Bcl-2 Attenuation of Oxidative Cell Death Is Associated with Up-regulation of γ-Glutamylcysteine Ligase via Constitutive NF-κB Activation

Oxidative stress induced by reactive oxygen intermediates often causes cell death via apoptosis, which is regulated by many functional genes and their protein products. The evolutionarily conserved protein Bcl-2 blocks apoptosis induced by a wide array of death signals. Despite extensive research, the molecular milieu that characterizes the anti-apoptotic function of Bcl-2 has not been fully clarified. In this work, we have investigated the role of bcl-2 in protecting against oxidative death induced by H2O2 in cultured rat pheochromocytoma PC12 cells. Transfection with the bcl-2 gene rescued PC12 cells from apoptotic death caused by H2O2. Addition of NF-κB inhibitors such as pyrrolidine dithiocarbamate and N-tosyl-l-phenylalanine chloromethyl ketone to the medium aggravated oxidative cell death. PC12 cells overexpressing bcl-2 exhibited relatively high constitutive DNA binding and transcriptional activities of NF-κB compared with vector-transfected control cells. Western blot analysis and immunocytochemistry revealed that bcl-2-transfected PC12 cells retained a higher level of p65 (the functionally active subunit of NF-κB) in the nucleus compared with vector-transfected controls. In addition, sustained activation of ERK1/2 (upstream of NF-κB) was observed in bcl-2-overexpressing cells. In contrast, the cytoplasmic inhibitor IκBα was present in lower amounts in cells overexpressing bcl-2. The ectopic expression of bcl-2 increased the cellular glutathione level and γ-glutamylcysteine ligase expression, which were attenuated by NF-κB inhibitors. These results suggest that NF-κB plays a role in bcl-2-mediated protection against H2O2-induced apoptosis in PC12 cells through augmentation of antioxidant capacity.

Oxidative stress refers to the mismatched redox equilibrium between the production of reactive oxygen intermediates (ROIs) 1 and the ability of the cell to defend against them.
Oxidative stress thus occurs when the production of ROIs increases or when the elimination of ROIs or the repair of oxidatively damaged macromolecules decreases or both. There is a growing body of data indicating that ROIs are a major cause of cellular injuries in a variety of clinical abnormalities, including cancer, diabetes, rheumatoid arthritis, and neurodegenerative disorders (1,2).
Multiple lines of evidence support that ROIs can cause cell death via apoptosis (3)(4)(5). The concentration of ROIs and the microenvironment appear to be important in determining the mode of cell death (6). Cells undergoing apoptosis exhibit shrinkage of the nucleus, blebbing of membranes, condensation or fragmentation of chromatin, and internucleosomal DNA degradation by endonucleases into fragments in multiples of 180 -200 bp. Apoptosis is a tightly regulated process that involves changes in the expression of a distinct set of genes (7,8). One of the major genes responsible for regulating apoptotic cell death is the proto-oncogene bcl-2, which encodes a 26-kDa protein found in the nuclear envelope, parts of the endoplasmic reticulum, and the outer mitochondrial membrane. The bcl-2 gene product has been shown to prolong cell survival by blocking apoptosis induced by a wide array of death signals (9 -11). In addition, Bcl-2 heterodimerizes with Bax to form a high molecular mass complex. Bcl-2 may prevent apoptosis through regulation of an antioxidant pathway and is considered to act as a free radical scavenger. For instance, the protein was found to inhibit lipid peroxidation and oxidative DNA and/or protein damage induced by diverse pro-apoptotic stimuli capable of triggering apoptosis (9 -11). Furthermore, induction or overexpression of bcl-2 is thought to confer resistance to oxidant injury (12)(13)(14).
The ubiquitous eukaryotic transcription factor NF-B/Rel is known to regulate expression of numerous cellular genes that play important roles in mediating/regulating immune and stress responses, inflammation, apoptosis, and proliferation (15,16). Recent studies have revealed that NF-B is involved in regulating cell survival. Overexpression of NF-B increases cell viability by suppressing induction of apoptosis in various cell types (17)(18)(19). Cells that become resistant to oxidative cell death exhibit constitutive activation of NF-B as an adaptive defense mechanism (20,21). Conversely, inhibitors of NF-B activity such as SN50, N-tosyl-L-phenylalanine chloromethyl ketone (TPCK), and pyrrolidine dithiocarbamate (PDTC) attenuate the survival of neurons, supporting that NF-B is required for neuronal protection (18,20,22,23). However, the molecular mechanisms underlying the anti-apoptotic effect of NF-B have not been clarified.
In this work, we investigated whether Bcl-2 can protect against H 2 O 2 -induced apoptosis through activation of NF-B in cultured rat pheochromocytoma PC12 cells. For this purpose, we compared the extent of NF-B activation and levels of antioxidative defense capacity, especially glutathione metabolism, in bcl-2-transfected and vector-treated control cells to link Bcl-2 and the NF-B signaling pathways in the context of their commitment to cellular protection against oxidative insults.
Cell Culture-PC12 cells transfected with a eukaryotic expression vector containing the human cytomegalovirus major immediate-early enhancer/promoter followed by a full-length human Bcl-2 cDNA sequence were kindly provided by Dr. Young J. Oh (Yonsei University, Seoul) and maintained in our laboratory. Briefly, DNA transfection was performed with 1 ϫ 10 5 PC12 cells cultivated on poly-D-lysine-coated 100-mm Petri dishes by adding a transfection mixture of 2 g of plasmid/10 l of LipofectAMINE (Invitrogen) in Dulbecco's modified Eagle's medium. Subsequently, single neomycin-resistant colonies were selected and expanded in Dulbecco's modified Eagle's medium supplemented with 10% fetal bovine serum, 5% horse serum, and 500 g/ml G418. Stable PC12 cell lines overexpressing bcl-2 were characterized by immunoblot analysis. PC12 cells transfected with a eukaryotic expression vector without a human bcl-2 cDNA sequence were utilized as a control cell line. The subsequent cultures were conducted as reported previously (6).
Determination of Cell Viability-PC12 cells were plated at a density of 4 ϫ 10 4 cells/300 l in 48-well plates, and cell viability was determined using the conventional MTT reduction assay. After incubation, cells were treated with MTT solution (1 mg/ml final concentration) for 2 h. The dark blue formazan crystals formed in intact cells were solubilized with lysis buffer (20% SDS in 50% aqueous N,N-dimethylformamide), and absorbance at 540 -595 nm was measured with a microplate reader (Molecular Devices, Inc., Sunnyvale, CA). Results are expressed as percent MTT reduction.
Terminal Deoxynucleotidyltransferase-mediated dUTP Nick End Labeling (TUNEL)-The commercially available in situ death detection kit was utilized to assess DNA fragmentation. PC12 cells (5 ϫ 10 5 cells/3 ml on a chamber slide) were fixed for 30 min in 10% neutral buffered formalin solution at room temperature. Endogenous peroxidase was inactivated by incubation with 0.3% hydrogen peroxide in methanol for 30 min at room temperature and further incubated in a permeabilizing solution (0.1% sodium citrate and 0.1% Triton X-100) for 2 min at 4°C. The cells were incubated with the TUNEL reaction mixture for 60 min at 37°C, followed by labeling with peroxidaseconjugated anti-goat antibody (Fab fragment) for an additional 30 min. The cells were rinsed with phosphate-buffered saline (PBS) and examined under a confocal microscope (Leica Microsystems Heidelberg GmbH, Heidelberg, Germany) with excitation at 488 nm and emission at 525 nm.
Measurement of the Mitochondrial Transmembrane Potential-To measure the mitochondrial membrane potential (⌬⌿ m ), the lipophilic cationic probes JC-1 and TMRE were used. The green fluorescent JC-1 probe exists as a monomer at low membrane potential. However, at higher potential, JC-1 forms red fluorescent J-aggregates that exhibit a broad excitation spectrum. Following treatment with H 2 O 2 for 6 h, cells (5 ϫ 10 5 cells/3 ml on a chamber slide) were rinsed with PBS, and JC-1 (10 g/ml) was loaded. After a 20-min incubation at 37°C, cells were examined under a confocal microscope with excitation at 488 nm and emission at 530/590 nm. Determination of ⌬⌿ m was also carried out using TMRE, which rapidly equilibrates between cellular compartments due to potential differences. Thus, a decrease in fluorescence is indicative of reduced ⌬⌿ m . Following the incubation, the cells were treated with TMRE (150 nM) for 30 min, rinsed, and examined by confocal microscopy in the same manner as done for JC-1, except that the TMRE fluorescence was measured at 590 nm.
Measurement of Intracellular ROI Accumulation-To monitor net intracellular accumulation of ROIs, the fluorescent probe DCF-DA was used. After treatment with H 2 O 2 for 30 min, cells (1 ϫ 10 6 cells/3 ml in 6-well plates) were rinsed with Krebs-Ringer phosphate solution, and 10 M DCF-DA was loaded. Following an additional incubation for 15 min at 37°C, cells were examined under a confocal microscope equipped with an argon laser (488 nm, 200 milliwatts).
Preparation of Nuclear Proteins-After treatment with H 2 O 2 , cells (1 ϫ 10 7 cells/7 ml in a 100-mm dish) were washed with PBS, centrifuged, and resuspended in ice-cold isotonic buffer (10 mM HEPES (pH 7.9), 1.5 mM MgCl 2 , 10 mM KCl, 0.5 mM dithiothreitol, and 0.2 mM phenylmethylsulfonyl fluoride). After incubation in an ice bath for 10 min, cells were centrifuged again and resuspended in ice-cold buffer containing 20 mM HEPES (pH 7.9), 20% glycerol, 420 mM NaCl, 1.5 mM MgCl 2 , 0.2 mM EDTA, 0.5 mM dithiothreitol, and 0.2 mM phenylmethylsulfonyl fluoride, followed by incubation at 0°C for 20 min. After Vortex mixing, the resulting suspension was centrifuged, and the supernatant was stored at Ϫ70°C for the NF-B DNA binding assay. The protein concentration was determined using the BCA protein assay kit.
Electrophoretic Mobility Shift Assay-A synthetic double-strand oligonucleotide harboring the NF-B-binding domain was labeled with [␥-32 P]ATP using T4 polynucleotide kinase and separated from unincorporated [␥-32 P]ATP by gel filtration using a nick spin column (Amersham Biosciences). Prior to addition of the ␥-32 P-labeled oligonucleotide (100,000 cpm), 10 g of the nuclear extract was kept on ice for 15 min in gel shift binding buffer (4% glycerol, 1 mM EDTA, 1 mM dithiothreitol, 100 mM NaCl, 10 mM Tris-HCl (pH 7.5), and 0.1 mg/ml sonicated salmon sperm DNA). DNA-protein complexes were resolved by 6% nondenaturing PAGE at 200 V for 2 h, followed by autoradiography.
Immunocytochemical Staining for p65 and Phospho-ERK-To detect the activation of p65 and ERK, we have adopted the immunocytochemical method with a monoclonal antibody recognizing p65 or phospho-ERK. Cells (10 5 cells/1 ml on a chamber slide) were fixed in 10% neutral buffered formalin solution for 30 min at room temperature. After rinsing with PBS, cells were blocked for 1 h at room temperature in fresh blocking buffer (0.5% PBST, pH 7.4) containing 10% normal goat serum). Dilutions (1:100) of primary antibodies were made in 0.1% PBST with 1% bovine serum albumin, and cells were incubated overnight at 4°C. Following two washes with 0.1% PBST, the cells were incubated with fluorescein isothiocyanate-conjugated goat anti-rabbit IgG second-ary antibody in 0.1% PBST with 3% bovine serum albumin for 1 h at room temperature. Cells were rinsed with PBS, and stained cells were analyzed under a confocal microscope and photographed.
Transient Transfection and Luciferase Assay-One day before transfection, PC12 cells were subcultured at a density of 1 ϫ 10 6 cells/60-mm dish to maintain ϳ60 -80% confluency. They were transiently transfected with the NF-B or GCL promoter-luciferase construct using the transfection reagent N- [1-(2,3-dioleoyloxy)propyl]-N,N,N-trimethylammonium (Roche Diagnostics) according to the instructions supplied by the manufacturer. After overnight transfection, the treated cells were harvested and lysed with reporter lysis buffer (Promega luciferase assay system). The cell extract (20 l) was mixed with 100 l of the luciferase assay reagent and analyzed with a luminometer (AutoLumat LB 953, EG&G Berthold, Bad Widbad, Germany). The ␤-galactosidase assay (Promega ␤-galactosidase enzyme assay system) was done according to the supplier's instructions to normalize the luciferase activity.
Assessment of Intracellular GSH Levels-The intracellular GSH levels were assessed using the commercially available colorimetric assay kit BIOXYTECH GSH-400 (OXIS Research, Portland, OR). Cells were harvested and homogenized in meta-phosphoric acid working solution. After centrifugation, 50 l of R1 solution (solution of the chromogenic reagent in HCl) was added to the 700-l supernatant, followed by gentle Vortex mixing. Following addition of 50 l of R2 solution (30% NaOH), the mixtures were incubated at 25 Ϯ 3°C for 10 min. After centrifugation, the absorbance of the clear supernatant was read at 400 nm. The protein concentration was determined using the BCA protein assay kit.

bcl-2-Overexpressing PC12 Cells Are Less Susceptible to the Cytotoxic Effects of H 2 O 2 -
To address whether ectopic expression of Bcl-2 can protect against H 2 O 2 -induced cell death, PC12 cells were stably transfected with a plasmid harboring the bcl-2 gene. The effects of bcl-2 overexpression on cell survival following exposure to H 2 O 2 were then examined by the MTT reduction assay. As illustrated in Fig. 1A, bcl-2 overexpression rescued PC12 cells from H 2 O 2 -induced cytotoxicity.
Bcl-2 Overexpression Attenuates Apoptotic Cell Death Induced by H 2 O 2 -One of the distinctive biochemical hallmarks of apoptotic cell death is the unique occurrence of internucleosomal DNA fragmentation. Apoptotic cells were confirmed by TUNEL staining, which is widely used to detect DNA fragmentation in situ. Bcl-2 overexpression reduced H 2 O 2 -induced DNA fragmentation as revealed by the decreased proportion of TUNEL-positive cells (Fig. 1B). H 2 O 2 treatment also led to the increased expression of Bax, activation of caspase-3, and cleavage of poly(ADP-ribose) polymerase, which were attenuated in bcl-2-overexpressing cells (Fig. 1C).
bcl-2 Mitigates the Dissipation of the Mitochondrial Transmembrane Potential and Intracellular Accumulation of Hydroperoxide in H 2 O 2 -treated PC12 Cells-When PC12 cells were exposed to H 2 O 2 (250 M), the mitochondrial membrane became rapidly depolarized, as shown by an increase in green fluorescence and the concomitant disappearance of red fluorescence derived from the JC-1 dye (Fig. 2A, panels a and c). Bcl-2 overexpression reduced the changes in mitochondrial membrane transition (⌬⌿ m ) as indicated by repression of green fluorescence and restoration of red fluorescence ( Fig. 2A, panels b and d). These findings were further supported by use of another voltage-sensitive dye, TMRE. Again, H 2 O 2 -induced dissipation of ⌬⌿ m ( Fig. 2A, panel e) was found to be blocked by bcl-2 overexpression (panel f). Accumulation of intracellular hydroperoxide was detected by use of DCF-DA, which is freely permeable to cell membranes. Once inside cells, the compound is hydrolyzed by an esterase activity to DCF and trapped intracellularly. DCF is then able to interact with peroxides to form fluorescent 2Ј,7Ј-dichlorofluorescin, which is readily detectable by confocal microscopy. The activation of DCF is relatively specific for the detection of H 2 O 2 and secondary or tertiary peroxides such as lipid peroxides. H 2 O 2 -derived intracellular peroxide accumulation decreased in PC12 cells overexpressing bcl-2 (Fig. 2B).

NF-B Inhibitors Render PC12 Cells More Vulnerable to H 2 O 2 -induced Cell
Death-NF-B activation in PC12 cells was assessed by electrophoretic mobility shift assay with an oligonucleotide harboring a consensus NF-B-binding element. Treatment of PC12 cells with H 2 O 2 (100, 250, and 500 M) caused a concentration-dependent increase in NF-B DNA binding activity in these cells (Fig. 3A). PDTC, an antioxidant reported to effectively block the IB degradation pathway (24,25),reducedtheDNAbindingactivityofNF-Binaconcentrationdependent manner (Fig. 3B). To examine the possible role of NF-B in protecting against H 2 O 2 -induced cytotoxicity, cells were exposed to H 2 O 2 for 9 h in the absence and presence of PDTC (10 M) or another NF-B inhibitor, TPCK (5 M), and cell viability was assessed by the MTT reduction assay. Both PDTC (Fig. 3C) and TPCK (Fig. 3D) (Fig. 4A). PC12 cells overexpressing bcl-2 exhibited relatively high levels of consti-tutively activated NF-B compared with vector-transfected control cells (Fig. 4A). Since the NF-B DNA binding activity is largely regulated by IB␣, which sequesters NF-B in the cytoplasm, we determined whether the observed increase in nuclear NF-B binding activity in bcl-2-overexpressing PC12 cells is due to increased IB␣ degradation. Protein extracts of both the nucleus and cytoplasm were subjected to Western blot analysis to measure p65, IB␣, or phospho-IB␣. Cytoplasmic IB␣ levels were profoundly reduced, whereas phospho-IB and nuclear p65 levels were constitutively increased in PC12 cells overexpressing bcl-2 compared with control cells (Fig. 4B). We also verified the nuclear accumulation of p65 by immunocytochemistry using anti-p65 antibody (Fig. 4C). The transcriptional activity of NF-B was also constitutively increased in bcl-2-transfected cells as assessed using an NF-B-reporter plasmid containing the consensus NF-B-binding DNA site linked to a luciferase reporter gene (pELAM-Luc). As illustrated in Fig. 4D, the base-line transcriptional activity of NF-B was found to be approximately six times higher in bcl-2-transfected cells than in vector-transfected control cells. To elucidate a molecular target for NF-B-mediated potentiation of cellular defense against oxidative insult in bcl-2-overexpressing cells, we examined the effect of GSH on H 2 O 2induced cell death. GSH, a ubiquitous tripeptide thiol, is a vital intra-and extracellular antioxidant against oxidative stress. N-Acetyl-L-cysteine undergoes a rapid deacetylation in cells

FIG. 2. Effect of bcl-2 overexpression on H 2 O 2 -induced dissipation of mitochondrial membrane potential (⌬⌿ m ) and intracellular accumulation of ROIs.
A, ⌬⌿ m was assessed by changes in JC-1 fluorescence (panels a-d) as described under "Experimental Procedures." TMRE fluorescence was monitored to confirm the changes in ⌬⌿ m (panels e and f). Cells transfected with the bcl-2-expressing vector (panels b, d, and f) or with the control neo gene plasmid (panels a, c, and e) were exposed to H 2 O 2 (250 M) for 6 h. B, intracellular ROI levels were determined based on DCF fluorescence as described under "Experimental Procedures." Cells were exposed to H 2 O 2 (250 M) for 30 min. and provides a rate-limiting amino acid (cysteine) needed for the intracellular synthesis of GSH, thereby replenishing the depleted levels of GSH. Pretreatment with GSH or N-acetyl-Lcysteine protected PC12 cells from H 2 O 2 -induced cell death, which was aggravated by depletion of GSH by L-buthionine-(SR)-sulfoximine, a specific inhibitor of ␥-glutamylcysteine ligase (Fig. 4E).

FIG. 3. Effect of NF-B inhibitors on H 2 O 2 -induced cytotoxicity in bcl-2-overexpressing PC12 cells. A,
Bcl-2 Potentiates Intracellular GSH Capacity through NF-B Activation-GSH is formed by the consecutive actions of GCL and glutathione synthetase. The rate-limiting enzyme in cellular GSH biosynthesis is GCL. bcl-2-overexpressing PC12 cells exhibited an increased cellular GSH pool (Fig. 5A) and increased GCLC mRNA levels compared with vector-transfected control cells. In contrast, the mRNA levels of the regulatory subunit (GCLM) remained almost the same in both bcl-2-and vector-transfected cells (Fig. 5B). The increased GSH production and mRNA expression of GCLC were attenuated by pretreatment with NF-B inhibitors (Fig. 5, A and B). The transcriptional activity of GCLC was measured using GCLC promoter-luciferase constructs. PC12 cells transfected with the empty construct pGL3 served as background controls. The luciferase activity of GCLC was significantly higher in bcl-2overexpressing cells (Fig. 5C) and was effectively suppressed by treatment with the NF-B inhibitors (Fig. 5D).
ERK Activation Is Increased in bcl-2-overexpressing PC12 Cells-To explore the intracellular signaling transduction pathways that mediate Bcl-2-induced constitutive NF-B activation in PC12 cells, we examined the activation of MAPKs such as ERK1/2, JNK, and p38 MAPK, which are important components of intracellular signaling cascades that regulate oxidative cell death and/or survival. The activation of ERK1/2 was assessed by Western blot analysis (Fig. 6A) and immunocytochemistry (Fig. 6B) using a phospho-specific antibody. PC12 cells overexpressing bcl-2 exhibited constitutive and sustained activation of ERK via phosphorylation (Fig. 6, A and B). The levels of unphosphorylated or total ERK1/2 remained unchanged even after H 2 O 2 treatment, and there were no significant differences in the levels of the parental/unphosphorylated form of this MAPK between cells transfected with bcl-2 and the blank vector (Fig. 6A). Pretreatment with 20 M U0126, an ultrapotent inhibitor of MEK1/2 upstream of ERK1/2, augmented the H 2 O 2 -induced cytotoxicity (Fig. 6C). H 2 O 2 stimulation also led to the transient activation of JNK and p38 via phosphorylation. Whereas JNK activation was slightly alleviated, p38 was not much affected by bcl-2 overexpression (Fig. 6D).
Bcl-2-mediated Activation of NF-B Is Suppressed by Inhibition of ERK1/2-In an attempt to elucidate the molecular mechanism underlying bcl-2-mediated NF-B activation, we examined the effect of a pharmacological MEK1/2 inhibitor (U0126) or a dominant-negative mutation of ERK1/2 on the activation of NF-B. The increased NF-B DNA binding in bcl-2-overexpressing PC12 cells was attenuated by pretreatment with U0126 or transient transfection with a dominantnegative mutant form of ERK1/2 (Fig. 7A). Both pharmacological and genetic inhibition of ERK abrogated the NF-B DNA binding activity, suggesting that NF-B activation in bcl-2overexpressing cells is mediated partly via the ERK signaling pathway. Consistent with the above gel shift data, the increased nuclear p65 and decreased cytoplasmic IB␣ levels in bcl-2-transfected PC12 cells were reduced by U0126 treatment (Fig. 7B).

DISCUSSION
Several protein families highly conserved during evolution are considered to be specifically involved in regulating programmed or apoptotic cell death. Examples are Bax and Bcl-2, which form homo-and heterodimers in vivo. The relative expression of these two proteins determines cell death or survival. When Bcl-2 is in excess, the Bcl-2 homodimer predominates and promotes cell survival. It has been reported that overexpression of bcl-2 or the related gene bcl-x L protects a variety of mammalian cells from apoptosis induced by diverse death stimuli, including chemotherapeutic agents, radiation, tumor necrosis factor-␣, glucocorticoids, glutamate, and withdrawal of serum or growth factors (9, 26 -28). Bcl-2 has been proposed to prevent apoptosis by regulating an antioxidant pathway (29). In line with this notion, Bcl-2 reduces the intracellular levels of ROIs associated with apoptosis (30,31), and disruption of the proper expression of this anti-apoptotic protein is likely to increase vulnerability of cells to apoptotic signals (32,33). In this work, Bcl-2 mitigated not only distinct morphological as well as biochemical changes related to apoptotic death, but also intracellular accumulation of ROIs in H 2 O 2 -treated PC12 cells.
The molecular events and genetic programs activated in response to oxidative stress and those involved in providing cells with resistance against oxidative insults remain to be unraveled. Accumulating evidence supports a role of the ubiquitous eukaryotic transcription factor NF-B in regulating oxidative stress-induced cell damage (34). Consistent with this notion, our present study has revealed that the DNA binding of NF-B and its transcriptional activity are constitutively increased in Bcl-2-overexpressing PC12 cells compared with vector-transfected control cells. Overexpression of NF-B/Rel promotes cell survival by hampering the induction of apoptosis (35)(36)(37). High constitutive NF-B activation also confers resistance to oxidative stress in neuronal cells (20). Conversely, NF-B inhibitors have been found to decrease cell viability by stimulating apoptosis. Inhibition of p65 nuclear translocation by PDTC, which is capable of blocking IB phosphorylation, and the peptide proteasome inhibitor or addition of SN50, a cell-permeable specific inhibitor of NF-B, all reduce NF-B activity and increase apoptosis (22,38). When PC12 cells were treated with NF-B inhibitors, H 2 O 2 -induced cell death was aggravated, supporting the involvement of NF-B in cell survival.
Although NF-B appears to be an important component of the cellular response to oxidative stress, the molecular mechanisms of its activation and its role in regulating transcription of genes involved in cellular antioxidant defense are complex. Such complexity reflects the multiplicity of binding proteins and the existence of a large number of different gene-specific sequences recognizing NF-B. There is a growing body of data supporting the role of NF-B in the regulation of anti-apoptotic gene expression and promotion of cell survival. GSH as a universal cellular thiol would be one of the potential target molecules for the anti-apoptotic functions of NF-B against oxidative stress. GSH has been shown to inhibit or retard apoptosis triggered by many different stimuli, including oxidants, cytokines, and anti-Fas/APO-1 (CD95) antibody (39). Conversely, depletion of intracellular GSH has been reported to occur with the onset of apoptosis and is frequently accompanied by a concomitant increase in the accumulation of ROIs. As a ratelimiting enzyme in GSH biosynthesis, GCL expression/activity is modulated by oxidants, antioxidants, growth factors, and inflammation-related agents (40). The GCL holoenzyme is composed of a heavy catalytic subunit (GCLC, 73 kDa) and a light regulatory subunit (GCLM, 30 kDa) (41). The 5Ј-flanking regions of both human GCL subunits have been characterized, and putative binding sites for NF-B have been identified in the promoter region of the heavy subunit (42,43). In our study, Bcl-2-overexpressing PC12 cells exhibited an increased GSH pool and increased GCLC expression. Moreover, blocking the activation of NF-B led to down-regulation of GCLC and subsequent depletion of GSH. Based on these findings, we propose that NF-B fortifies the cellular protection against oxidative stress through augmentation of antioxidative capacity, particularly GSH biosynthesis. Additional studies will be necessary to search for antioxidant molecules or enzymes other than GCL that are involved in cellular antioxidant defense mechanisms regulated by NF-B in PC12 cells overexpressing bcl-2.
Because Bcl-2 possesses no inherent kinase activity, we hypothesized that bcl-2 possibly impacts on cellular factors that directly or indirectly lead to NF-B activation. The critical regulatory step in the activation of NF-B is the phosphorylation of IB␣ and other IB proteins, which is mediated by a high molecular mass multiprotein complex called IB kinase. In bcl-2-overexpressing cells, the relatively high constitutive NF-B activity we observed was associated with enhanced degradation of IB␣, consistent with findings in mature murine B-cells (44). The N-terminal region of IB␣ has been proposed to be an important regulatory site for Bcl-2 (45,46). However, the molecular mechanism by which Bcl-2 mediates NF-B activation through interaction with IB is not completely clarified. One possibility is that Bcl-2 directly or indirectly modulates IB activity by interacting with one of the cellular factors that are involved in the activation of IB kinase. IB kinase is phosphorylated and activated by one or more upstream activating kinases, which are likely to be the members of the MAPK kinase kinase family of enzymes (also known as MAP3Ks and MEKKs). MEKK1, which phosphorylates the upstream kinase of MAPKs, was shown to bind and phosphorylate IB kinase (47). Alternatively, MEKK2 and MEKK3 also have the potential to activate IB kinase and thereby stimulate NF-B activation (48). Other evidence supports that regulation of the IB activity by Bcl-2 may be mediated by a mechanism that involves the Raf-1/MEKK1 signaling pathway (49). This study suggests that Bcl-2, through the Bcl-2 homology 4 domain, interacts with Raf-1, leading to the downstream activation of MEKK1 and subsequent IB kinase-dependent NF-B activation. However, under certain experimental conditions, bcl-2 overexpression can negatively regulate the activation of NF-B (50,51). It depends on the conditions of the manipulated cells, including the transfection system and differentiation status. It has been reported that a sustained low level of bcl-2 expression resembles stable transfected cell lines; however, transiently increased high level expression of bcl-2 may result in immediate cellular alterations, which have not yet been characterized in stable clonal cell lines (52). The basal activity of NF-B in nerve growth factor-treated cells is high, which is required for the survival of neurons (53), so bcl-2 overexpression can differentially regulate stress stimuli-induced NF-B activation.
Recently, ROIs have gained special attention because of their potential role as second messengers in the intracellular signaling network (54,55). Although oxidative stress may directly activate redox-sensitive transcription factors including NF-B, ROIs may function as secondary messengers in various cellular signaling cascades. Exogenous H 2 O 2 has been demonstrated to activate MAPKs such as ERK, JNK, and p38 MAPK (56,57). However, the roles of MAPKs in cell death or survival are controversial. In general, activation of ERK occurs in response to growth factor stimulation (58), whereas JNK and p38 MAPK are activated after exposure to environmental stress such as ROIs, UV irradiation, hyperosmolarity, and endotoxins (59).

FIG. 7. Effect of ERK inhibition on increased NF-B activation in bcl-2-overexpressing PC12 cells.
To block the activation of ERK, dominant-negative mutant ERK and U0126 were utilized. A, Bcl-2overexpressing PC12 cells were transiently transfected with dominantnegative mutant ERK (DN ERK) or control wild-type (WT) plasmid or treated with increasing concentrations of U0126 (10 and 25 M) for 6 h. Nuclear protein was extracted as described under "Experimental Procedures." NF-B DNA binding activity was measured by electrophoretic mobility shift assay. P, probe only. B, PC12 cells overexpressing bcl-2 were incubated with increasing concentrations of U0126 (5, 10, and 25 mM) for 6 h. IB␣ in a cytosolic extract (CE) and p65 in a nuclear extract (NE) was compared by Western blot analysis.
ERK is regarded as an anti-apoptotic kinase, although it may control proliferation of certain cells, either positively or negatively, depending on the duration of its activation (60). In our study, treatment of PC12 cells with H 2 O 2 led to transient activation of ERK, which was constitutively up-regulated in bcl-2-overexpressing PC12 cells. However, bcl-2 overexpression did not cause any substantial alterations in the phosphorylation of p38 MAPK, whereas JNK activation was slightly diminished. The ERK signaling cascade has been implicated in NF-B activation through phosphorylation of inhibitory IB (61). The association between the ERK signaling cascade and NF-B activation is also supported by the finding that the ERK-regulated kinase p90 RSK phosphorylates and thereby inactivates IB in response to mitogenic stimuli (62). Overexpression of bcl-2 in the PC12 cell line leads to phosphorylation of c-Jun at Ser 73 via the ERK pathway, which contributes to the anti-apoptotic function of bcl-2 (63). bcl-2 overexpression increases expression of neural differentiation-associated genes through the TrkA/MEK/ERK pathway (64). Neuroprotection by transforming growth factor-␤1 involves activation of NF-B through the Akt/protein kinase B and ERK1/2 signaling pathways, further supporting that NF-B is a target of ERK signaling (65).
In summary, ectopic expression of bcl-2 in PC12 cells protected these cells from apoptotic death induced by H 2 O 2 . Of particular interest is that PC12 cells overexpressing bcl-2 exhibited relatively high levels of constitutively activated NF-B and the upstream kinase ERK1/2 compared with vector-transfected control cells. Pharmacological inhibition of NF-B or ERK1/2 aggravated H 2 O 2 -induced PC12 cell death, suggesting that both NF-B and ERK are responsible for protection of PC12 cells from oxidative stress.
Taken together, these findings suggest that constitutive activation of the redox-sensitive transcription factor NF-B is part of a self-defense program that enables neuronal cells to protect themselves against oxidative stress. Considering the notion that some neuronal cells can survive accumulating oxidative damages and degenerative processes, an understanding of the molecular mechanisms that can alleviate the vulnerability of neurons and consequently increase their resistance to oxidative stress is of great interest in the context of establishing therapeutic strategies for the management of neurodegenerative disorders.