Activation of GluR6-containing Kainate Receptors Induces Ubiquitin-dependent Bcl-2 Degradation via Denitrosylation in the Rat Hippocampus after Kainate Treatment*

We previously showed that Bcl-2 (B-cell lymphoma 2) is down-regulated in a kainate (KA)-induced rat epileptic seizure model. The underlying mechanism had remained largely unknown, but we here report for the first time that denitrosylation and ubiquitination are involved. Our results show that the S-nitrosylation levels of Bcl-2 are down-regulated after KA injection and that the GluR6 (glutamate receptor 6) antagonist NS102 can inhibit the denitrosylation of Bcl-2. Moreover, the ubiquitin-dependent degradation of Bcl-2 was found to be promoted after KA treatment, which could be suppressed by the proteasome inhibitor MG132 and the NO donors, sodium nitroprusside and S-nitrosoglutathione. In addition, experiments based on siRNA transfections were performed in the human SH-SY5Y neuroblastoma cell line to verify that the stability of Bcl-2 is causal to neuronal survival. At the same time, it was found that the exogenous NO donor GSNO could protect neurons when Bcl-2 is targeted. Subsequently, these mechanisms were morphologically validated by immunohistochemistry, cresyl violet staining, and in situ TUNEL staining to analyze the expression of Bcl-2 as well as the survival of CA1 and CA3/DG pyramidal neurons. NS102, GSNO, sodium nitroprusside, and MG132 contribute to the survival of CA1 and CA3/DG pyramidal neurons by attenuating Bcl-2 denitrosylation. Taken together, our data reveal that Bcl-2 ubiquitin-dependent degradation is induced by Bcl-2 denitrosylation during neuronal apoptosis after KA treatment.

The agitation of glutamate receptors is thought to be the primary cause of epileptic seizures. Glutamate receptors are classified into metabolic glutamate receptors (GluRs) 3 and ionic GluRs. Glutamate itself gates three types of ionotropic receptors: NMDA, AMPA, and kainate (KA) receptors. There are also five types of kainate receptor subunits: GluR5, GluR6, GluR7, KA1, and KA2 (1). These subunits can be arranged in different ways to form a tetramer. GluR5-7 can form homomers (for example, a receptor composed entirely of GluR5) and heteromers (for example, a receptor composed of both GluR5 and GluR6). However, KA1 and KA2 can only form functional receptors by combining with one of the GluR5-7 subunits (2). Kainic acid (KA) is a potent exogenous agonist of the KA receptors, and the systemic administration of KA produces epilepsy in rats or mice accompanied by neuronal damage, mainly in limbic structures such as the hippocampal pyramidal neurons in particular (3). KA-induced seizures in rodents have been widely used as a model of human temporal lobe epilepsy.
We have reported previously that the Bcl-2 (B-cell lymphoma gene 2) levels are down-regulated in a KA-induced rat epileptic seizure model (3). The Bcl-2 family proteins are mainly located in the mitochondrial outer membrane and are divided into pro-apoptotic and the anti-apoptotic groups (4 -7). Several mechanisms have been reported to underlie the anti-apoptotic function of Bcl-2, including the regulation of Ca 2ϩ homeostasis or action as an antioxidant. Furthermore, as a heterodimer, the pro-apoptotic protein Bax is attenuated by Bcl-2. Moreover, Bcl-2 prevents the release of cytochrome c from mitochondria and inhibits the activation of caspase-9 and caspase-3 (8).
It has been well established that the Bcl-2 protein levels are essential for its anti-apoptotic function. The regulation of these levels mainly occurs via post-translational modifications and degradation (9 -14). More recently, protein S-nitrosylation has emerged as the principal post-translational modification by which nitric oxide exerts a myriad of biological effects. S-Nitrosylation, the covalent attachment of a NO group to a Cys thiol side chain, has been postulated to be a fundamental mechanism in cellular signal transduction. The stability of proteins, cleavage of zymogens, and modification of active proteins can be controlled by this post-translational modification process (15,16). More recently, nitrosylation has been found to be reversible. Nitrosylated Cys thiol groups can be reduced to free thiols, depending on the conditions, which is defined as denitrosylation and plays an important role in signal regulation in a broad range of diseases (17). Mutational analysis of Bcl-2 has shown that two cysteine residues in this * This work was supported by Project Grant 30870543 from the National Natural Science Foundation of China and Education Departmental Nature Science Funds of Jiangsu Province of China (09KJB310015). 1 Both authors contributed equally to this work. 2  protein (Cys-158 and Cys-229) are important for the S-nitrosylation process (18). Bcl-2 degradation is mainly mediated through ubiquitin-proteasome pathway. All four lysine residues in Bcl-2 (K17R, K22R, K218R, and K239R) target this protein for ubiquitin-dependent degradation (19). A more recent report has indicated that the S-nitrosylation of Bcl-2 prevents its ubiquitin-proteasomal degradation during the apoptotic cell death induced by chromium (VI) in lung cancers (18). Similarly, it has been shown that the mechanism of cisplatin resistance involves the up-regulation of Bcl-2 expression by NO, which occurs by preventing its ubiquitin-dependent degradation in human lung carcinoma H-460 cells (20).
In this study, we demonstrate that KA induces Bcl-2 denitrosylation through the GluR6-KA receptor pathway, which is proposed to facilitate Bcl-2 ubiquitination and finally downregulate its protein level. The GluR6 antagonist NS102, NO donor S-nitrosoglutathione (GSNO), sodium nitroprusside (SNP), and the proteasome inhibitor MG132 prevent the denitrosylation and the degradation of Bcl-2, playing an important role in neuron protection in hippocampal CA1 and CA3/DG regions.
Seizure Model-Adult male Sprague-Dawley rats weighing 250 Ϯ 10 g were used. Seizures were induced by an intracerebroventricular injection of KA (0.6 g/10 l) dissolved in sterile saline. The animals were monitored behaviorally for seizures for at least 6 h after injection. The seizures were scored as follows: 1) behavioral arrest and staring spells, 2) head bobbing and gnawing, 3) unilateral forelimb clonus, 4) bilateral forelimb clonus, 5) severe seizures with loss of postural control, and 6) seizure-induced death (21). Only animals with stage 4 or 5 seizures were used in the analyses.
Sample Preparation-The rats subjected to KA for 3 or 6 h post-treatment were decapitated immediately, and the hippocampal CA1 or CA3/DG region was isolated from each animal and quickly frozen in liquid nitrogen. Tissues were homogenized in an ice-cold homogenization buffer containing 50 mM MOPS, pH 7.4, 100 mM KCl, 320 mM sucrose, 50 mM NaF, 0.5 mM MgCl 2 , 0.2 mM dithiothreitol, 1 mM EDTA, 1 mM EGTA, 1 mM Na 3 VO 4 , 20 mM sodium pyrophosphate, 20 mM ␤-phosphoglycerol, 1 mM p-nitrophenyl phosphate, 1 mM benzamidine, 1 mM phenylmethylsulfonyl fluoride, and 5 g/ml each of leupeptin, aprotinin, and pepstatin A. The homogenates were centrifuged at 800 ϫ g for 10 min at 4°C. Supernatants were collected, and protein concentrations were determined using the BCA method. The samples were stored at Ϫ80°C until use.
S-Nitrosylation Assay-S-Nitrosylation was detected using the biotin switch method as described by Jaffrey et al. (22) in which the free thiols in S-nitrosylated proteins are blocked and the S-nitrosothiols are reduced by ascorbic acid sodium salt to yield free thiols, which can be covalently linked to biotin derivatives and assayed using a biotin-based analysis. Briefly, the cells are lysed in HEN buffer (250 mM HEPES, pH 7.7, 1 mM EDTA, 0.1 mM neocuproine, 1% Nonidet P-40, 150 mM NaCl, 1 mM PMSF, protease inhibitor mixture), and the resulting lysates are mixed with an equal volume of MMTS buffer (25 mM HEPES, pH 7.7, 0.1 mM EDTA, 10 M neocuproine, 5% SDS, 20 mM MMTS) and incubated at 50°C for 20 min with frequent vortexing. After the free MMTS is removed by cold acetone precipitation, the precipitates are resuspended in HENS buffer (25 mM HEPES, pH 7.7, 0.1 mM EDTA, 10 M neocuproine, 1% SDS). After the addition of 2 volumes of neutralization buffer (20 mM HEPES, pH 7.7, 1 mM EDTA, 100 mM NaCl, 0.5% Triton X-100), the samples are then modified with biotin in the following buffer (25 mM HEPES, pH 7.7, 0.1 mM EDTA, 1% SDS, 10 M neocuproine, 10 mM ascorbic acid sodium salt, and 0.2 mM biotin-HPDP). After free biotin-HPDP was removed by cold acetone precipitation, biotinylated proteins are absorbed to streptavidin-agarose. The streptavidin absorbates are then eluted by ␤-mercaptoethanol (100 mM), separated by SDS-PAGE, and immunoblotted with an anti-Bcl-2 antibody.
Immunoprecipitation and Western Blotting Analysis-For immunoprecipitation, cytosolic fractions (each containing 400 g of proteins) were diluted 4-fold with HEPES buffer containing 50 mM HEPES, pH 7.4, 150 mM NaCl, 10% glycerol, 1% Triton X-100, and 1 mM each of EGTA, EDTA, PMSF, and Na 3 VO 4 . The samples were then preincubated for 1 h with 20 l of protein A-Sepharose CL-4B (Amersham Biosciences) and centrifuged to remove any nonspecifically adhered proteins from the protein A-Sepharose. The supernatant was then incubated with 2-5 g of specific antibodies for 4 h at 4°C. After the addition of protein A-Sepharose, the mixture was incubated at 4°C for an additional 2 h. The samples were triple washed with HEPES buffer and eluted by SDS-PAGE loading buffer then boiled at 100°C for 5 min.
Western blot analysis was carried out following 12.5% SDS-PAGE. Briefly, the proteins were electrotransferred onto PVDF filters (pore size, 0.45 m). After blocking for 3 h in Tris-buffered saline with 0.1% Tween 20 (TBST) and 3% bovine serum albumin, the membranes were incubated overnight at 4°C with primary antibodies in TBST containing 3% bovine serum albumin. The membranes were then washed and incubated with horseradish peroxidase-conjugated secondary antibodies in TBST for 30 min and visualized by the enhanced chemiluminescence detection system (BeyoECL Plus). The densities of the bands were subsequently scanned and analyzed with an image analyzer (LabWorks Software; UVP, Upland, CA).
Proteasome Activity Assay-20 S proteasome activity was measured using an assay kit from Chemicon (Temecula, CA), in accordance with the manufacturer's protocol but with some modifications. Briefly, tissues were homogenized in icecold extract buffer (50 mM Tris-HCl, pH 7.6, 1 mM EDTA, 10% glycerol, 1 mM EGTA) and then exposed to ultrasound, followed by centrifugation at 800 ϫ g for 15 min at 4°C. The supernatants were collected and determined for protein content using the BCA method. 100 g of proteins were incubated with 5 M proteasome substrate LLVY-AMC in 1 ml of assay buffer at 37°C for 2 h. The AMC fluorophore obtained after cleavage from the labeled substrate was quantified with Hitachi fluorescence spectrophotometer F-7000 at excitation and emission wavelengths of 335 and 460 nm, respectively (23).
Immunohistochemistry-The rats were anesthetized with chloral hydrate and then subjected to transcardial perfusions with 0.9% saline followed by 4% paraformaldehyde in 0.1 M PBS. The brains were then removed, postfixed overnight in paraformaldehyde, processed, and embedded in paraffin. Coronal brain sections (6 m thick) were cut on a microtome (RM2155; Leica, Nussloch, Germany). The sections were deparaffinized in xylene and rehydrated in a gradient of ethanol and distilled water. High temperature antigen retrieval was then performed in 1 mM citrate buffer. To block endogenous peroxidase activity, the sections were incubated for 6 min in a solution of 0.1% H 2 O 2 in PBS. To reduce nonspecific staining, the sections were incubated for 1 h in a blocking solution containing 1% BSA, 2% normal goat serum, 0.3% Triton X-100, and 5% nonfat dry milk in PBS. The sections were then incubated with a rabbit polyclonal antibody against Bcl-2 (1: 50) and 0.3% Triton X-100 overnight at 4°C. After washing three times in PBS, the sections were incubated for 2 h in biotinylated goat anti-rabbit secondary antibody (1:200) made up in 0.1% BSA, 0.3% Triton X-100, and 1% normal goat serum in PBS. The sections were washed and incubated with avidinconjugated horseradish peroxidase for 1 h at 37°C. To visualize bound antibodies, the sections were incubated with a 3,3Јdiaminobenzidine peroxidase substrate kit and examined under a light microscope.
Histology-For histological analyses, rats subjected to KA post-treatment for 7 days were perfusion-fixed with 4% paraformaldehyde in 0.1 M phosphate buffer (pH 7.4) under anesthesia. Paraffin-embedded brain sections (6 m) were then prepared and stained with 0.1% (w/v) cresyl violet to assess neuronal damage in the hippocampus. The number of surviving hippocampal CA1 or CA3 pyramidal cells/mm was counted as the neuronal density.
Cell Culture-The human neuroblastoma cell SH-SY5Y was cultured in DMEM containing 10% fetal bovine serum at 37°C in humidified 8% CO 2 atmosphere. For transfection experiments, the cells were seeded onto 6-or 24-well plates. Twenty-four h after inoculation, either Bcl-2-targeted or scrambled siRNAs (nonsilencing control) were transfected into the cells, which were at 40 -50% confluence. The stock siRNA was diluted in reduced serum Opti-MEM to form complexes with Lipofectamine 2000 at a 1:2 ratio (3 g of siRNA formulated with 6 l of Lipofectamine 2000/well for 6-well plates; 0.75 g of siRNA formulated with 1.5 l of Lipofectamine 2000/well for 24-well plates). The mixtures were then incubated at room temperature for 20 min before transfection in a final volume of 2 ml/well in 6-well plates and 500 l/well in 24-well plates. The final concentration of siRNA in the medium was 125 nM. The cells were cultured in Opti-MEM medium for 6 h without antibiotics during the transfection period. Then cells were cultured in DMEM containing 10% fetal bovine serum for 18 h and stimulated with 500 ⌴ KA for an additional 24 h. Apoptotic levels after treatments were determined by DAPI in 24-well plates. Total cell homogenates were extracted from 6-well plates for total protein quantification before Western blot analysis. Total RNA was also extracted from 6-well plates by TRIzol for RT-PCR analysis of Bcl-2.
TUNEL Staining-TUNEL staining was performed using an ApopTag peroxidase in situ apoptosis detection kit according to the manufacturer's protocol with minor modifications. Briefly, paraffin-embedded coronal sections were deparaffinized, rehydrated, and then treated with protease K at 20 mg/ml for 15 min at room temperature. The sections were then incubated with reaction buffer containing TdT enzyme and at 37°C for 1 h. After washing with stop/ wash buffer, the sections were further treated with antidigoxygenin conjugate for 30 min at room temperature, and the color reaction was subsequently developed in peroxidase substrate. The nuclei were lightly counterstained with 0.5% methyl green.
Data Analysis and Statistics-The values are expressed as the means Ϯ S.D. The statistical analysis of the results was carried out using the Student's t test or one-way analysis of the variance followed by the Duncan's new multiple range method or Newman-Keul's test. p values Ͻ0.05 were considered significant.

RESULTS
The Degradation of Bcl-2 in the Hippocampal CA1 and CA3/DG Regions Is Induced by KA through the Activation of GluR6-containing Kainate Receptors-We previously reported that KA treatment through intracerebroventricular infusion or intraperitoneal injection down-regulates the Bcl-2 protein level gradually in hippocampal CA1 and CA3/DG regions in a concentration-dependent manner (3,24). In our current experiments, we focused on both the KA concentration and time course. As shown in Fig. 1 (A and B), the treatment of rats with KA caused a dose-dependent decrease in Bcl-2, with the lowest level reached at a dose of 0.6 g/10 l. We also observed a time-dependent decrease, with this effect clearly noticeable at 6 h post-treatment and thereafter.
KA activates not only KA receptors but also AMPA receptors and even NMDA receptors at some higher doses. We therefore pretreated the rats with the AMPA receptor antagonist GYKI53655, NMDA receptor antagonist MK801 and GluR6 antagonist NS102 before administering a 0.6 g/10 l KA injection to investigate whether any of the corresponding receptors is involved in the degradation of Bcl-2 (25,26). As shown in Fig. 1C, only pretreatment with NS102 could prevent the degradation of Bcl-2, indicating that KA at the dosage selected exerts its effects mainly through GluR6-KA receptors.
To confirm that the down-regulation of Bcl-2 by KA was at the post-translational level, we examined the Bcl-2 mRNA levels in the rat hippocampal CA1 and CA3/DG regions at 6 h post-treatment. As shown in Fig. 1D, KA has no affect on the Bcl-2 transcript levels that were equivalent to the saline control group.
GluR6-KA Receptors Mediate Bcl-2 Denitrosylation-It has been reported that Bcl-2 is denitrosylated in some circumstances (18). A few studies have also recently indicated that protein denitrosylation can be induced by membrane receptors (17). To test this potential function of KA receptors in our present study, we examined the Bcl-2 S-nitrosylation level using the biotin switch method at 3 h post-treatment when Bcl-2 degradation is not clearly evident. We also pretreated the animals with NS102 to block the activation of GluR6-KA receptors. As shown in Fig. 2A, KA infusions induce the denitrosylation of Bcl-2, which is prevented by NS102, suggesting that the Bcl-2 denitrosylation is indeed mediated by GluR6-KA receptors. We next administrated the exogenous NO donors SNP and GSNO by intracerebroventricular infusion prior to KA injection to suppress the denitrosylation of Bcl-2. As shown in Fig. 2B, both GSNO and SNP were found to be effective Bcl-2 NO donors under our experimental conditions, and Bcl-2 denitrosylation was thus rescued by pretreatment with these agents. This rescue effect was negated by co-administration with the reducing agent DTT.
The dephosphorylation of Bcl-2 Ser-87 after the stimulation of human endothelial cells with TNF-␣ has been shown to promote Bcl-2 ubiquitin-dependent degradation (19). It has also been elucidated that the down-regulation Bcl-2 Ser-87 phosphorylation by reactive oxygen species is critical for this ubiquitin-dependent degradation in manganese-superoxide dismutase antisense-transfected squamous cell carcinoma cell line OSC-4 (27). Our previous study also showed that the administration of KA could increase the production of reactive oxygen species (24). Consequently, in our present analyses we examined whether the Bcl-2 Ser-87 phosphorylation levels are influenced by KA and various NO modulators. The rats were treated with KA in the presence or absence of GSNO, SNP, DTT, and the antioxidant NAC to attenuate reactive oxygen species induction. The effects of these compounds on Bcl-2 phosphorylation were then determined by Western blot using a phospho-specific Bcl-2 (Ser-87) antibody. As shown in Fig. 2C, KA had minimal effects on Bcl-2 phosphorylation at 3 h post-treatment compared with the saline group or the groups pretreated with SNP, GSNO, DTT, or NAC. These results support the notion that KA induced Bcl-2 degradation is independent of Bcl-2 Ser-87 phosphorylation.
KA Mediates Bcl-2 Ubiquitin-dependent Degradation-To next investigate whether Bcl-2 degradation is ubiquitin-proteasome dependent, we pretreated rats with the proteasome inhibitor MG132 at 25 g/10 l by intracerebroventricular infusion at 30 min before KA injection. As shown in Fig. 3A, the degradation of Bcl-2 was completely rescued by MG132, implicating the ubiquitin-proteasome system in Bcl-2 degradation. Furthermore, immunoprecipitation with anti-ubiqui-tin antibody was performed to examine the Bcl-2 ubiquitin complex at 3 h post-treatment. The results showed that the up-regulation of Bcl-2 ubiquitination was promoted by KA but impaired by pretreatment with SNP or GSNO (Fig. 3B). DTT co-treatment negated the effects of these NO donors. These data suggest that the up-regulation of Bcl-2 nitrosylation impairs its ubiquitination and vice versa.
Further analyses were performed to examine the Bcl-2 levels in SNP-or GSNO-pretreated groups. As shown in Fig. 3C, both SNP and GSNO prevent the Bcl-2 degradation induced by KA. We have previously observed that caspases are activated after KA infusion, and given the finding that caspases cleave Bcl-2 in some cell lines (29), we pretreated the KA group with the caspase inhibitor z-VAD-FMK (150 ng/10 l) (30). As shown in Fig. 3D, however, z-VAD-FMK did not attenuate the Bcl-2 degradation caused by KA, indicating that caspases are not involved in this pathway. To next investigate whether treatment with these compounds alone (i.e. in the absence of KA) had any effects on Bcl-2 S-nitrosylation and stability, the rats were treated with NS102, GSNO, SNP, or MG132, respectively. The Bcl-2 S-nitrosylation and relative expression levels were then determined. As shown in Fig. 3E, treatments with these compounds alone had no effect on Bcl-2 S-nitrosylation or stability (p Ͼ 0.05). The mechanism of Bcl-2 degradation by exposure to KA is therefore a stress response rather than a general Bcl-2 degradation pathway involved in homeostasis.
Immunohistochemical analysis further revealed the neuroprotective effects of NS102, GSNO, SNP, and MG132 upon Bcl-2 degradation at 6 h post-treatment. In the KA group (Fig.  3F, panels d-f), Bcl-2 immunoreactivity was barely detectable in the CA1 and CA3 pyramidal neurons but was significantly visible in the saline group (Fig. 3F, panels a-c). The protective effects of NS102, GSNO, SNP, and MG132 were also clearly evident from the stronger immunostaining (Fig. 3F, panels  g-r), in contrast to the KA group (Fig. 3F). Our data thus suggest that NS102, GSNO, SNP, and MG132 cause an increase in the levels of Bcl-2 by blocking the degradation induced by KA in hippocampal CA1 and CA3 pyramidal neurons.
GluR6-KA Receptor Coupled Bcl-2 Denitrosylation Does Not Require Fas Activation-We have reported previously that FasL is up-regulated and that more Fas receptors are activated by JNK in a KA-induced rat epileptic seizure model (3). Moreover, it has been reported that the activation of Fas receptors gives rise to procaspase-3 denitrosylation, which facilitates cleavage to its active form (i.e. activated caspase-3) (31). To investigate whether the Bcl-2 denitrosylation triggered by KA injection is mediated by Fas activation, we administrated FasL antisense oligodeoxynucleotides to the rats prior to KA treatment. As shown in Fig. 4A, however, inhibition of the Fas pathway did not ameliorate the degradation of Bcl-2. This result indicates that the Fas pathway does not play a role in GluR6-KA receptor-coupled Bcl-2 denitrosylation.
It has been reported that the E3 ubiquitin ligase parkin is nitrosylated under conditions of NOS stress, which has a strong effect upon the ubiquitin-proteasome system (first promotes and then reduces its activity) (32). To test the possibility that ubiquitin-proteasome system activity affects Bcl-2 degradation, hippocampal homogenates were analyzed for proteasome activity by spectrofluorometry. As shown in Fig.  4B, however, KA and NO modulators did not interfere with proteasome activity.
Exogenous NO Donors and Proteasome Inhibitors Prevent the Activation of Caspase-3-A previous study has demonstrated that KA infusion down-regulates Bcl-2, which releases mitochondrial cytochrome c into the cytosol where it activates caspase-3 and induces neuronal apoptosis (3). In our present study, we have demonstrated that exogenous NO donors and the proteasome inhibitor MG132 can attenuate the Bcl-2 degradation induced by KA. To further investigate the possible neuroprotective effects of these agents, rats were pretreated with NO donor GSNO and MG132 by cerebroventricular injection 30 min prior to KA administration. The expression of cytochrome c in both the mitochondria and the corresponding cytosol was then examined by Western blotting. In addition, the activation of caspase-3 was determined by analyzing the procaspase-3 (32 kDa) and cleaved caspase-3 (17/19 kDa) levels. The results shown in Fig. 5 reveal that both GSNO and MG132 prevent the translocation of cytochrome c and the activation of caspase-3.   ϫ40 in panels a, d, g, j, m, and p; magnification was ϫ400 in panels b, c, e, f, h, i, k, l, n, o, q, and r. Scale bars, 200 m (panel p) and 10  m (panels q and r). The data are expressed as the means Ϯ S.D. from three independent animals (n ϭ 3). *, p Ͻ 0.05 versus the saline group; #, p Ͻ 0.05 versus the KA group or the respective DTT-pretreated group (n ϭ 3).

Knockdown of Bcl-2 in SH-SY5Y Cells by siRNA and Antisense ODNs in Hippocampi Promotes Cellular Apoptosis-Ex-
periments were performed to confirm whether the stability of Bcl-2 is causal to KA resistance in both cultured human neuroblastoma SH-SY5Y cells and hippocampal neurons in vivo. We suppressed Bcl-2 expression in SH-SY5Y cells using an RNA interference approach. Transfection of a scrambled siRNA control resulted in no gross morphological alterations under light microscopy (Fig. 6, A and B), and no toxicity was observed. Fluorescently labeled siRNA was successfully transfected, whereas no positive signal was evident in the nontransfection control group (Fig. 6, AЈ and BЈ). Several siRNA sequences that were designed to target to Bcl-2 were next screened for their efficacy. We found that siRNA-3 and -4 effectively suppressed the expression of Bcl-2 (Fig. 6, C and  D), whereas the siRNA-1 and -2 molecules had no effect (Fig.   6, C and D). After siRNA-4 transfection into cultured SH-SY5Y cells, cellular apoptosis was induced by KA stimulation (Fig. 6E). The effects of enhanced Bcl-2 stability was further confirmed in hippocampal neurons in vivo, in which the cell density of CA1 and CA3 pyramidal neurons was more reduced by Bcl-2 AS-ODNs administration in the presence of KA (Fig. 6G). Similar to the effects of siRNA-4, the downregulation of Bcl-2 expression by Bcl-2 AS-ODNs also promoted the effects of KA on neuronal survival. Hence, Bcl-2 is vital to neuronal survival in our rat seizure model. GSNO Protects Neurons Mainly by Enhancing Bcl-2 Stability-We investigated whether NO donors could upregulate the S-nitrosylation of Bcl-2 and maintain its stability and also whether the down-regulation of Bcl-2 could promote KA-induced cellular apoptosis. To examine the role of NO donors in neuronal survival, in one test, we pretreated SH- SY5Y cells with 100 M GSNO at 30 min before KA stimulation. In another test, we performed siRNA knockdown of Bcl-2 before pretreatment with GSNO followed by KA treatment. According to the DAPI staining pattern, GSNO was found to protect neurons from KA-induced neuronal apoptosis (Fig. 6E, panels b and d). Significantly also, the down-regulation of Bcl-2 by siRNA attenuated the protective effects of GSNO (Fig. 6E, panels d and e), indicating that Bcl-2 is a key GSNO-targeted protein that protects neurons from KA stimulation. A similar experiment was performed in an animal seizure model system (Fig. 6G). Cresyl violet staining also indicated the neuroprotective effects of GSNO against KA-induced brain injury in rat hippocampal CA1 and CA3 regions, which was attenuated following Bcl-2 down-regulation by AS-ODNs. Taken together, these observations reveal that Bcl-2 stability is causal to KA resistance and underpins the mechanism by which NO functions.
The Protective Effects of Some Drugs against the Delayed Injury Induced by KA in Hippocampal CA1 and CA3 Pyramidal Neurons Are Exerted through the Prevention of Bcl-2 Degradation-To investigate whether NS102, GSNO, SNP, and MG132 play protective roles against KA-induced neuronal cell death, cresyl violet staining was performed to examine the survival of pyramidal neurons of the rat hippocampal CA1 and CA3 regions. Photomicrographs of cresyl violet-stained brain sections from the rats revealed the state of the neuronal cells. Normal neurons in the pyramidal layer of the hippocampi, as shown in the saline operation group, appear rounded with pale stained nuclei, whereas shrunken cells with pyknotic nuclei were considered dead. Samples from rats pretreated with NS102 (Fig. 7A, panels g-i), GSNO (Fig. 7A, panels j-l), SNP (Fig. 7A, panels m-o), or MG132 (Fig. 7A, panels p-r) were significantly protected compared with those administered with KA only. The numbers of viable cells were counted within a 1-mm length in the KA alone and saline-, NS102-, GSNO-, SNP-, and MG132-treated groups. TUNEL staining was also used to examine the level of apoptosis in CA1 and CA3 pyramidal cells in the hippocampus. As shown in Fig. 7B (panels d-f), a significant number of TUNEL-positive cells was observed on the seventh day post-treatment, with some of them showing characteristic appearances such as shrunken condensed nuclei and apoptotic bodies. Pretreatment with NS102, GSNO, SNP, and MG132 significantly reduced the number of TUNEL-positive cells (Fig. 7B, panels g-r). The numbers of viable TUNEL-positive cells were counted within a 1-mm length in the saline, KA, NS102, GSNO, SNP, and MG132 treatment groups. The above results suggest that NS102, GSNO, SNP, and MG132 protect rat hippocampal CA1 and CA3 pyramidal cells from apoptosis induced by KA infusion.

DISCUSSION
In our present report, we demonstrate that GluR6-KA receptor-mediated Bcl-2 denitrosylation facilitates Bcl-2 ubiquitination and proteasomal degradation during KA-induced neuronal apoptosis. Pretreatment with the GluR6 antagonist NS102, the NO donors GSNO and SNP, as well as the proteasome inhibitor MG132 exerted neuroprotective effects by attenuating Bcl-2 denitrosylation and ubiquitin-dependent degradation.
S-Nitrosylation and denitrosylation are crucial protein post-translation modifications that involve the regulation of protein function, such as phosphorylation and dephosphorylation, and that play important roles in the activity regulation of protein kinases under both physiological and pathological FIGURE 5. Effects of the exogenous NO donor GSNO and the proteasome inhibitor MG132 on the release of cytochrome c and the activation of caspase-3 induced by KA. A, effects of the exogenous NO donor GSNO and proteasome inhibitor MG132 on the redistribution of cytochrome c (Cytc) in the cytosol and mitochondria at 6 h post-treatment with KA (0.6 g/10 l) in rat hippocampal CA1 and CA3/DG regions. Cytochrome c oxidase subunit IV (COX 4) was found to be strongly expressed in the mitochondria fraction and did not decrease after KA treatment, but virtually no immunoreactivity was seen in the cytosolic fraction in any of the groups. B, effects of GSNO and MG132 on the activation of caspase-3 induced by KA. The expression of procaspase-3 (32 kDa) and its larger cleaved product (17/19 kDa) caspase-3 at 6 h post-treatment was examined by immunoblotting. The corresponding bands were scanned, and the optical density was determined as the fold change versus the saline control. The data are expressed as the means Ϯ S.D. from three independent animals. *, p Ͻ 0.05 versus saline control; #, p Ͻ 0.05 versus KA (6 h) group. conditions. S-Nitrosylation can positively or negatively regulate the function of proteins, and denitrosylation has shown this same capacity (16). For example, the pro-apoptosis proteinase matrix metalloproteinase 9 is activated by S-nitrosylation and induces neuronal apoptosis (33). Our previous study showed that JNK3, a pro-apoptosis kinase, is activated via S-nitrosylation during cerebral ischemia and reperfusion in the rat hippocampus, whereas the anti-apoptosis kinase, Akt/ PKB, can be inactivated by S-nitrosylation (34,35). As for denitrosylation, when this occurs for caspase-3, a pro-apoptosis proteinase, caspase activity is inhibited, whereas Bcl-2 is hydrolyzed by this modification (17,18,31,36). In accordance with previous reports and our present analyses, it is clear that Bcl-2 is also regulated by S-nitrosylation and denitrosylation. Bcl-2 is S-nitrosylated to a certain extent in its basal state, maintaining its stability. On this basis, the regulation or dysregulation of Bcl-2 S-nitrosylation could be employed in diverse cellular responses. For example, on the one hand the effects of stress inducers, including Cr (VI), FasL, and BSO, up-regulate Bcl-2 S-nitrosylation (18). On the other hand, the dysregulation of Bcl-2 S-nitrosylation impairs the anti-apoptotic function of cells and reduces their resistance to KAinduced neuronal apoptosis and cisplatin-induced cell death in the case of human lung carcinoma H-460 cells (20).
Degradation mechanisms also play important roles in the regulation of protein levels, cell apoptosis, and survival and signaling pathways. Among the different protein degradation systems, the ubiquitin-dependent proteasomal degradation pathway is the most well studied (37). The ubiquitination system functions in a wide variety of cellular processes, including antigen processing, apoptosis, the cell cycle, DNA repair, immune responses, the modulation of cell surface receptors, the response to stress, and extracellular modulators. Our findings here indicate that in response to KA stress, Bcl-2 denitrosylation initiates the ubiquitin-dependent degradation of Bcl-2, giving rise to the apoptosis of hippocampal neurons. When Bcl-2 expression was decreased by AS-ODNs or siRNA, the neuroprotection of exogenous NO donor GSNO was weak-  6). B, representative photomicrographs of TUNEL stained hippocampal samples counterstained with methyl green are shown. The rats were injected with saline (panels a-c), KA (panels d-f), or pretreated with NS102 (panels g-i), GSNO (panels j-l), SNP (panels m-o), or MG132 (panels p-r). The results were obtained from six independent animals in each experimental group, and the results of a typical experiment are shown. The boxed areas in the left column are shown at higher magnification in the right columns. The original magnifications were: ϫ40 in panels a, d, g, j, m, and p; and ϫ400 in panels b, c, e, f, h, i, k, l, n, o, q, and r. Scale bars, left column, 200 m; right column, 10 m. The viable TUNEL-positive cells were counted within a 1-mm area. *, p Ͻ 0.05 versus the saline group; #, p Ͻ 0.05 versus the KA injection group (n ϭ 6). ened (as shown in Fig. 6). As is well established, the ubiquitinproteasome system recycles proteins with a short half-life. It would be interesting to measure whether the ubiquitin-proteasome system controls the basal Bcl-2 levels in homeostasis. However, based on our present results (Fig. 3E), the Bcl-2 levels should therefore be noticeably increased at the indicated time points in the presence of MG132. Although our present data do not exclude the possibility of this regulatory mechanism, they cannot confirm that the ubiquitin-proteasome system is a general regulator of Bcl-2 degradation or whether another pathway performs this function.
Although numerous theories have sought to explain the mechanism of protein S-nitrosylation, it remains largely unknown. Recently, two specific enzymatic systems of protein denitrosylation have been identified: the thioredoxin (Trx) system, which comprises Trx proteins, Trx reductase (TrxR) proteins, and NADPH; and the GSNO reductase system, which comprises GSH and GSNO reductase. Theoretically, the denitrosylation of target proteins is mediated either by catalyzing the conversion of reduced Trx-(SH) 2 to oxidized Trx-S 2 or by oxidizing glutathione (GSH). Moreover, candidate denitrosylases have been proposed, such as protein disulfide isomerase, xanthine oxidase, superoxide dismutase, glutathione peroxidase, and carbonyl reductase (17). Additionally, receptor-coupled denitrosylation mechanisms have been identified that are mediated by Fas receptors, TNF␣ receptors, VEGF receptors, insulin receptors, and ␤-adrenergic receptors. In addition, a rise in intracellular calcium has been proven to play a crucial role in protein denitrosylation (28). However, more studies are required to elucidate the mechanisms underlying GluR6-KA signaling. It will also be of great interest to determine the denitrosylase involved in Bcl-2 degradation and any other regulators that function in this process, such as calcium.
In summary, we propose from our findings that the intracerebroventricular infusion of KA results in ubiquitin-dependent Bcl-2 degradation, which is facilitated by GluR6-KA receptor coupled Bcl-2 denitrosylation and is independent of its phosphorylation. Moreover, NS102, GSNO, SNP, and MG132 exert neuroprotective effects by preventing Bcl-2 degradation through the suppression of denitrosylation or ubiquitination. Further studies are needed to elucidate the mechanisms underpinning GluR6-KA receptor coupled denitrosylation, which may lead to better therapeutic approaches to the future treatment of epileptic seizures.