletion and calcium release from the ER (Harriman et al., 2002), we have hypothesized that interruption of mitochondrial function causes a rapid depletion of intracellular ATP by TFEC, which leads to ER calcium release and an unfolded protein response (UPR), which is an energy-dependent process (Fig. 8). It is also possible that ATP depletion directly inhibits Ca-ATPase such that calcium is released from ER stores. Our ongoing efforts will include detecting activation of PERK, as well as other calcium-dependent signaling pathways, in Nrf2 phosphorylation and activation.
Overall, to the best of our knowledge, this is the first evidence of Nrf2 activation in TFEC-induced cytotoxicity. The apparent lack of oxidative stress despite clear transcriptional upregulation of classical antioxidant response genes suggests that a pathway independent of classical oxidative stress might be involved in this mitochondrially-initiated pathway of toxicity. A strong ER response following ATP deprivation provides important clues to support further investigations of ER stress–mediated Nrf2 activation.
ACKNOWLEDGMENTS
This work was supported by NIH grants GM51916 (SAB), GM32165 (SDN), CA74131 (NF), UW NIEHS sponsored Center for Ecogenetics and Environmental Health: NIEHS P30ES07033, 5 R01 ES010849 (TJK) and Pfizer Inc. Authors wish to thank the technical help of Yankai Jia, Fred Farin, Theo Bammler, Dick Beyer, and Jasmine Wilkerson for microarray and RT-PCR support, and Keck Center for fluorescence microscopy. Conflict of interest: none declared.
REFERENCES
Botta, D., Franklin, C. C., White, C. C., Krejsa, C. M., Dabrowski, M. J., Pierce, R. H., Fausto, N., and Kavanagh, T. J. (2004). Glutamate-cysteine ligase attenuates TNF-induced mitochondrial injury and apoptosis. Free Radic. Biol. Med. 37, 632–642.
Bruschi, S. A., Lindsay, J. G., and Crabb, J. W. (1998). Mitochondrial stress protein recognition of inactivated dehydrogenases during mammalian cell death. Proc. Natl. Acad. Sci. U.S.A. 95, 13413–13418.
Bruschi, S. A., West, K. A., Crabb, J. W., Gupta, R. S., and Stevens, J. L. (1993). Mitochondrial HSP60 (P1 protein) and a HSP70-like protein (mortalin) are major targets for modification during S-(1,1,2,2-tetrafluoroethyl)-L-cysteine-induced nephrotoxicity. J. Biol. Chem. 268, 23157–23161.
Buckley, B. J., Marshall, Z. M., and Whorton, A. R. (2003). Nitric oxide stimulates Nrf2 nuclear translocation in vascular endothelium. Biochem. Biophys. Res. Commun. 307, 973–939.
Chen, C., Pung, D., Leong, V., Hebbar, V., Shen, G., Nair, S., Li, W., and Tony Kong, A. N. (2004). Induction of detoxifying enzymes by garlic organosulfur compounds through transcription factor Nrf2: Effect of chemical structure and stress signals. Free Radic. Biol. Med. 37, 1578–1590.
Cooper, A. J., Bruschi, S. A., and Anders, M. W. (2002). Toxic, halogenated cysteine S-conjugates and targeting of mitochondrial enzymes of energy metabolism. Biochem. Pharmacol. 64, 553–564.
Cullinan, S. B., and Diehl, J. A. (2004). PERK-dependent activation of Nrf2 contributes to redox homeostasis and cell survival following endoplasmic reticulum stress. J. Biol. Chem. 279, 20108–20117.
Cullinan, S. B., Zhang, D., Hannink, M., Arvisais, E., Kaufman, R. J., and Diehl, J. A. (2003). Nrf2 is a direct PERK substrate and effector of PERK-dependent cell survival. Mol. Cell. Biol. 23, 7198–7209.
Dietze, E. C., Schafer, A., Omichinski, J. G., and Nelson, S. D. (1997). Inactivation of glyceraldehyde-3-phosphate dehydrogenase by a reactive metabolite of acetaminophen and mass spectral characterization of an arylated active site peptide. Chem. Res. Toxicol. 10, 1097–1103.
Groves, C. E., Lock, E. A., and Schnellmann, R. G. (1991). Role of lipid peroxidation in renal proximal tubule cell death induced by haloalkene cysteine conjugates. Toxicol. Appl. Pharmacol. 107, 54–62.
Harriman, J. F., Liu, X. L., Aleo, M. D., Machaca, K., and Schnellmann, R. G. (2002). Endoplasmic reticulum Ca(2+) signaling and calpains mediate renal cell death. Cell Death Differ. 9, 734–741.
Ho, H. K., Hu, Z. H., Tzung, S. P., Hockenbery, D. M., Fausto, N., Nelson, S. D., and Bruschi, S. A. (2005). BCL-xL overexpression effectively protects against tetrafluoroethylcysteine-induced intramitochondrial damage and cell death. Biochem. Pharmacol. 69, 147–157.
Itoh, K., Tong, K. I., and Yamamoto, M. (2004). Molecular mechanism activating Nrf2-Keap1 pathway in regulation of adaptive response to electrophiles. Free Radic. Biol. Med. 36, 1208–1213.
Itoh, K., Wakabayashi, N., Katoh, Y., Ishii, T., O'Connor, T., and Yamamoto, M. (2003). Keap1 regulates both cytoplasmic-nuclear shuttling and degradation of Nrf2 in response to electrophiles. Genes Cells 8, 379–391.
Jaiswal, A. K. (2004). Nrf2 signaling in coordinated activation of antioxidant gene expression. Free Radic. Biol. Med. 36, 1199–1207.
James, E. A., Gygi, S. P., Adams, M. L., Pierce, R. H., Fausto, N., Aebersold, R. H., Nelson, S. D., and Bruschi, S. A. (2002). Mitochondrial aconitase modification, functional inhibition, and evidence for a supramolecular complex of the TCA cycle by the renal toxicant S-(1,1,2,2-tetrafluoroethyl)-L-cysteine. Biochemistry 41, 6789–6797.
Kang, M. I., Kobayashi, A., Wakabayashi, N., Kim, S. G., and Yamamoto, M. (2004). Scaffolding of Keap1 to the actin cytoskeleton controls the function of Nrf2 as key regulator of cytoprotective phase 2 genes. Proc. Natl. Acad. Sci. U.S.A. 101, 2046–2051.
Katsuoka, F., Motohashi, H., Engel, J. D., and Yamamoto, M. (2005). Nrf2 transcriptionally activates the mafG gene through an antioxidant response element. J. Biol. Chem. 280, 4483–4490.
Lee, J. M., and Johnson, J. A. (2004). An important role of Nrf2-ARE pathway in the cellular defense mechanism. J. Biochem. Mol. Biol. 37, 139–143.
Lee, J. M., Shih, A. Y., Murphy, T. H., and Johnson, J. A. (2003). NF-E2-related factor-2 mediates neuroprotection against mitochondrial complex I inhibitors and increased concentrations of intracellular calcium in primary cortical neurons. J. Biol. Chem. 278, 37948–37956.
Leist, M., Single, B., Castoldi, A. F., Kuhnle, S., and Nicotera, P. (1997). Intracellular adenosine triphosphate (ATP) concentration: A switch in the decision between apoptosis and necrosis. J. Exp. Med. 185, 1481–1486.
Leist, M., Single, B., Naumann, H., Fava, E., Simon, B., Kuhnle, S., and Nicotera, P. (1999). Inhibition of mitochondrial ATP generation by nitric oxide switches apoptosis to necrosis. Exp. Cell Res. 249, 396–403.
Levonen, A. L., Landar, A., Ramachandran, A., Ceaser, E. K., Dickinson, D. A., Zanoni, G., Morrow, J. D., and Darley-Usmar, V. M. (2004). Cellular mechanisms of redox cell signalling: Role of cysteine modification in controlling antioxidant defences in response to electrophilic lipid oxidation products. Biochem. J. 378, 373–382.
Lieberthal, W., Menza, S. A., and Levine, J. S. (1998). Graded ATP depletion can cause necrosis or apoptosis of cultured mouse proximal tubular cells. Am. J. Physiol. 274, F315–F327.
Liu, X. M., Peyton, K. J., Ensenat, D., Wang, H., Schafer, A. I., Alam, J., and Durante, W. (2005). Endoplasmic reticulum stress stimulates heme oxygenase-1 gene expression in vascular smooth muscle. Role in cell survival. J. Biol. Chem. 280, 872–877.
Lock, E. A., and Ishmael, J. (1998). The nephrotoxicity and hepatotoxicity of 1,1,2,2-tetrafluoroethyl-L-cysteine in the rat. Arch. Toxicol. 72, 347–354.
Nakaso, K., Yano, H., Fukuhara, Y., Takeshima, T., Wada-Isoe, K., and Nakashima, K. (2003). PI3K is a key molecule in the Nrf2-mediated regulation of antioxidative proteins by hemin in human neuroblastoma cells. FEBS Lett. 546, 181–184.
Nguyen, T., Sherratt, P. J., and Pickett, C. B. (2003). Regulatory mechanisms controlling gene expression mediated by the antioxidant response element. Annu. Rev. Pharmacol. Toxicol. 43, 233–260.
Numazawa, S., Ishikawa, M., Yoshida, A., Tanaka, S., and Yoshida, T. (2003). Atypical protein kinase C mediates activation of NF-E2-related factor 2 in response to oxidative stress. Am. J. Physiol. Cell Physiol. 285, C334–C342.
Numazawa, S., and Yoshida, T. (2004). Nrf2-dependent gene expressions: A molecular toxicological aspect. J. Toxicol. Sci. 29, 81–89.
Pierce, R. H., Franklin, C. C., Campbell, J. S., Tonge, R. P., Chen, W., Fausto, N., Nelson, S. D., and Bruschi, S. A. (2002). Cell culture model for acetaminophen-induced hepatocyte death in vivo. Biochem. Pharmacol. 64, 413–424.
Sakurai, A., Nishimoto, M., Himeno, S., Imura, N., Tsujimoto, M., Kunimoto, M., and Hara, S. (2005). Transcriptional regulation of thioredoxin reductase 1 expression by cadmium in vascular endothelial cells: Role of NF-E2-related factor-2. J. Cell Physiol. 203, 529–537.
Schnellmann, R. G., and Williams, S. W. (1998). Proteases in renal cell death: Calpains mediate cell death produced by diverse toxicants. Ren. Fail. 20, 679–686.
Wolfgang, C. D., Chen, B. P., Martindale, J. L., Holbrook, N. J., and Hai, T. (1997). gadd153/Chop10, a potential target gene of the transcriptional repressor ATF3. Mol. Cell. Biol. 17, 6700–6707.
Wu, J. C., Merlino, G., Cveklova, K., Mosinger, B., Jr., and Fausto, N. (1994). Autonomous growth in serum-free medium and production of hepatocellular carcinomas by differentiated hepatocyte lines that overexpress transforming growth factor alpha 1. Cancer Res. 54, 5964–5973.
Yu, R., Chen, C., Mo, Y. Y., Hebbar, V., Owuor, E. D., Tan, T. H., and Kong, A. N. (2000). Activation of mitogen-activated protein kinase pathways induces antioxidant response element-mediated gene expression via a Nrf2-dependent mechanism. J. Biol. Chem. 275, 39907–39913.
Zhang, D. D., and Hannink, M. (2003). Distinct cysteine residues in Keap1 are required for Keap1-dependent ubiquitination of Nrf2 and for stabilization of Nrf2 by chemopreventive agents and oxidative stress. Mol. Cell. Biol. 23, 8137–8151.(Han K. Ho, Collin C. Whit)