BRAIN RESEARCH ELSEVIER
Brain Research 703 (1995) 63-71
Research report
Carbamazepine induction of apoptosis in cultured cerebellar neurons: effects of N-methyl-D-aspartate, aurintricarboxylic acid and cycloheximide Xiao-Ming Gao, Russell L. Margolis, Peter Leeds, Christopher Hough, Robert M. Post, De-Maw Chuang * Section on Molecular Neurobiology, Biological Psychiatry Branch, NIMH, NIH Bldg. 10, Rm. 3N212, 10 Center Dr. MSC 1272, Bethesda, MD 20892-1272, USA
Accepted 8 August 1995
Abstract
We have previously demonstrated that carbamazepine (CBZ) at concentrations above the therapeutic range is toxic to cultured cerebellar granule ceils. Here, we ask whether the effect of CBZ involves neuronal apoptosis or necrosis. Treatment of cultured cerebellar granule cells with CBZ for 3 days resulted in a concentration-dependent fragmentation of DNA revealed as a laddered pattern in agarose gel electrophoresis, a phenomenon characteristic of apoptosis. Pretreatment of cells with N-methyl-D-aspartate (NMDA) blocked CBZ-induced DNA fragmentation and neurotoxicity as assayed by loss of mitochondrial activity with MqT or by [3H]ouabain binding to Na+/K+-ATPase. Aurintricarboxylic acid (ATA), a polyanionic dye, also markedly suppressed DNA fragmentation and cell death detected by morphological ex~tmination. A considerable level of DNA ladder formation was detected in untreated cells and this basal DNA fragmentation was also blocked by NMDA and ATA. Moreover, NMDA and ATA prevented CBZ-induced chromatin condensation as revealed by DNA binding with the fluorescent dye Hoechst 33258. Pretreatment of cells with cycloheximide, a protein synthesis inhibitor, prevented CBZ-induced cell death detected morphologically and attenuated CBZ-induced neurotoxicity assessed by mitochondrial activity and [3H]ouabain binding assays. Taken together, our results suggest that CBZ-induces death of cerebellar granule cells by an apoptotic process that is sensitive to NMDA, ATA and cycloheximide. Keywords: Carbamazepine; Apoptosis; DNA fragmentation; Cerebellar granule cell; N-Methyl-D-aspartate; Aurintricarboxylic acid; Cycloheximide
1. Introduction
Apoptosis, sometimes referred to as programmed cell death, is a process by which a cell dies through activation of intrinsic cellular systems; in essence, the cell kills itself. This form of cell death is typified by defined morphological changes (including cell shrinkage, condensed nuclei and early sparing of the plasma membrane)[28], dependence on gene transcription [23], and early cleavage of DNA between nucleosomes (resulting in a laddered pattern when the DNA fragments are separated electrophoretically) [8,22,24]. These distinct features are absent in another form of cell death termed necrosis (for a review see [25,32]). Apoptosis is probably caused by a heterogeneous
* Corresponding author. Fax: (1) (301) 402-0052. 0006-8993/'95/$09.50 © 1995 Elsevier Science B.V. All rights reserved SSDI 0006- 8993(95)01066- ]
mixture of pathways that vary among cell types. However, activation of certain pathways, especially that leading to endonuclease cleavage of DNA, may be necessary for apoptotic death in many different cell types. Activation of such pathways therefore serves to mark apoptotic cell death and provide specific clue about the mechanism of death in a particular cell type. Carbamazepine (CBZ) is a drug used in neurology for treatment of epilepsy [1,2] and paroxysmal pain syndromes such as trigeminal neuralgia [5,6]. It is also used in psychiatry for the treatment of acute mania and for prophylactic treatment of manic-depressive illness [33,34]. When used in combination, CBZ potentiates the therapeutic action of lithium or valproate for bipolar depressive patients (for a review, see [34]). However, CBZ has side effects such as teratogenesis and an overdose can induce ataxia, nystagmus, vertigo, coma, respiratory depression, and even seizures [13,18]. Differential mechanisms involved in the
X.-M. Gao et al. / Brain Research 703 (1995) 63-71
64
therapeutic and toxic effects of CBZ have not been adequately delineated. Using cerebellar granule cells as a model, we have recently reported that CBZ at a concentration above the therapeutic plasma level induces neurotoxicity [12]. Neurotoxicity of these cells, as assessed morphologically and biochemically, occurs only after a 2-3-day exposure to CBZ and is completely blocked by N-methylD-aspartate (NMDA) through an NMDA receptor-mediated mechanism. To determine if CBZ induces neurotoxicity by an apoptotic mechanism, we sought evidence of DNA fragmentation after CBZ treatment of cerebellar granule cells. We also examined the effect of aurintricarboxylic
A
CBZ
Marker DNA (bp)
o
r
20 - -
(pM) 50
I i
100
300 A.
WIT Assay
tO 4.a
-~
140 t
o tO
120,
o
m
: ~c-
200
< v
100
100 '
0
g
80'
>,
60
z
a
e~
•-~
40
®
20
0 0
0
B.
3H-ouabain
CBZ
Binding
NMDA
NMDA+CBZ
Assay
120
I O0
80
g m
50
100
CBZ Concentration (pM) Control
o tO
20
Fig. 2. CBZ induces DNA fragmentation in a concentration-dependent manner. Cells grown in 60-mm dishes were treated with the indicated concentrations of CBZ for 72 h. Soluble DNA was prepared from cell extracts and subjected to agarose gel electrophoresis as described in the Methods. A: photograph of ethidium bromide stained agarose gel of fragmented DNA; (B), quantification of DNA fragmentation by image analysis. Band density was integrated over all nucleosomal bands and corrected for background fluorescence. Shown in the left margin of (A) are DNA markers of different sizes. The results in B are means + ranges of duplicate determinations and are expressed as % of vehicle control. The experiment has been reproduced four times with essentially the same results.
60
~
4o
=
20
O
0 Control
CSZ
NMDA
NMDA+CBZ
Fig. 1. NMDA protects against CBZ-induced neurotoxicity of cerebellar granule cells as determined by MTI" and [3H]ouabain binding assays. Cells grown in 24-well plates were treated with 100 NM CBZ or vehicle for 72 h as described in the Methods. When used, 100 ~M NMDA was added to the culture 2 h prior to CBZ exposure. Cell survival was determined by using M T r assay (A) or [3H]ouabain binding to Na +/K+-ATPase (B). Data are means + S.E.M. of % of control of four determinations. The experiment has been repeated four times with nearly identical results. The 100% values are 0.746+0.019 absorbance unit at 540 nm for MTT assays and 14327+572 @ m / w e l l for [3H]ouabain binding assays. *, P < 0.05; * * * P < 0.001 when compared with the vehicle control, using the Student's t-test.
acid (ATA), a polymer of salicylic acid purported to be an inhibitor of calcium- and magnesium-activated DNA endonucleases [4,16,29,35], on CBZ-induced morphological changes and DNA fragmentation. Finally, we investigated whether CBZ-induced neurotoxicity requires de novo protein synthesis, a marker for many cases of apoptosis [23,25,32].
2. Materials and methods 2.1. Chemicals
CBZ was obtained from Research Biochemicals International (Natick, MA). ATA (trisodium salt) was obtained
X.-M. Gao et al. / Brain Research 703 (1995) 63-71
from Aldrich Chemical (Milwaukee, WI). NMDA, cytosine arabinoside, RNase and SDS were from Sigma Chemical (St. Louis, MO). Proteinase K, agarose, phenol and DNA ladder standards were, purchased from GIBCO-BRL Life Technologies (Gaithersburg, MD). Chloroform, isomyl alcohol and isopropyl alcohol were from J.T. Baker (Phillisburg, NJ).
2.2. Cell cultures and carbamazepine treatment Cerebellar granule cells were prepared from 8-day-old Sprague-Dawley rats and cultured in the presence of 25 mM KCI on 24-well plates or 60-mm dishes as previously described [12]. Cytosine arabinoside (10 /xM) was added approximately 24 h after plating to arrest the growth of non-neuronal cells. Routinely, 100/zM CBZ or its vehicle was added to the growth medium on the 5th or 6th day in culture, and the granule cells were harvested 72 h later for assessment of morphology, DNA fragmentation and neuronal survival (see below). It is well established that cerebellar granule cells differentiate in culture to form a nearly homogenous population of excitatory neurons using glutamate as the primary transmitter and that contaminating glial cells represent only 2-3% of the total cell population. [7,11,12].
used to determine cell survival in a quantitative colorimetric assay [30]. The tetrazolium ring of MTT is cleaved by various dehydrogenase enzymes in active mitochondria into a blue colored product, formazan. Cerebellar granule cells grown in 24-well plates were incubated with MqT (125 /~g/ml) in the growth medium for 1 h at 37°C. The formazan formed was solubilized with 1.2 ml dimethlsulfoxide (DMSO) per well and quantified spectrophotometrically using a wavelength of 540 nm. Neuronal survival was also quantified by measuring high affinity binding of [3H]ouabain to plasma membrane Na+/K+-ATPase of neurons, as previously described [27] and slightly modified [12]. In essence, cells grown in 24-well plates were washed and incubated with 25 nM [3H]ouabain at 37°C for 1 h, washed again three times, dissolved in 3% Triton X-100, and counted for the determination of specific binding of [3H]ouabain.
A Marker
DNA (bp)
2072 1500
2.3. Analysis of DNA fragmentation DNA fragmentation was assessed by using either a genomic or soluble DNA preparation. Total genomic DNA was isolated from cerebellar granule cells grown on 60 mm dishes using a standard procedure. Briefly, cells were lysed in a Tris-HCl buffer containing 1 mM EDTA and 0.2% Triton X-100 and the lysates were treated with RNase (50 /zg/ml) and proteinase K (100 /zg/ml). The DNA was further deproteinated with 0.2% SDS, extracted with phenol-chloroform-isoamyl alcohol and then precipitated with ethanol. The i:~olated DNA fragments were electrophoretically separated in a 1.5% agarose gel along with molecular size markers and were visualized by staining with ethidium bromide. Soluble DNA was prepared using the procedures of Ishida et al. [20]. Briefly, cells were scraped off 60-mm dishes and then isolated by a low speed centrifugation. The resulting cell pellets were lysed in a Tris-EDTA-Triton X-100 solution and the lysates centrifuged at 13,000 g for 10 min. The DNA fragments in the high speed supernatant were precipitated with ethanol. The isolated nucleic acids fragments were treated with RNase prior to agarose gel electrophoresis in the presence of ethidium bromide.
2.4. Measurement of neurotoxicity The mitochondrial activity that cleaves Mq'T (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide) was
65
600
100
B ~)
140
120
8~o ~oo Eu
80
~-~
60
'~
40
Z Q
20 Control
CBZ
NMDA
NMDA +
CBZ Fig. 3. NMDA inhibits basal and CBZ-induced DNA fragmentation. Cells were pretreated with 100 /zM NMDA for 2 h and then exposed to 100 # M CBZ for 72 h. Total genomic DNA was isolated as described in the Methods. An aliquot containing 1 /.Lg DNA was applied to each lane for agarose gel electrophoresis. A: photograph of ethidium bromide stained agarose gel of fragmented DNA; B: quantification of DNA fragments as described in the legend to Fig. 2. Note that NMDA attenuated the level of the vehicle control and completely blocked CBZ-induced DNA fragmentation.
66
X.-M. Gao et al. /Brain Research 703 (1995) 63-71
No CBZ
A
with 100 pM CBZ
B
(0 pM)
(0 pM)
C
D
(5 pM)
(s pM)
E
F
(50 pM)
G (100 pM)
(so pM)
H (lOO pM)
Fig. 4. ATA protects against CBZ-induced neurotoxicity, as revealed by phase contrast microscopy. Cells were pretreated with the indicated concentrations of ATA for 2 h and were then exposed to 100/xM CBZ for 72 h. Cell morphology was examined by phase contrast microscopy, at 200 × magnification. Note the marked disaggregation of neuronal clumps, disruption of neuronal processes and appearance of smaller translucent dead cells after CBZ treatment. ATA, at 5 / z M partially and at 50 or 100/xM, completely prevented this aspect of CBZ neurotoxicity. ATA did not significantly affect the morphology of untreated cells.
X.-M. Gao et al. / Brain Research 703 (1995) 63-71
67
3. Results Marker DNA (bp)
Fig. 5. ATA inhibits CBZ-inducedand basal DNA fragmentation.Cells were pretreated with 50 or 100 jtLMATA for 2 h and then exposed to CBZ for 72 h. Soluble DNA was prepared and then subjected to agarose gel electrophoresis as described in the Methods. Shown in the left margin are DNA size markers. Note that ATA inhibitedboth CBZ-inducedand basal DNA fragmentation.The experimenthas been repeated four times with essentiallythe same results. The apparent increasein the mobilityof DNA shown on the top of the gel in ATA-treated samples may reflect either ATA-induced degradation of high molecular weight DNA or merely binding of ATA to this DNA species.
Exposure of cultured cerebellar granule cells on the 5th or 6th day in culture to 100 tzM CBZ for 3 days resulted in an approximately 60% loss of neuronal viability assessed by the measurement of mitochondrial activity (MTT assay) or by [3H]ouabain binding to N a + / K ÷ - A T P a s e (Fig. 1A, B). Both assays showed that the CBZ neurotoxicity was largely blocked by the presence of 100 /xM N M D A which by itself did not induce neurotoxicity and appeared to slightly enhance the mitochondrial activity. To discriminate between the possible apoptosis and necrosis induced by CBZ, D N A fragmentation was examined. CBZ treatment for 3 days increased the levels of D N A laddering as assessed by electrophoresis through an agarose gel (Fig. 2). The size of the bands forming the ladders were multiples of 180 bp, indicating the occurrence of internucleosomal cleavage which is a major characteristic of apoptosis but not of necrosis [25,32]. Considerable amounts of D N A fragmentation were also detected in untreated cells. The CBZ induced nucleosomal cleavage was completely blocked by NMDA, which also reduced basal DNA fragmentation (Fig. 3). ATA, an inhibitor of Ca2÷,Mg2+-dependent DNase thought to be responsible for internucleosomal cleavage [4,29,35] was also examined for its effect on CBZ-induced
ii
Fig. 6. CBZ induces increasednuclear staining.Cells grown on 12-mmpolylysine-coatedglass covemlipswere pretreated with 100/a,M NMDA or 50/zM ATA for 2 h and then exposed to 100 IzM CBZ for 3 days. Cells were then washed with ice-cold phosphate-bufferedsaline (pH 7.2), fixed with 4% formaldehyde and stained with Hoechst 33258 (5 p,g/ml) for 5 min at 4°C. Cells were washed with distilledwater and visualizedunder UV illumination using a Zeiss Axiophot microscope at 1000 × magnification.A: untreated control; B: CBZ alone; C: NMDA in combinationwith CBZ; D: ATA in combinationwith CBZ. Scale bar represents 10/~m. Note the appearance of brightened 'apoptoticbodies' presumablydue to chromatincondensationafter CBZ treatment and the suppressionof this appearance by NMDA and ATA. The experimenthas been repeated three times with similar results.
68
X.-M. Gao et al. / Brain Research 703 (1995) 63-71
neurotoxicity. ATA effectively prevented CBZ-induced morphological changes that are typified by the appearance of smaller translucent cells, disaggregation of neuronal clumps and fragmentation of neuronal processes (Fig. 4). The protection by ATA was evident at 5/xM and appeared to be complete at 50 and 100 /~M. ATA at these concentrations did not significantly affect the morphology of untreated cells. CBZ-induced DNA fragmentation was blocked by ATA in a concentration-dependent manner (Fig. 5). Basal level of DNA fragmentation in untreated cells was also reduced by ATA. Staining the nuclear chromatin with the fluorescent dye Hoechst 33258 revealed the appearance of 'brightened' nuclei after CBZ treatment (Fig. 6A, B), suggesting that condensation of nuclear chromatin had occurred [31]. This apparent chromatin condensation was almost completely suppressed by the presence of NMDA (Fig. 6C) or ATA (Fig. 6D). Apoptotic cell death is also typified by the requirement of new macromolecular synthesis (for a review, see
[25,32]). In an attempt to evaluate the role of de novo protein synthesis in CBZ-induced cell death, cells were treated with 0 . 2 / z g / m l cycloheximide (CHX) before CBZ exposure. Visual inspection of granule cell morphology revealed that the CBZ-induced cell death was essentially blocked by the presence of CHX which alone showed little or no morphological alteration (Fig. 7). The effect of CHX on CBZ neurotoxicity was further examined by using MTT and [3H]ouabain binding assays. The presence of CHX dose-dependently decreased the mitochondrial activity and the binding of [3H]ouabain to Na+/K+-ATPase in untreated cells, but increased these activities in cells treated with CBZ (Fig. 8A,B). When the M T r results of CBZtreated cells were normalized to the respective control at each concentration of CHX, the cell viability was increased from about 30% to 75% by the presence of 0.2 /xg/ml CHX (Fig. 8A). Using the same method of normalization for control conditions, [3H]ouabain binding results also demonstrated a dose-dependent protection against CBZ
Fig. 7. CHX protects against CBZ-induced cell death: morphological evidence. Cells grown in 24-well plates were pretreated with 0 . 2 / x g / m l CHX for 2 h before exposure to CBZ for 3 days. Cell morphology was examined using a phase contrast microscope, at 200 x magnification. A: untreated control; B: CBZ alone; C: CHX alone; D: CHX in combination with CBZ. Scale bar represents 25 brm. Note the reversal of CBZ-induced cell death by CHX which did not significantly affect the cell morphology in untreated cells. The results are from a typical experiment reproduced five times.
X.-M. Gao et al. /Brain Research 703 (1995) 63-71 Ao
MTT Assay
120 CHX alone
o
100
~
0
CHX ÷ CBZ
o
80
•
laamma...IIDIm
60 m
40 >
20 o
o
0
i
i
i
i
i
0.0
0.1
0.2
0.3
0.4
CHX Concentration (gg/ml)
B. 3H-ouabain Binding 120 loo
O
o ,6
so
g
so
'~
40
~_.
20 0
i
u
i
i
I
0.00
0.05
010
0.15
0.20
CHX Concentration
(pg/ml)
Fig. 8. CHX reduces CBZ-induccd neurotoxicity as determined by mitochondrial activity and [3 H]ouabain binding assays. Cells grown on 24-well plates were pretreated with the indicated concentrations of CHX for 2 h and were then exposed to 100 #,M CBZ or DMSO vehicle for 3 days. Cell viability was determined by using the M'Iq" assay (A) or [3H]ouabain binding assay (B). The data are means + S.E.M. of quadruplicate determinations from a typical experiment and are expressed as % of the untreated control. The dotted lines represem the normalized value when the result of CBZ at each CHX concentration was normalized to the respective control of CHX alone at each point. The 100% values of the untreated controls are 0.738 + 0.015 absorbance unit at 540 nm for the MTT assay and 12904+436 dpm/well for the [3H]ouabain binding assay. The experiment has been repeated seven times with essentially the same results.
neurotoxicity by CHX, further restoring the viability to a level greater than 90% of the value in the corresponding control (Fig. 8B).
4. Discussion
In this study we confima our previous report that CBZ induces neurotoxicity of cerebellar granule cells and that this CBZ toxicity is protected by NMDA. Moreover, we have provided several lines of evidence that this neurotoxicity involves apoptotic cell death: (a) CBZ dose-depen-
69
dently induces DNA fragmentation, indicating CBZ induction of internucleosomal cleavage; (b) both CBZ-induced neurotoxicity and DNA fragmentation can be blocked by NMDA and ATA, an inhibitor of endonuclease; (c) the neurotoxicity is accompanied by the detection of dead cells which have reduced volume and whose nuclei are intensively labeled with Hoechst 33258, suggesting that chromatin condensation has occurred; (d) CBZ-induced cell death is largely suppressed by the presence of CHX in a dose-dependent manner, suggesting an involvement of de novo protein synthesis. A considerable level of internucleosomal DNA cleavage was detected in untreated cells and that this basal DNA fragmentation is also reduced by NMDA and ATA. These results suggest that a moderate level of apoptosis occurs in mature cerebellar granule cells at the time of cell harvest (8-9 DIV) and that this spontaneous apoptosis is exacerbated by CBZ. ATA in a dose-dependent manner blocks CBZ-induced DNA fragmentation and cell death (Figs. 4-6). ATA is a polymeric polyanionic dye composed of a heterogeneous mixture of salicylic acid and methylene-bis-salicylic acid and is known to display pleiotropic effects on cellular activity including nuclease inhibition [4,16,29,35]. If the effect of ATA is attributed to inhibition of DNAse activity, our results suggest that DNA fragmentation is an early event and is required for CBZ-induced apoptosis. This possibility is in contrast to neuronal apoptosis in the nematode C. elegans where cell death precedes corpse phagocytosis and DNA degradation [17]. At relatively high doses, ATA also inhibits NMDA receptor-mediated excitotoxicity in a retina preparation [38]. However, it seems unlikely that NMDA receptor blockade is related to ATA's neuroprotection against CBZ toxicity. Although ATA at 50 /xM partially inhibits NMDA/glycine induced increase in intracellular calcium in cerebellar granule cells, the NMDA receptor antagonists, MK-801 (10 /xM) and aminophosphovaleric acid (100 /xM), did not affect CBZ-induced neurotoxicity (data not shown). Nevertheless, we can not exclude the possibility that the neuroprotective effects of ATA are related to its other actions, e.g., inhibition of nucleic acid polymerase activity or protein synthesis [14,15,37]. In this context, it is of interest to note that ATA has variable effects on excitotoxin-induced cell death in neurons. For example, ATA protects against glutamate toxicity in hippocampal neurons [36], but not the NMDA toxicity in cerebellar granule cells [9]. The protein synthesis inhibitor, CHX, effectively attenuates CBZ neurotoxicity based on morphological examinations and assays for mitochondrial activity with MTI" and binding of [3H]ouabain to Na+/K+-ATPase in the plasma membrane (Figs. 7 and 8). Interestingly, CHX appears to completely reverse neurotoxic effect of CBZ detected by microscopic inspection and [3H]ouabain binding, while only partially protecting against the effect assayed by MTI" cleavage. This discrepancy suggests that the loss of mitochondrial function is an early event associated with CBZ-
70
X.-M. Gao et al. /Brain Research 703 (1995) 63-71
induced death of cerebellar granule cells in culture but that not all mitochondrial activity is essential for cell survival. The sensitivity of CBZ-induced apoptosis to CHX suggests an essential role of de novo protein synthesis; however, the nature of the protein(s) involved is unknown. We have recently found that the spontaneous death of cerebellar granule cells as cultures age is closely correlated with the over-expression of a 38 kDa protein in the particulate fraction [21]. The role of this protein in CBZ-induced granule cell death is being investigated. CBZ neurotoxicity is inhibited by NMDA (Figs. 1, 3 and 6), thus confirming our previous report [12]. We have also shown that NMDA protects against glutamate toxicity in cerebellar granule cells [7]. The mechanisms underlying the protective effects of NMDA are still unclear. It is known that the survival of cerebellar granule cells in culture requires the presence of depolarizing concentrations of KC1 which in turn induces the release of glutamate to activate NMDA receptors [3,11]. We have recently shown that CBZ at a relevant concentration inhibits NMDA receptor-mediated increase of [Ca2÷ ]i in cerebellar granule cells [19]. Thus it is conceivable that CBZ neurotoxicity is due to the inhibition of [Ca2÷ ]i maintained by endogenous glutamate acting on NMDA receptors and that NMDA promotes cell survival by alleviation of the CBZ inhibition of [Ca2+] i in these neurons. Alternatively, the NMDA protection may be related to its ability to induce gene expression of neuroprotective proteins, such as brain-derived neurotrophic factor in cerebellar granule cells [10]. Consistent with this possibility is the report that inhibitors of RNA and protein synthesis block the NMDA protection against glutamate and MPP + toxicity in cerebellar granule cells [26]. Clearly, the mechanism(s) underlying NMDA protection requires further investigation.
References [1] Albright, P.S. and Burnham, W.M., Development of a new pharmacological seizure model: effects of anticonvulsants on corticaland amygdala-kindled seizures in the rat, Epilepsia 21 (1980) 681-689. [2] Ashton, D. and Wauquier, A., Behavioral analysis of the effects of 15 anticonvulsants in the amygdaloid kindled rat, Psychopharmacology, 65 (1979) 7-13. [3] Balazs, R., Hack, N. and Jorgensen, O.S., Stimulation of N-methylt)-aspartate receptor has atrophic effect on differentiating cerebellar granule cells, Neurosci. Lett., 87 (1988) 80-86. [4] Batistatou, A. and Green, L.A., Aurintricarboxylic acid rescues PC12 cells and sympathetic neurons from cell death caused by nerve growth factor deprivation: correlation with suppression of endonuclease activity, J. Cell Biol., 115 (1991) 461-471. [5] Bonduelle, M., Current approaches to the treatment of trigeminal neuralgia. In W. Birkmayer (Ed.), Epileptic Seizures-Behavior Pain, Huber, Bern, 1976, pp. 321-326. [6] Cereghino, J.J., Brock, J.T., Van Meter, J.C., Penry, J.N., Smith, L.D. and White, B.G., Carbamazepine for epilepsy: a controlled prospective evaluation, Neurology, 24 (1974) 401-410. [7] Chuang, D.-M., Gao, X.-M. and Paul, S.M., N-Methyl-D-aspartate
exposure blocks glutamate toxicity in cultured cerebellar granule cells, Mol. Pharmacol., 42 (1993) 210-261. [8] Deshpande, J., Bergstedt, K., Linden, T., Kalimo, H. and Wieloch, T., Ultrastructural changes in the hippocampal CA1 region following transient cerebral ischemia: evidence against programmed cell death, Exp. Brain Res., 88 (1992) 91-105. [9] Dessi, F., Charriaut-Marlangue, C., Khrestchatisky, M. and Ben-Aft, Y., Glutamate-induced neuronal death is not a programmed cell death in cerebellar culture, J. Neurochem., 60 (1993) 1953-1955. [10] Favaron, M., Manev, R.M., Rimland, J.M., Candeo, P., Beccaro, M. and Manev, H., NMDA-stimulated expression of BDNF mRNA in cultured cerebellar granule neurons, Neuroreport, 4 (1993) 11711174. [11] Gallo, V., Kingsburg, A., Balazs, R. and Jorgensen, O.S., The role of depolarization in the survuval and differentiation of cerebellar granule cells in culture, J. Neurosci., 7 (1987) 2203-2213. [12] Gao, X.-M. and Chuang, D.-M., Carbamazepine-induced neurotoxicity and its prevention by NMDA in culture cerebellar granule cells, Neurosci. Lett., 135 (1992) 159-162. [13] Gerson, W.T., Fine, D.G., Spielberg, S.P. and Sensenbrenner, L.L., Anticonvulsant-induced aplastic anemia: increased susceptibility to toxic drug metabolites in vitro, Blood, 61 (1983) 889-893. [14] Givens, J.F. and Manly, K.F., Inhibition of RNA-directed DNA polymerase by aurintricarboxylic acid, Nucleic Acids Res., 3 (1976) 405-418. [15] Grollman, A.P. and Stewart, M.L., Inhibition of the attachment of messenger ribonucleic acid to ribosomes, Proc. Natl. Acad. Sci. USA., 61 (1968) 719-725. [16] HaUick, R.B., Chelm, B.K., Gray, P.W. and Orozco, E.M., Use of aurintricarboxylic acid as an inhibitor of nuclease during nucleic acid isolation, Nucleic Acids Res. 4 (1977) 3055-3064. [17] Hengartner, M.O. and Horvitz, H.R., The ins and outs of programmed cell death during C. elegans development, Phil, Trans. R. Soc. Lond., B 345 (1994) 243-246. [18] Horowitz, S., Patwardham, B. and Marcell, E., Hepatotoxic reactions associated with carbamazepine therapy, Epilepsia, 29 (1988) 149-154. [19] Hough, C., Irwin, R.P., Gao, X.-M., Rogawski, M.A. and Chuang, D.-M., Carbamazepine inhibition of NMDA-stimulated calcium influx in rat cerebellar granule cells, Neurosci. Abstr., 19 (1993) p. 1781. [20] Ishida Y., Agata, Y., Shibahara, K. and Honjo, T., Induced expression of PD-1, a novel member of the immunoglobulin gene superfamily, upon programmed cell death, EMBOJ. 11(1992) 3887-3895. [21] Ishitani, R., Sunaga, K., Hirano, A., Katsube, N. and Chuang, D.-M., A 38-KDa protein is associated with apoptosis in mature cerebellar neurons, Neurosci. Abstr. 20 (1994) p. 689. [22] Kure, S., Tominaga, T., Tada, K. and Narisawa, K., Glutamate triggers internucleosomal DNA cleavage in neuronal cells, Biochem. Biophys. Res. Commun., 179 (1991) 39-45. [23] Lockshin, R.A. and Zakeri, Z. In L.D. Tomei and F.O. Cope (Eds), Apoptosis. The Molecular Basis of Cell Death, 1991, pp. 44-60. [24] Loo, D.T., Copani, A., Pike, C.J., Whittemore, E.R., Walencewicz, A.J. and Cotman, C.W., Apoptosis is induced by B-amyloid in cultured central nervous, Proc. Natl. Acad. Sci. USA, 90 (1993) 7951-7955. [25] Margolis, R.L., Chuang, D.-M. and Post, R.M., Programmed cell death: implication for neuropsychiatric disorders, Biol. Psychiatry, 35 (1994) 946-956. [26] Marini, A.M. and Paul, S.M., N-Methyl-o-aspartate receptor-mediated neuroprotection in cerebellar granule cells requires new RNA and protein synthesis, Proc. Natl. Acad. Sci. USA. 89 (1992) 6555-6559. [27] Markwell, M.A.K., Sheng, H.Z., Brenneman, D.E. and Paul, S.M., A rapid method to 3quantify neurons in mixed cultures based on the specific binding of H-ouabain to neuronal Na+,K+-ATPase, Brain Res., 538 (1991) 1-8.
X.-M. Gao et al. /Brain Research 703 (1995) 63-71 [28] Martin, D.P. Johnson Jr., E.M., Cytosine arabinoside kills postmitotic neurons in a fashion resembling trophic factor deprivation: evidence that a deoxycytidine-dependent process may be required for nerve growth factor signal transduction, J. Neurosci., 10 (1990) 184-193. [29] McConkey, D.J., Hartzell, P., Nicotera, P. and Orrenius, S., Calcium-activated DNA fragmentation kills immature thymocytes, FASEB J., 3 (1989) 1843-1849. [30] Mosmann, T.R., Rapid colourmetric assay for cellular growth and survival: application to proliferation and cytotoxicity assays, J. lmmunol. Meth., 65 (1983) 55-63. [31] Oberhammer, F.A., Pavell~:a, M., Sharma, S., Tiefenbacher, R., Purchio, A.F., Bursch, W. and Schulte-Hermann, R., Induction of apoptosis in cultured hepatocytes and in regressing liver by transforming growth factor beta 1, Proc. Natl. Acad. Sci. USA, 89 (1992) 5408-5412. [32] Oppenheim, R.W., Cell death during development of the nervous system, Ann. Rev. Neurosc/., 14 (1991) 453-501. [33] Post, R.M., Anticonvulsants for the lithium-resistant bipolar patient. In J. Amsterdam (Ed.), Pharmacology of Depression: Application
71
for the Outpatient Practioner, Marcel Dekker, New York, 1990, pp. 129-158. [34] Post, R.M., Weiss, S.R.B., Chuang, D.-M. and Ketter, T.A., Mechanisms of action of carbamazepine in seizure and affective disorders. In R.J. Joffe and J.R. Calabrese (Eds.), Anticonvulsants in Psychiatry, Marcel Dekker, 1994, pp. 43-92. [35] Ribeciro, J.M. and Carson, D.A., Ca2+/MgZ+-dependent endonuclease from human spleen: purification, properties, and role of apoptosis, Biochemistry, 32 (1993) 9129-9136. [36] Samples, S.D. and Dubinsky, J.M., Aurintricarboxylic acid protects hippocampal neurons from glutamate excitotoxicity in vitro, J. Neurochem., 61 (1993) 382-385. [37] Siegelman, F. and Apirion, D., Aurintricarboxylic acid, a preferential inhibitor of initiation of protein synthesis, J. Bacteriol., 105 (1971) 902-905. [38] Zeevalk, G.D., Schoepp, D. and Nicklas, W.J., Aurintricarboxylic acid prevents NMDA-mediated excitotoxicity: evidence for its action as an NMDA receptor antagonist, J. Neurochem., 61 (1993) 386-389.