Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Cyanide is a rapid-acting neurotoxic compound that initiates a complex series of intracellular reactions leading to neuronal dysfunction and eventually cell death. Cyanide enhances N-methyl-D-aspartate (NMDA) receptor function (Patel et al., 1992; Sun et al., 1997) and mobilizes intracellular calcium stores (Yang et al., 1997). In cultured neurons, cyanide-induced cytotoxicity is linked to the NMDA receptor-mediated rise in cytosolic Ca²⁺ that in turn activates a series of biochemical reactions leading to generation of reactive oxygen species (ROS) and nitric oxide (NO). These oxidant species then initiate peroxidation of cellular lipids (Gunasekar et al., 1996). It is concluded that oxidative stress plays a pivotal role in cyanide-induced neurodegeneration (Kanthasamy et al., 1994). Oxidative stress can lead to either necrosis or apoptosis, based on the cell model and the level of oxidative insult (Bonfoco et al., 1995; Nicotera et al., 1996).

Depending on the cell type and the stimulus, a cell may die in either of two distinct ways: apoptosis or necrosis (Grooten et al., 1993). Apoptosis, considered the physiological form of cell demise, is an active process with distinct morphological and biochemical features (Vaux and Strasser, 1996; Darzynkiewicz et al., 1997). Apoptotic cells are characterized by condensed and fragmented nuclei, whereas necrotic cells show loss of plasma membrane integrity without apparent damage to nuclei. These two apparently opposite forms of cell death can be elicited by the same stimuli according to the intensity of the effect (Bonfoco et al., 1995; Hampton and Orrenius, 1997), suggesting that initial common events could be shared by apoptosis and necrosis.

We have shown previously that cyanide exposure induces different modes of cell death in different brain areas (Mills et al., 1999). Cell death occurs predominantly via apoptosis in the cortical region, whereas necrosis is observed in substantia nigra after the same dose of cyanide in mice. Selective vulnerability of different brain areas to cyanide may be explained by triggering of region-specific toxic pathways in which oxidative stress may be a common activator (Vaux and Strasser, 1996).

The objective of the present study was to determine differences in cell death responses between primary rat cortical (CX) and mesencephalic (MC) cell cultures to cyanide exposure. In addition, we determined the effects of antioxidants and caspase inhibition on the induction of neuronal cell death by cyanide.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Materials

Dulbecco's modified Eagle's medium (DMEM) and penicillin/streptomycin solutions were purchased from Invitrogen (Carlsbad, CA); fetal calf serum was from Hyclone Laboratories (Logan, UT); six- and 24-well plastic culture dishes were from Costar (Cambridge, MA); coverslips were from Fisher Scientific (Pittsburgh, PA); poly-L-lysine, catalase (CAT), and superoxide dismutase (SOD) were from Sigma-Aldrich (St. Louis, MO); N-nitro-L-arginine methyl ester (L-NAME) and 5-methyl-10,11-dihydro-5H-dibenzo[a,d]cyclohepten-5,10-imine (MK-801) were from Sigma/RBI (Natick, MA); 2,7-dichlorofluorescin diacetate (DCF-DA) was from Molecular Probes (Eugene, OR); Apoptag kit was from Oncor (Gaithersburg, MD); Z-VAD and caspase 3 substrate Ac-DEVD-AMC were obtained from Bachem Biosciences (King of Prussia, PA); anti-cytochrome c antibody was from BD PharMingen (San Diego, CA); and SYTO-13 and propidium iodide were purchased from Molecular Probes.

Cell Culture

Ventral mesencephalon tissue was removed from fetal Sprague-Dawley rats (15-17 days gestation) as described by Zeevalk et al. (1995). Briefly, the tissue was placed in Hanks' balanced salt solution), and the cells were dissociated with the addition of 0.025% trypsin at 37°C for 15 min. Trypsin digestion was stopped by adding 100 µl of trypsin inhibitor and 100 µl of DNase I. Dissociated cells were centrifuged at 1000g for 10 min and suspended in DMEM supplemented with 22 mM glucose, 2 mM glutamine, 2.2 g/l bicarbonate, 10% fetal bovine serum, 10% horse serum, and 1% penicillin/streptomycin (5000 U/ml). Cells were plated at a density of 5 × 10⁵ cells/cm² in 24-well culture plate or on microscope cover slides in six-well culture plates precoated with 0.1% poly-L-lysine and maintained in DMEM under controlled incubation conditions (37°C, 5% CO₂). On day 4, 10 µM cytosine arabinofuranoside was added to the medium for 24 h and DMEM was changed every 48 to 72 h until cells reached confluence (12-16 days). Primary CX cells were prepared from embryonic day 16 Sprague-Dawley rats. In brief, cerebral cortices were dissected and cells were dissociated in 0.025% trypsin at 37°C for 15 min. Trypsin digestion was stopped by adding trypsin inhibitor and DNase I. Cells were passed through a Pasteur pipette several times and plated at a density of 5 × 10⁵ cells/cm² onto 24-well plates or six-well culture plates containing sterile coverslips precoated with 10 µg/ml poly-L-lysine. Cells were grown in 80% DMEM supplemented with 10% fetal bovine serum, 10% horse serum, 22 mM glucose, 2 mM glutamine, and 1 ml penicillin/streptomycin (5000 U/ml/l) at 37°C in an atmosphere of 5% CO₂/95% O₂. Cytosine arabinofuranoside (10 µM) was added at 7 days in vitro to limit growth of glial cells. Medium was changed twice a week until experiments were carried out at 14 to 16 days in vitro

Evaluation of Apoptosis

Terminal Deoxynucleotidyl Transferase dUTP Nick-End Labeling (TUNEL) Staining. TUNEL was performed on paraformaldehyde (4%)-fixed cells using the Apoptag in situ apoptosis detection kit (Oncor). Briefly, cells were preincubated in equilibration buffer containing 0.1 M potassium cacodylate pH 7.2, 2 mM CaCl₂, and 0.2 mM dithiothreitol for 10 min at room temperature. This was then incubated in TUNEL reaction mixture containing (200 mM potassium cacodylate pH 7.2, 4 mM MgCl₂, 2 mM 2-mercaptoethanol, 30 µM biotin-16-dUTP, and 300 U/ml terminal deoxynucleotidyl transferase) in a humidified chamber at 37°C for 1 h. After incubating in stop/wash buffer for 10 min, the elongated digoxigenin-labeled DNA fragments were visualized using anti-digoxigenin peroxidase antibody solution followed by staining with 0.2 mg/ml diaminobenzidine tetrachloride and 0.005% H₂O₂ in PBS, pH 7.4. Cells were then counterstained with hematoxylin. The selectivity of the assay is based on the presence of 3-OH DNA fragment ends in apoptotic cells.

Extraction and electrophoresis of DNA. Intracellular DNA was extracted according to the protocol described by Herget et al. (1998) with slight modification. Cells (2 × 10⁷) were washed with PBS, pH 7.4, twice and collected by centrifugation. Cell pellets were lysed with 0.5 ml of lysis buffer (10 mM Tris-HCl pH 7.4, 10 mM EDTA, and 0.5% sodium dodecyl sulfate) for 10 min on ice. After treatment with RNase A (final concentration 100 µg/ml) for 1 h at 37°C, proteinase K (final concentration 100 µg/ml) was added and samples were incubated for 4 h at 50°C. DNA was precipitated with 0.1 volumes of 3 M sodium acetate, pH 5.2, and 2.5 volumes of precooled ethanol and then resuspended in Tris-EDTA buffer. For analysis, 10 or 20 µg of DNA was loaded on to a 1.2% agarose gel containing ethidium bromide (10 µg/ml) and electrophoresed in Tris-boric acid-EDTA buffer at 70 V for 2 h. DNA was visualized under ultraviolet light and photographed.

Quantitation of Necrotic Cell Death

Necrotic cell death was quantitated using two DNA fluorescent dyes, SYTO-13 and propidium iodide (Ankarkrona et al., 1995). Both dyes stain DNA but only SYTO-13 is membrane-permeable. Thus, SYTO-13 stains normal cells with a green fluorescence, whereas only cells with disrupted plasma membranes stain red with propidium iodide. The percentage of cells stained positive for propidium iodide was determined as an estimate of necrosis.

Cytotoxicity was also estimated by measurement of LDH efflux from damaged cells in the medium over a 24-h exposure to cyanide. LDH activity was determined by the spectrophotometric method of Vassault (1983).

Measurement of Mitochondrial Membrane Potential

Mitochondrial membrane potential was monitored following the procedure of Satoh et al. (1997). Cells grown on coverslips were incubated with KCN, washed twice with Krebs-Ringer buffer. Then cells were loaded with 10 µM rhodamine 123 (R123) and incubated at 37°C for 30 min in the dark. Uptake of R123 into mitochondria is a direct reflection of its permeability. Therefore, increases in R123 fluorescence reflect losses of mitochondrial membrane potential. Loaded cells were washed twice with Krebs-Ringer buffer and mounted in a cell chamber on a microscope attached to a spectrofluorometer. Changes in R123 fluorescence were monitored for 10 min at 498-nm excitation and 525-nm emissions. All the inhibitors were added 10 min before cyanide and fluorescence intensity was recorded.

Monitoring ROS

2,7-Dichlorofluorescein (DCF), the fluorescent, oxidized product of DCF, was assayed to monitor generation of ROS (Gunasekar et al., 1995). Neurons grown on glass coverslips were loaded with DCF-DA and fluorescence was monitored with an SLM-8000 spectrofluorometer (SLM-Aminco, Urbana, IL) attached via fiber optics to a diaphot TMD microscope (Nikon, Tokyo, Japan). To load cells with DCF-DA, culture medium was replaced with 1 ml of prewarmed Krebs-Ringer solution and 10 µl of 30 mM DCF-DA was added and incubated for 15 min at room temperature in the dark. Coverslips containing the neurons loaded with DCF were placed in a cell chamber (Medical System, Inc., Greenvale, NY) mounted on a heated (37°C) microscope stage. Fluorescence of single cell was monitored over a 10-min period after addition of cyanide at excitation and emission wavelengths of 475 and 525 nm. All the inhibitors were added 10 min before cyanide and fluorescence intensity was recorded.

Measurement of NOS Activity

As an index of NOS activity, nitrite formation (NO breakdown product) was measured by the method of Ignarro et al. (1987). Nitrite was quantitated after exposure to KCN (100-400 µM) for 24 h. Cells grown in six-well culture dishes were used for these experiments. Nitrite levels were estimated at 548 nm after adding 500 µl of 1% sulfanilic acid dissolved in 5% (v/v) H₃PO₄ and 500 µl of 0.1% N-(1-napthyl)-ethylenediamine to 1 ml of incubation medium of control and treated cells.

Western Blot Analysis for Cytochrome c

Cytochrome c was assessed with Western blots (Bradham et al., 1998). Cells were washed with ice-cold PBS and harvested by centrifugation at 1000 rpm for 5 min. Mitochondria-free cytosolic extracts were prepared by homogenizing cells suspended in a buffer containing 220 mM mannitol, 68 mM sucrose, 20 mM HEPES pH 7.4, 50 mM KCl, 5 mM EGTA, 1 mM EDTA, 2 mM MgCl₂, 1 mM dithiothreitol, and protease inhibitors on ice (4°C). Homogenates were centrifuged at 15,000g for 40 min, and mitochondria-free cytosolic supernatants were frozen at 70°C until further analysis. Protein content of mitochondria-free cytosolic supernatant was determined by the Bradford assay (Bio-Rad, Hercules, CA). Samples containing 25 µg of protein were boiled in Laemmli buffer for 5 min and subjected to electrophoresis in 12% SDS-polyacrylamide gel, followed by transfer to a nitrocellulose membrane. After blocking with phosphate-buffered saline containing 5% nonfat dry milk and 0.1% Tween 20, the membrane was exposed to the primary antibody (anti-mouse cytochrome c monoclonal antibody) for 3 h at room temperature on a shaker. Reactions were detected with the fluorescein-linked anti-mouse Ig (second antibody) conjugated to horseradish peroxidase using a Storm 860 fluorescence-PhosphorImager (Molecular Dynamics, Sunnyvale, CA). Densitometry analysis was performed using the ImageQuant software (Molecular Dynamics).

Caspase-3-Like Protease Activity Measurement

Caspase-3-like protease activity was measured as described by Yoshimura et al. (1998). After treatment cells were harvested with PBS and centrifuged at 1500 rpm for 5 min. Cell pellets were resuspended in 2 ml of buffer containing 50 mM Tris-HCl, 1 mM EDTA, and 10 mM EGTA and lysed with 10 µM digitonin. After incubation at 37°C for 10 min, lysates were centrifuged at 900g for 3 min, and the supernatants incubated with 50 µM caspase-3 substrate Ac-DEVD-MCA at 37°C for 1 h. Substrates were cleaved by active caspase-3 into 7-amino-4-methylcoumarin, which is fluorescent. The level of fluorescence measured using an F-2000 spectrofluorometer (Hitachi, Yokohama, Japan), with excitation and emission wavelengths of 380 and 460 nm, indicated caspase activity.

Statistics

Data were expressed as mean ± S.E. and statistical significance was assessed by one-way analysis of variance followed by Tukey-Kramer multiple range test. Differences were considered significant at p < 0.05.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Cyanide-Induced Apoptosis and Necrosis. As shown in Fig. 1, cyanide exposure (400 µM) for 24 h induced primarily apoptosis in CX cells and necrosis in MC cells. Exposure of CX cells to varying concentrations of cyanide (100-400 µM) for 24 h induced a concentration-dependent apoptotic cell death (TUNEL staining), whereas MC cells did not undergo apoptosis (Figs. 1A and 2). To further confirm cyanide-induced apoptosis, DNA fragmentation was evaluated with gel electrophoresis. CX cells exposed to cyanide (100-400 µM) for 24 h produced distinct DNA laddering characteristic of apoptosis (Fig. 3).

View larger version (59K): [in this window] [in a new window]   Fig. 1.   A, cyanide-induced apoptosis. Neurons were treated with 400 µM KCN for 24 h and photographed after TUNEL staining. Cyanide induced an apoptotic death in CX cells as noted by increased TUNEL-positive (darkly stained) cells. B, cyanide-induced necrosis. Cells were treated with 400 µM KCN for 24 h and photographed after SYTO-13 and propidium iodide (PI) staining. Cyanide induced extensive necrotic cell death in MC cells as indicated by increased number of PI-positive (stained in red) cells.

View larger version (22K): [in this window] [in a new window]   Fig. 2.   Cyanide-induced apoptosis in CX and MC cells. Cells were exposed to cyanide for 24 h and apoptosis confirmed by counting the number of TUNEL-positive cells. Four randomly selected fields of TUNEL-stained cells were counted and the average percentage of apoptotic cells per total cells were determined. Percentage of cells stained for apoptosis (mean ± S.E.) is indicated for three experiments. Asterisks indicate significant difference from respective control; *, p < 0.05.

View larger version (94K): [in this window] [in a new window]   Fig. 3.   Agarose gel electrophoretic detection of KCN-induced DNA fragmentation in CX cells. Extracted DNA was loaded on to a 1.2% agarose gel containing ethidium bromide (0.5 mg/ml) and electrophoresed at 70 V for 2 h. Experiments were repeated three times and a representative gel image showing DNA laddering is presented.

View larger version (24K): [in this window] [in a new window]   Fig. 4.   Cyanide-induced necrosis in CX and MC cells. Cells were treated with KCN for 24 h and the percentage of necrotic cells noted. Four randomly selected fields of SYTO-13 and propidium iodide-stained cells were counted and the average percentage of necrotic cells per total cells was determined. Data represent mean ± S.E. of four experiments. *, significant difference from respective control group at p < 0.05.

View larger version (33K): [in this window] [in a new window]   Fig. 5.   Cyanide-induced LDH release and blockade of LDH release by antioxidants and NMDA receptor antagonist. Cells were pretreated with MK-801 (1 µM), L-NAME (300 µM), SOD (100 units), and CAT (100 units) 10 min before cyanide (400 µM) treatment and LDH release was determined after 24 h. Data represent mean ± S.E. of four or more experiments. Asterisks indicate significant difference at p < 0.05 for control versus KCN and #, p < 0.05 for various pretreatment + KCN versus KCN treatment alone.

View larger version (32K): [in this window] [in a new window]   Fig. 6.   Effect of antioxidants and NMDA receptor antagonist on cyanide-induced apoptosis and necrosis. Cells were pretreated with MK-801 (1 µM), L-NAME (300 µM), SOD (100 units), CAT (100 units), and Z-VAD (100 µM) for 10 min before addition of cyanide (400 µM). Data represent mean ± S.E. of four experiments. Asterisks indicate significant difference at p < 0.05 for control versus KCN treatment and #, p < 0.05 for various pretreatment + KCN versus KCN alone.

View larger version (38K): [in this window] [in a new window]   Fig. 7.   Nitrite level in CX and MC cells after cyanide exposure for 24 h. Data represent mean ± S.E. of four experiments. Asterisks indicate significant difference from control at p < 0.05.

Cyanide-Induced ROS Generation. Intracellular oxidant species (ROS and NO) are initiating signals of cyanide-induced cell death (Gunasekar et al., 1996; Shou et al., 2000). In CX cells cyanide (100-400 µM) produced a concentration-dependent increase in fluorescence units reaching an 8-fold increase at 400 µM compared with control (Fig. 8A). MK-801 inhibited cyanide-induced ROS generation by more than 50%, indicating that ROS generated after cyanide treatment is associated with NMDA receptor activation. Pretreatment with the antioxidants SOD and catalase decreased cyanide-induced ROS generation, whereas L-NAME and Z-VAD failed to block the ROS generated after cyanide in CX cells (Fig. 8B).

View larger version (32K): [in this window] [in a new window]   Fig. 8.   ROS generation induced by cyanide. A, cells were treated with 100 to 400 µM cyanide and fluorescence monitored over 10-min period. B, inhibition of cyanide-induced ROS generation by MK-801, antioxidants, and Z-VAD. Cells were pretreated with MK-801 (1 µM), L-NAME (300 µM), SOD (100 units), CAT (100 units), and Z-VAD (100 µM) for 10 min before treating with cyanide (400 µM). DCF fluorescence intensity was monitored over a 10-min period after cyanide exposure. Data represent mean ± S.E. of four experiments. Asterisks indicate significant difference at p < 0.05 for control versus cyanide treatment and at #, p < 0.05 for various pretreatment + KCN and KCN alone.

Cyanide-Induced Mitochondrial Membrane Potential Change. Changes in mitochondrial membrane potential are an early indicator of impending cell death (Vayssiere et al., 1994; Ankarkrona et al., 1995). To assess whether cyanide alters mitochondrial membrane potential, CX and MC cells were loaded with rhodamine 123. On exposure to cyanide the mitochondrial membrane potential dropped moderately in CX cells. In contrast, MC cells showed a rapid and extensive reduction in mitochondrial membrane potential after cyanide exposure (Fig. 9).

View larger version (31K): [in this window] [in a new window]   Fig. 9.   Effect of cyanide on mitochondrial membrane potential. Cells were loaded with rhodamine 123 after addition of cyanide and fluorescence was noted over 10-min periods. Data represent mean ± S.E. of three or more experiments. Asterisks indicate significant difference from respective control at p < 0.05.

Cyanide-Induced Release of Cytochrome c from Mitochondria. Loss of mitochondrial membrane integrity permits release of cytochrome c into the cytosol (Tatton and Olanow, 1999), which then promotes caspase activation (Li et al., 1997). Cytosolic cytochrome c seems to be an obligate component of the cytotoxic response in certain cells. Quantitative analysis of Western blots on nuclear cell lysates of CX and MC cells exposed to cyanide was performed to assess cyanide-induced changes in cytosolic cytochrome c level. Treatment of CX cells with cyanide produced a concentration-dependent increase in cytochrome c level, but no change in cytochrome c was observed in MC cell necrosis induced by cyanide (Fig. 10).

View larger version (19K): [in this window] [in a new window]   Fig. 10.   Effect of KCN on cytochrome c levels in CX and MC cells. A, representative results from cytochrome c analysis. The cells were treated with 100 to 400 µM of cyanide for 6 h and the cytosolic extract (25 µg of protein) were subjected to Western blot analysis using an anti-mouse cytochrome c antibody. B, results in A and in parallel experiments were quantified by densitometry using ImageQuant software and are shown as mean ± S.E. (n = 3). Maximal value is set at 100%. Asterisks indicate significant difference from control at p < 0.05.

Cyanide-Induced Caspase 3-Like Activity. The caspase family of cysteine proteases plays an important role in apoptosis but the relationship of these proteases to necrosis has not been described. To measure the extent of caspase-3-like protease activity, CX and MC cells were loaded with Ac-DEVD-MCA, a fluorogenic tetrapeptide substrate that is cleaved by caspase-3-like proteases to release 7-amino-4-methylcoumarin. Caspase-3-like protease activity showed a concentration-dependent increase in CX cells. In contrast, caspase was not involved in cyanide-induced necrosis because no change in caspase activity was observed in MC cells (Fig. 11).

View larger version (22K): [in this window] [in a new window]   Fig. 11.   Caspase-3-like protease activity induced by cyanide. Cells were treated with varying concentrations of cyanide for 24 h, and caspase-3-like protease activity was measured as the fluorescence level of cleaved substrates. Data represent mean ± S.E. for six experiments. Asterisks indicate significant difference from control at p < 0.05.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

Apoptosis and necrosis have traditionally been regarded as two different modes of death induced by different stimuli (Wyllie, 1980; Duval and Wyllie, 1986). Recent studies indicate that some stimuli, such as tumor necrosis factor, can induce either apoptosis or necrosis, depending on the cell type (Vercammen et al., 1998; Fiers et al., 1999). Genotoxic agents that cause apoptosis also cause necrosis in several cell lines when apoptosis is blocked by caspase inhibitors (Lemaire et al., 1998). Present results show that cyanide exposure produces cell death, either by apoptosis or by necrosis, depending on the cell type. In MC cells, necrotic cell death as assessed by propidium iodide staining and LDH measurement, was the predominant response to 400 µM cyanide, whereas at the same concentration of cyanide CX neurons showed extensive apoptotic death. Between 400 and 500 µM cyanide, CX cells switch their mode of death and die of necrosis rather than apoptosis (Shou et al., 2000). Some critical change occurs in CX cells when cyanide concentrations increase above 400 µM. In contrast MC cells do not die of apoptosis but succumb to necrosis when cyanide concentration is sufficiently high. These findings are consistent with our previous results obtained in vivo (Mills et al., 1999).

In these two modes of cell death, oxidative stress was a common factor, but the level of oxidative species generation varied with each cell type. Others report that oxidative stress can be involved in either apoptosis or necrosis (Zeevalk et al., 1998; Mason et al., 1999; Shou et al., 2000). It is possible that when high ROS levels accumulate in the cell, direct and irreversible damage of cellular components can lead to necrosis. Moderate ROS levels, on the other hand, may function as second messengers and regulating molecules, which mediate apoptotic death (Denecker et al., 2001). In MC cells, cyanide induced a progressive but rather small increase in necrosis at the concentration range of 100 to 300 µM and at 400 µM, the response showed a steep rise in necrotic index. The reason for this steep concentration dependence may be related to a similar pattern of ROS generation in response to cyanide at the indicated concentrations. The sudden burst of oxidative stress at 400 µM of cyanide may be responsible for this effect.

Antioxidants that prevented cyanide-induced death also prevented accumulation of oxidative species, supporting the hypothesis that toxicity is initiated by ROS. The concentrations used in the present study for various antioxidants and blockers were chosen based on the previous work from our laboratory (Gunasekar et al., 1995; Shou et al., 2000) and other laboratories (Bonfoco et al., 1995), which are known to be maximally effective.

Catalase pretreatment prevented cyanide-induced CX apoptosis and ROS formation, but not necrotic death of MC cells. On the other hand, L-NAME, a blocker of NOS, prevented necrosis and ROS formation in MC cells but had a lesser effect in CX cells. It seems that CX cell apoptosis is largely mediated by hydrogen peroxide and superoxide radicals, whereas necrotic actions of cyanide are mediated by NO and superoxide in MC cells. NO generated after NOS stimulation can further react with superoxide to form peroxynitrite, a potent cytotoxic oxidant. Because NO inhibits the mitochondrial respiratory chain, the resultant energy depletion would contribute to cell death in MC cells. Similarly, Leist et al. (1999) showed that NO prevents caspase activation by inhibiting mitochondrial respiration, thereby decreasing intracellular ATP levels eventually leading the cell death by necrosis.

Changes in mitochondrial membrane potential and membrane depolarization are associated with cellular oxidant accumulation (Satoh et al., 1997). Mitochondrial dysfunction with insufficient energy levels has been proposed as a determinant for cell death modes (Ankarkrona et al., 1995; Bonfoco et al., 1995; Nicotera et al., 1999). In the present study CX cells responded to cyanide treatment with moderate reduction in mitochondrial membrane potential, whereas in MC cells the membrane potential dropped dramatically. A relatively minor mitochondrial dysfunction, as seen with CX cells, may act as a signal for apoptosis and initiate release of mitochondrial factors (cytochrome c) that could activate downstream degradative processes (Nicotera et al., 1996). On the other hand, extensive collapse of mitochondrial membrane potential as seen in MC cells and impaired respiration are likely to result in energy dissipation leading to necrosis (Nicotera and Leist, 1997; Tsujimoto, 1997).

Recently, it has been proposed that cell death also occurs as an apoptosis-necrosis continuum (Martin, 2001). In this continuum, neuronal cell death occurs as intermediate or hybrid forms of cell death with coexisting characteristics that lie along a structural continuum with apoptosis and necrosis at the extremes (Portera-Cailliau et al., 1997). Studies indicate that excitotoxin-induced neuronal death can appear as apoptosis, necrosis, or as hybrids of apoptosis and necrosis. Ischemia-induced degeneration of selectively vulnerable neurons is phenotypically necrotic, but apoptosis also occurs in some neurons and non-neuronal cells (Martin et al., 1998). Thus, a single insult to the central nervous system can cause death or injury to different populations of cells, resulting from multiple, distinct causal processes that temporally overlap. Taken together, these data suggest a complex picture of cytotoxin-related death mechanisms in neurons.

It is clear, however, that blockade of cyanide-induced cell death by MK-801 shows the involvement of NMDA receptor activation in both CX and MC cells. This confirmed our previous observation that blockade of NMDA receptors protected against cyanide cytotoxicity in CX (Shou et al., 2000) and cerebellar granule (Gunasekar et al., 1996) cell cultures.

In summary, our results demonstrate that exposure to cyanide damages cultured CX cells primarily through an apoptotic pathway, whereas the same dose of cyanide resulted in necrosis of MC cells. The intensity and nature of the oxidative stress and the degree of mitochondrial dysfunction may be related to the different modes of death exhibited by these two cell types.