Materials and Methods

Drugs and Other Materials.

PCP was obtained from the National Institute on Drug Abuse (Rockville, MD). PCP was dissolved in Dulbecco's modified Eagle's medium (DMEM). NMDA was purchased from Tocris Neuramin (Bristol, UK). MTT [3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide] was purchased from Sigma (St. Louis, MO). The medium and fetal bovine serum were purchased from Life Technologies (Grand Island, NY). All other kits and enzymes were obtained from Boehringer Mannheim (Indianapolis, IN). Other chemicals were obtained from Sigma.

Primary Cell Culture.

Primary forebrain cultures were prepared from newborn rats (Sprague-Dawley) as described by Kiss et al. (1994). Briefly, forebrains were dissected from brains and dissociated in cold Hanks solution without Mg²⁺ and Ca²⁺. Cultures were grown on polylysine-coated coverslips in DMEM supplemented with 10% (v/v) fetal bovine serum. Glial proliferation was stopped with a mitotic inhibitor, cytosine β-d-arabinofuranoside (beginning on the third day of culture). Cells were treated with 3 μM PCP or normal medium at 37°C for 48 h (days 4–6). After the treatment, cultures were rinsed with serum-free defined medium. Cultures were then exposed to NMDA (30 or 100 μM, or control) on day 6 in serum-free defined medium (prepared by adding a supplement mixture consisting of 15 μg/ml insulin, 20 μg/ml transferrin, 20 nM progesterone, 100 μM putrescine, and 30 nM sodium selenite to DMEM). Neurotoxicity was evaluated in four different assays 20 h after the addition of NMDA in Mg²⁺-containing medium.

Cytotoxicity Detection Assay.

The release of the cytosolic enzyme lactate dehydrogenase (LDH) into the medium was used as a generic index of cell death. Twenty hours after exposure to NMDA or control medium, the medium was collected and assayed for LDH activity with a cytotoxicity detection kit from Boehringer Mannheim. Briefly, LDH catalyzes the conversion of lactate to pyruvate on reduction of NAD⁺ to NADH/H⁺; the added tetrazolium salt (yellow) is then reduced to formazan (red). The amount of formazan formed correlates to LDH activity. The formazan product is measured with a microtiter plate reader at an absorption wavelength of 490 nm.

MTT Reduction Cell Viability Assay.

The dye MTT is taken up and metabolized to a colored product by viable mitochondria. Thus, the measurement of this metabolic reduction reaction was used as a marker of mitochondrial viability. Briefly, 100 μl of MTT (5 mg/10 ml of medium) was added to each well, and the plate was incubated for 4 h at 37°C. The MTT solution was removed, 100 μl of dimethyl sulfoxide was added to each well, and the color intensity was assessed with an enzyme-linked immunosorbent assay (ELISA) plate reader at a wavelength of 590 nm.

Fragmented DNA Detection by ELISA.

Although the LDH release assay and the MTT reduction assay are reliable indices of cell death, neither is specific for apoptotic cells. Such cells are better characterized by DNA fragmentation that is the result of internucleosomal cleavage of DNA by apoptosis-specific activation of endonucleases. The presence of fragmented DNA associated with nucleosomal histone in chronic PCP-treated and control cells was assessed by a specific two-site ELISA using an antihistone primary antibody and a secondary anti-DNA antibody according to the manufacturer's instructions (Boehringer Mannheim). Briefly, cells were grown in 10-cm tissue culture dishes. After the treatment regimen, cells were spun and resuspended in 3 ml of lysis buffer and incubated for 30 min at room temperature. After centrifugation, the supernatants (cytosol containing low-molecular mass, fragmented DNA) were diluted 1:2 (v/v) with lysis buffer. Then, 20 μl from each sample was transferred to a plate reader well precoated with antihistone antibody, and 80 μl of immunoreagent mix, including the secondary antibody, was added. After incubation and washes, the wells were treated with the chromogen substrate, and the intensity of the color that developed was assayed at 405/490 nm wavelength.

Terminal dUTP Nick-End Labeling (TUNEL) Assay.

This assay is widely used to assess apoptosis in situ. It relies on the detection of fragmented DNA strands, but because fragmentation can occur via nonapoptotic mechanisms, it is not absolutely specific for apoptosis. After the treatment regimen, the cells were rinsed with PBS; fixed by ice-cold (4°C) 4% paraformaldehyde in 0.1 M phosphate buffer, pH 7.2; and processed for evaluation of nuclei containing fragmented DNA in situ. Terminal deoxynucleotidyl transferase, a template-independent polymerase, was used to incorporate biotinylated nucleotides at sites of DNA breaks as previously described (Johnson et al., 1998). The cells were photographed with the use of an Olympus light microscope. The percentage of TUNEL-positive cells was estimated in five 0.24-mm² fields in each of three dishes in each treatment condition. Each condition was assessed at least in triplicate and experiments were repeated three times independently. Data are presented as the mean ± S.E. A probability of P< .05 was considered significant (one-way ANOVA).

Light-Microscopic Immunocytochemistry and Nuclear Staining.

This experiment evaluated the nuclear morphology of PCP-treated and control cells, with Hoechst 33258 to visualize the nucleus and polysialic acid-neuronal cell adhesion molecule (PSA-NCAM) immunocytochemistry to mark neurons. A mouse monoclonal antibody (Meningococcus group B) to PSA-NCAM was used (1:500 dilution). After the treatment regimen, the cells were rinsed with PBS; fixed by ice-cold (4°C) 4% paraformaldehyde in 0.1 M phosphate buffer, pH 7.2; and the indirect immunofluorescence technique was used to visualize immunoreactivities. The cells were incubated with the primary antibody (diluted in PBS/0.5% BSA solution) at 4°C overnight. Bound antibodies were revealed with fluorescein isothiocyanate-conjugated sheep anti-mouse IgG (Boehringer Mannheim) secondary antibody (diluted 1:80 in PBS/0.5% BSA solution). The cells were examined with an Olympus light microscope equipped with epifluorescence.

To assess nuclear morphology, cultured cells were stained with bisbenzimide solution (Hoechst 33258; Sigma). Bisbenzimide (0.1 μg/ml) was dissolved in PBS/glycerol (1:1) solution. After putting two drops of bisbenzimide solution on microscope slide, the coverslips with cells were mounted onto the slides and observed under a fluorescence microscope with an excitation wavelength of 365 nm.

Western Blot Analysis.

After the treatment regimen, the medium was first removed and the attached cells were washed with PBS. Protein extraction was accomplished by cell lysis with SDS. Protein samples were measured for protein concentration with BCA Protein Reagent (Pierce, Rockford, IL). Equal amounts of total protein (10 μg) were loaded on each lane and run on SDS-polyacrylamide gels with a Tris/glycine running buffer system and then transferred to a polyvinylidene difluoride membrane (0.2 μm) in a Mini Electrotransfer Unit (Bio-Rad, Richmond, CA). The blots were probed with an anti-Bcl-X_L (1:1000, polyclonal; Santa Cruz Biotechnology, Santa Cruz, CA) antibody, anti-Bax (1:1000, polyclonal; Santa Cruz Biotechnology) antibody, and anti-actin (1:3000, monoclonal, house-keeping protein; Amersham). Immunoblot analysis was performed with horseradish peroxidase-conjugated anti-rabbit and anti-mouse IgG with the enhanced chemiluminescence Western blotting detection reagents (Amersham). The Bcl-X_L/Bax ratio was analyzed by the Lynx 5000 Imagine Analysis System.

In Situ Hybridization of NR1 mRNA.

An oligonucleotide probe complementary to the mRNA encoding the NMDA glutamate receptor NR1 subunit was selected on the basis of cloned cDNA sequences. The sequence of the probe used for in situ hybridization was as follows: 5′-TTCCTCCTCCTCCTCACTGTTCACCTTGAATC-GGCCAAAGGGACT (this corresponds to a region that is constant across all NR1 splice variants, amino acids 566–580). It was 3′-end labeled by incubation with³⁵S-deoxy-ATP (New England Nuclear, Boston, MA) and terminal deoxynucleotidyl transferase (Boehringer Mannheim) to attain specific activities of ∼5–8 × 10⁸cpm/μg. The specificity of the probe has been previously described (Monyer et al., 1992).

In situ hybridization experiments were carried out on day 6, with cells fixed in 4% paraformaldehyde, as described previously (Wang et al., 1999). After an overnight hybridization at 41°C, slides were washed successively in 4×, 1×, and 0.1× standard saline citrate, quickly dehydrated in ethanol (70%), and air dried. Autoradiography was performed with Kodak NTB3 emulsion and the slides were exposed for 3 weeks at 4°C. Analysis of in situ hybridization autoradiographs was done on hematoxylin-eosin-counterstained sections. The negative controls were performed by adding an excess amount (50-fold) of unlabeled probe.

Quantitation of In Situ Autoradiographs.

Images were acquired with a digital microscopy apparatus (Image Tools Software, University of Texas Health Science Center at San Antonio), saved as 640 × 480 × 8-bit grayscale Tif files and sent to a University core image laboratory for analysis. Briefly, the images were smoothed with a 3 × 3 square filter to remove noise, and regions of interest were selected using a threshold technique that segments the image into labeled cells and background. The threshold was determined interactively by the consensus of two trained observers (one was blind to the treatment) and then held constant for the cells in both the PCP group and the control group. The density of silver grains that exceeded the threshold was estimated in five 0.24-mm²fields in each of three dishes treated with either PCP or normal serum-free defined medium. Each condition was assessed at least in triplicate and experiments were repeated three times independently. A monotonic relationship was assumed to exist between measured labeling and the amount of mRNA labeled with the radioactive probe. This technique is similar to that used by several laboratories except that a fixed size rather than a variable size region of interest is used (Rudolf et al., 1996). Data are presented as the mean ± S.E. Statistical differences were determined by Student's ttest. A probability of P < .05 was considered significant.

Results

Morphological Characterization of Primary Cell Cultures and NMDA-Induced Neurotoxicity.

Two randomly selected regions of 1.8 mm² were selected from three culture dishes from independent experiments to assess the phenotype of the forebrain cells used in this study. Approximately 130 cells were counted in each region. The use of immunofluorescent staining of PSA-NCAM, a neuron-specific marker, and glial fibrillary acidic protein (a glial marker) in control rat forebrain culture revealed that 57 ± 3% of the cells in culture were neurons. Figure1A illustrates the neuron-specific staining of cultured cells with PSA-NCAM. Figure 1B shows that all the nuclei of PSA-NCAM-positive neurons were stained with Hoechst 33258 as visualized with a fluorescence microscope. Glial fibrillary acidic protein-positive glia also were positively stained by Hoechst 33258 (data not shown). Neurons remained viable up to 2 weeks in culture, but were used in these experiments 7 days after plating. At day 6, in vitro, cells were exposed to NMDA for 20 h. Figure 1C shows NMDA-induced nuclear condensation and fragmentation, which are hallmarks of apoptosis. We also found that most of the apoptotic cells were neurons because those cells with condensed and fragmented nuclei were costained by PSA-NCAM (Fig. 1C).

• Download figure
• Open in new tab
• Download powerpoint

Figure 1

Effect of NMDA on nuclear morphology of forebrain neurons in primary culture. A, neuron-specific staining of cultured cells with PSA-NCAM as revealed by immunofluorescence of anti-mouse IgG conjugated to fluorescein isothiocyanate (scale bar, 110 μm). B and C, higher magnification of nuclei stained by Hoechst 33285 (with an excitation wavelength of 365 nm) after treatment with control medium (B) or 100 μM NMDA (in the presence of Mg²⁺) for 20 h (C). Closed arrows show fragmented nuclei and the open arrow points to a condensed nucleus, both of which are hallmarks of apoptosis. The faint outline of neurons in B and C is due to the partial overlap in PSA-NCAM fluorescence at 365 nm. Scale bars in B and C, 13.75 μm.

Quantitation of NMDA-Induced Cell Death in Control and PCP-Treated Cultures.

In these experiments, we evaluated two concentrations of NMDA (30 and 100 μM) in control cultures and in cultures pretreated for 48 h with PCP. In control cultures, after washout of PCP, 30 μM NMDA had no significant effect on the percentage of TUNEL-positive cells, but 100 μM NMDA caused a significant 3-fold increase in this measure (Figs. 2 and3). In contrast, in PCP-treated cells, both 30 and 100 μM NMDA had a large and significant effect on the percentage of TUNEL-positive cells (Figs. 2 and 3A). These data suggest that pretreatment with PCP sensitizes these cells to the toxic effects of NMDA. This suggestion was validated by similar results from an experiment that measured the effects of PCP pretreatment on NMDA-induced increases in LDH release. Herein, the effect of both concentrations of NMDA were potentiated significantly by PCP pretreatment (Fig. 3B).

• Download figure
• Open in new tab
• Download powerpoint

Figure 2

Effect of treatment with 3 μM PCP for 2 days on NMDA-induced apoptosis measured with TUNEL staining in primary cultures of forebrain. A and B, in normal medium; C and D, with 30 μM NMDA; and E and F, with 100 μM NMDA for 20 h immediately after PCP washout. Chronic PCP treatment (B, D, F) appeared to increase the number of TUNEL-positive neurons after NMDA treatment (quantitated in Fig. 3). Scale bar, 22.5 μm.

• Download figure
• Open in new tab
• Download powerpoint

Figure 3

Effect of 48-h incubation of forebrain cultures with 3 μM PCP. After incubation, PCP was washed out and the cultures were treated with either control medium or medium containing either 30 or 100 μM NMDA in the presence of physiological magnesium for 20 h. A, effect of PCP pretreatment on NMDA-evoked apoptosis expressed as the percentage of TUNEL-positive cells relative to the total number of cells measured in five randomly selected areas (0.055 mm²) from each culture. B, effect of PCP pretreatment on NMDA-induced LDH release expressed as percentage release of LDH relative to basal values in untreated control cells (defined as 100%). C, effect of chronic PCP treatment on the ability of 30 μM NMDA to reduce MTT uptake. D illustrates the effect of chronic PCP on the ability of NMDA to elicit an increase in fragmented DNA as measured by ELISA. Data are mean ± S.E. of three independent experiments, each performed in triplicate (*P < .05). In all four assays of neurotoxicity, 30 μM NMDA was nontoxic in normal control cultures but was markedly toxic after chronic PCP.

Mitochondrial viability and apoptosis after exposure to 30 μM NMDA were further investigated by measuring MTT uptake and histone-associated, fragmented DNA by ELISA (Fig. 3, C and D). Results from these assays verify that 30 μM NMDA (in the presence of Mg²⁺) did not significantly affect either measure in control. Thus, 30 μM NMDA is below the threshold for toxicity in this system. However, as shown in Fig. 3, the ability of 30 μM NMDA to cause either a decrease in mitochondrial viability or an increase in histone-associated DNA fragments was significantly greater in cultures pretreated with 3 μM PCP (Fig. 3, C and D). In summary, these data provide evidence that PCP pretreatment increases the vulnerability of cultured forebrain neurons to the apoptotic effects of 30 and 100 μM NMDA.

To determine whether the enhanced effect of NMDA could be due to PCP-induced up-regulation of the NMDA receptor, we again treated the forebrain cultures for 48 h (days 4–6) with 3 μM PCP, and then after washing out the PCP, the NR1 subunit mRNA level was measured by in situ hybridization. We used a ³⁵S-labeled oligonucleotide probe specific to the NR1 sequence (Monyer et al., 1992). The NR1 subunit mRNA was found to be present in both control and PCP-pretreated cultured neurons. Compared with cultures treated with control medium, PCP-pretreated cultures exhibited a marked increase of the mRNA for the obligatory NR1 subunit of NMDA receptor. Figure4 compares the density of NMDA receptor NR1 subunit mRNA present in cells treated with PCP and control medium. The signal, autoradiographic silver grains, is much denser in cells treated with PCP (Fig. 4C) than in cells treated with control medium (Fig. 4B). Quantitative analysis revealed that chronic PCP resulted in a 47% increase (P < .05) in mRNA for the NR1 subunit compared with control cultures (Fig. 4A). Although the NR1 subunit protein was not measured in these experiments, it is likely that the increased neuronal cell death after chronic PCP is a consequence of enhanced NMDA receptor function.

• Download figure
• Open in new tab
• Download powerpoint

Figure 4

Effect of chronic PCP treatment on the in situ hybridization signal of NMDAR1 subunit mRNA level in rat forebrain cultures. Cultures were grown on 2.2 × 2.2-cm coverslips in individual Petri dishes for 4 days before adding either control medium or medium containing 3 μM PCP for 2 days. NR1 mRNA was estimated in five 0.24-mm² fields in each of three dishes under each treatment condition. A, quantitation of the NR1 mRNA signal from control and PCP-treated cultures. Data are mean ± S.E. of three independent experiments, each performed in triplicate. Representative autoradiographs from control and PCP-treated cultures are shown in B and C, respectively. *P < .05 (Student'st test).

Role of Bcl-2 Family Proteins and Reactive Oxygen Species (ROS) in NMDA Receptor-Mediated Apoptosis.

To determine whether the expression pattern of key members of Bcl-2 family genes are correlated with the morphological and biochemical results, cell lysates were immunoblotted with rabbit antisera to Bcl-X_L and Bax. Figure 5 shows that the polyclonal anti-Bcl-X_L antibody recognized a single protein band at ∼29 kDa, and the anti-Bax antibody recognized a single protein band at ∼21 kDa. Visual inspection of lanes 1, 2, and 3 reveals that NMDA produced a concentration-dependent increase in Bax and concomitant decrease in Bcl-X_L. A similar effect was observed in PCP-treated cultures (lanes 4, 5, and 6). However, quantitative densitometry revealed that the effect of 30 μM NMDA on both Bax and Bcl- X_L was more marked in PCP-treated cultures than in control (data not shown). These densitometry measurements were used to calculate a ratio of Bcl-X_L to Bax in each lane in three independent experiments and the mean ± S.E. of these ratios is shown in Fig.6. This ratio was markedly decreased by 100 μM NMDA in both control and PCP-treated cultures; however, the effect of 30 μM NMDA was equivalent to 100 μM NMDA after PCP treatment, whereas it had no significant effect in control cultures. This pattern is very similar to that observed when the neurotoxic effect of NMDA was assessed by either LDH release or MTT uptake (Table1).

• Download figure
• Open in new tab
• Download powerpoint

Figure 5

Western blot analysis of Bcl-X_L and Bax expression in primary forebrain culture. Control cells were exposed to 100 (lane 1), 30 (lane 2), and 0 μM NMDA (lane 3). Chronic PCP-treated cells were exposed to 100 (lane 4), 30 NMDA (lane 5 and 7), and 0 μM NMDA (lane 6). Total cellular protein was extracted and analyzed by polyacrylamide gel electrophoresis (10 μg of protein per lane), electrotransferred, and then probed with anti-Bcl-X_L(top) and anti-Bax (bottom). By comparison with molecular mass standards (Santa Cruz Biotechology), a band of ∼29 kDa was identified as Bcl-X_L and another band of ∼21 kDa was identified as Bax according to their labeling by anti-Bcl-X_L and anti-Bax antibodies, respectively. Results shown are representative of three independent experiments.

• Download figure
• Open in new tab
• Download powerpoint

Figure 6

Densitometric measures of the ratio of Bcl-X_L to Bax. Values represent mean ± S.E. of three independent experiments. Statistical comparisons consisted of a one-way ANOVA (*P < .05).

The processes that link excessive activation of the NMDA receptor to eventual cell death are not fully understood. To elucidate the potential role of ROS such as superoxide anion, we exposed chronic PCP-treated cells to 30 μM NMDA in presence of superoxide dismutase (SOD; 30 U/ml of medium) and catalase (50 U/ml of medium). Comparison of lanes 5 and 7 in Fig. 5 shows that SOD/catalase pretreatment prevented the NMDA-induced increase in Bax. Densitometry showed that this treatment also prevented the NMDA-induced decrease in Bcl-X_L. This agrees with the ratiometric measure shown in Fig. 6. A similar effect also was noted with either LDH release or MTT uptake as markers of cell viability (Table1).

Discussion

Several laboratories have reported that a chronic regimen of PCP results in a sensitized locomotor response to PCP challenge, but the mechanism is still largely unknown. This laboratory has demonstrated that such sensitization is associated with apoptotic cell death in the cortex, an increase in NMDA receptor NR1 subunit protein and mRNA, and an increase in NMDA receptor function (Johnson et al., 1998; Hanania et al., 1999; Wang et al., 1999). This study sought to extend these findings to an in vitro model and to begin to define the possible underlying mechanisms at the cellular and molecular level. The goals of this study were to determine: 1) whether PCP treatment alone was neurotoxic; 2) whether PCP treatment increased NMDA receptor mRNA and/or function (i.e., NMDA-induced neurotoxicity); 3) whether NMDA treatment in the presence of physiological Mg²⁺resulted in apoptotic neuronal death, and, if so; 4) its association with alterations in steady-state levels of Bcl-2 family proteins; and 5) whether the apoptotic mechanism involved the production of superoxide anion or other ROS.

PCP Neurotoxicity.

Compared with buffer-treated cultures, we never observed a significant effect of treatment with 3 μM PCP alone. Thus, in this sense, the forebrain culture model does not mimic what we observe in vivo. Our interpretation of this finding is that even though the NMDA receptor is up-regulated (see below) as we observed in vivo, there is insufficient “NMDAergic” tone in the culture to produce toxicity. This conclusion is in keeping with the relative sparseness of synaptic connections in culture compared with the in vivo situation.

Activation of NMDA receptors is well known to kill neurons via a necrotic mechanism characterized by excessive sodium and calcium entry, accompanied by chloride and water entry, that leads to cell swelling and death (Rothman et al., 1985). More recently, it has been realized that NMDA receptor activation also can lead to apoptotic cell death (Ankarkona et al., 1995; Lesort et al., 1997). The characteristics of an excitotoxic insult that leads to necrosis or apoptosis are not clear-cut and may depend on the concentration of the glutamate agonist, the duration of the treatment, the receptor subtype activated, and the cell type and its stage of development or maturity (Portera-Cailliau et al., 1997; Cheung et al., 1998). In general, it is thought that a mild insult, given sufficient time, will result in apoptosis, whereas a more severe insult will lead to necrotic cell death. However, it is now becoming apparent that glutamate-mediated cell death is often not exclusively either necrosis or apoptosis, but presents with features characteristic of both (van Lookeren Campagne et al., 1995; Lesort et al., 1997; Sohn et al., 1998). Although it was not the intent of this study to distinguish absolutely between necrosis and apoptosis, the similarity between NMDA-induced alterations in mitochondrial MTT uptake, internucleosomal (histone-associated) DNA fragments, and TUNEL staining suggests that this treatment protocol produces cell death by a mostly apoptotic mechanism. For example, 100 μM NMDA increased LDH release by ∼2.5-fold, whereas the number of TUNEL-positive neurons increased by ∼5-fold. Although comparisons between assays are difficult, this suggests that the NMDA-induced decrease in cell viability (LDH release) can be accounted for by the increase in TUNEL-positive neurons. Although the TUNEL assay is not a completely certain marker of apoptosis, our assessment of nuclear condensation and fragmentation by Hoechst 33285 staining strongly supports the correlation between positive TUNEL staining and apoptotic nuclear morphology. In addition, the NMDA-induced decrease in mitochondrial viability assessed by MTT uptake is virtually the same magnitude as the increase in internucleosomal DNA fragments, which is thought to be a specific index of apoptosis. Thus, most of the loss of cell viability appears to be due to an apoptotic mechanism. We suggest that modest activation of NMDA receptors in the presence of Mg²⁺ over a relatively long time period produces a dose-related increase in neuronal apoptosis and that this is a valid model for the study of apoptotic cell death.

Effects of Chronic PCP on NMDA Receptor.

In this study, an in situ hybridization technique was used to investigate whether NMDA NR1 subunit mRNA levels were changed after PCP treatment. The results of this study indicate that there is an increased density of the NR1 subunit mRNA after 3 μM PCP treatment for 48 h. NR1 mRNA density previously has been shown to be increased in cultured cortical neurons after exposure to several antagonists, including D-AP5, CGS 19755, MK-801, and ethanol, but not after exposure to the α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid/kainate receptor antagonist 6-cyano-2,3-dihydroxy-7-nitroquinoxaline (Grant et al., 1990; Williams et al., 1992; Hu et al., 1996). Although the NR1 protein was not measured herein, these data are consistent with the increase in NR1 mRNA and protein that was observed in vivo after chronic PCP treatment (Hanania et al., 1999; Wang et al., 1999). The mechanism by which PCP and other NMDA antagonists are able to increase NR1 mRNA and protein is currently unknown, but is certainly deserving of further research because this mechanism could be key to understanding neurotoxicity in this and other models.

In a previous study, we also had observed that PCP treatment in vivo resulted in enhanced NMDA receptor function as assessed by measuring NMDA-stimulated neurotransmitter release (Hanania et al., 1999). This also appears to be true after PCP treatment in vitro. The observation that PCP treatment of cultures for 48 h resulted in an increased LDH release in response to 30 μM NMDA was confirmed by measuring the NMDA-induced alterations in the number of TUNEL-positive cells, MTT uptake, DNA fragmentation, Bax, and Bcl-X_L. Although PCP treatment also resulted in an increased response to 100 μM NMDA when LDH release and the number of TUNEL-positive neurons was measured, this difference was not great. Furthermore, no significant effect of the higher NMDA concentration was observed on the Bcl-X_L/Bax ratio. The differential effect of PCP treatment on the effect of 30 and 100 μM NMDA could be the result of a PCP-induced increase in the affinity of the receptor for NMDA. Such a result could occur as a consequence of altered receptor subunit stoichiometry after an increase in the NR1 subunit. In fact, we have previously observed that PCP treatment in vivo significantly reduced the ability of competitive glycine and NMDA antagonists, as well as PCP itself, to inhibit NMDA-induced transmitter release in an ex vivo paradigm (Wang et al., 1999). However, the affinity change observed in that study was opposite in direction to that required to explain the effect of PCP on NMDA responses observed in this study. Indeed, we suggest that a more likely explanation is that PCP increases the NMDA receptor number, but the effect of 100 μM NMDA is at or very near the ceiling of neurotoxicity in this model. A more detailed pharmacological analysis will be required to distinguish these possibilities. However, the rather steep concentration response to NMDA with either cell viability or apoptotic end points precludes such analysis in this study.

Mechanisms of NMDA-Induced Cell Death after Chronic PCP.

In addition to increases in intracellular sodium, chloride, and water observed after high concentrations of glutamate or NMDA, milder stimuli have been shown to be associated with a calcium-dependent increase in the formation of nitric oxide and other ROS or free radical generators such as superoxide anion and hydrogen peroxide (Gunasekar et al., 1995;Nicholls and Budd, 1998). NMDA receptor-mediated formation of ROS is blocked by inhibitors of mitochondrial electron transport such as rotenone (Dugan et al., 1995) and by a proton ionophore that blocks the uptake of Ca²⁺ into the mitochondria by dissipating the pH gradient (Reynolds and Hastings, 1995). This suggests that Ca²⁺ uptake by the mitochondria triggers the formation of ROS. How this occurs is not clear, but it has been suggested that superoxide anion is derived from complex I of the mitochondrial respiratory chain, at least in brain (Nicholls and Budd, 1998). Our observation that SOD, particularly in the presence of catalase, completely prevented NMDA-induced neurotoxicity, as well as the changes in Bcl-2 family proteins, suggest that superoxide anion may play a role in the regulation of Bax and/or Bcl-X_L. How exogenously administrated SOD/catalase retards this effect of intracellular ROS is not known, but this observation is not without precedence (Gunaseker et al., 1995).

The link between the formation of ROS and regulation of either Bax or Bcl-X_L in this system is unknown. Two early transcription factors, nuclear factor-κB (NF-κB) and activator protein-1 are known to be sensitive to the redox state of the cell (Gius et al., 1999). Both also are induced by glutamate in cerebellar cultures (Grilli et al., 1996). As such, they are possible candidates for mediating this linkage. The activation (and nuclear translocation) of NF-κB is mediated by phosphorylation of an inhibitory protein called I-κB, which in the unphosphorylated state keeps NF-κB sequestered in the cytoplasm. Although there is some evidence that I-κB also can be phosphorylated by other kinases, the two specific kinases responsible for phosphorylation of I-κB (and subsequent activation of NF-κB) contain critical cysteine residues in the kinase domain that are postulated to sense the local redox status (Gius et al., 1999). In this fashion, stimuli that increase ROS and the oxidative stress level are able to increase the translocation of NF-κB to the nucleus where it can act at specific DNA-binding sites to affect the transcription of many target genes. The presence of NF-κB-binding sites within the Bcl-X_L promoter region suggests that NF-κB most likely plays a role in the regulation of Bcl-X_L. Although highly speculative, NMDA-induced increases in [Ca²⁺]_i could result in mitochondrial calcium overload and subsequent increased production of ROS. This, in turn, could lead to the inhibition of Bcl-X_L synthesis through the pathway outlined above.

The impact of such a reduction is realized by the ability of Bcl-X_L to form heterodimers with Bax, thereby preventing Bax from promoting the release of cytochrome cfrom the mitochondria into the cytoplasm (although in some systems this antiapoptotic effect is thought to be mediated farther downstream;Rosse et al., 1998). Because the release of cytochrome cleads to the activation of caspases, a decrement in Bcl-X_L could then contribute to the ultimate demise of the cell. Increased [Ca²⁺]_i and ROS also can affect cell survival by altering the permeability of the mitochondrial inner membrane by causing transient or even permanent changes (Zamzami et al., 1998; Crompton, 1999). Alteration of the so-called mitochondrial permeability transition pore can lead to mitochondrial depolarization, depletion of ATP, and further loss of Ca²⁺ homeostasis. This results in mitochondrial swelling and stress of the outer membrane in such a way that permits the formation of Bax-induced pores that ultimately lead to cytochromec release and caspase activation.

In addition to this potential mechanism, it is possible that NMDA receptor activation also might result in an increased expression of Bax. The proapoptotic tumor suppressor protein p53 is known to up-regulate Bax in several settings (Chao and Korsmeyer, 1998). Furthermore, both glutamate and oxidative stress have been demonstrated to up-regulate p53 expression (Uberti et al., 1998, 1999). Finally, Bax has been implicated in several p53-dependent models of apoptotic cell death (Xiang et al., 1998; Cregan et al., 1999).

In conclusion, these data demonstrate that although PCP treatment is not neurotoxic by itself, it leads to an up-regulation of the obligatory NR1 subunit as well as an enhanced NMDA receptor function as assessed by measuring NMDA-induced loss of cell viability. Evidence also is presented suggesting that the ensuing cell death is largely apoptotic in nature and that it involves the production of ROS and a decrease in the Bcl-X_L/Bax ratio, which is known to play a crucial role in determining the fate of a cell in certain situations. Furthermore, evidence is presented that suggests that ROS formation is critical for the NMDA-induced alterations in Bcl-2 family proteins. Finally, hypothetical mechanisms linking the synthesis of ROS to changes in Bcl-X_L and Bax are proposed.