98

;V~rosci~~e Lrrrers, I26 (1991) 98-101 :(’ 1991 Elsevier Scientific Publishers Ireland Ltd. 0304-3940/91/$03.50 ADONIS0304394091002040

NSL 07722

Ganglioside GM 1 prevents death induced by excessive excitatory neurotransmission in cultured hippocampal pyramidal neurons Stephen D. Skaper, Alberta Leon and Laura Facci Fidia Research Laboratories, Abano Terme (Italy) (Received Key word&

Excitotoxicity;

Rat hippocampal degeneration

10 December

Pyramidal

pyramidal

neurons

neuron;

1990; Revised version received 29 January Glutamate;

in culture,

over the next 24 h. This neuronal

methyl-D-aspartate

(NMDA)

Ganglioside

which has recently

GMl,

rons associated

receptors,

Ganglioside;

exposed

to result from excitatory

as cell death could be blocked

with the above phenomenon.

to protect

Gangliosides

against

Correspondence: SD. Skaper, della Fabbrica,

Fidia Research

3/A, 35031 - Abano

Laboratories,

Terme, Italy.

by tetrodotoxin excitotoxin-induced

may be a novel therapeutic

Glutamate, a major excitatory transmitter in the mammalian central nervous system (CNS), can kill nerve cells under certain circumstances. The neurotoxic effect of glutamate has been described in energy crisis situations, such as ischemia and hypoglycemia [3,25]. Excitatory amino acids (EAAs) may also play a role in the pathogenesis of neuronal injury from status epilepticus and chronic epilepsy. Local injection of EAA receptor agonists and stimulation of EAA afferent pathways produce a pattern of acute and chronic morphological changes in the brain similar to those found in patients with epilepsy [20, 23, 281. Antagonists of EAA receptors possess anticonvulsant properties and prevent the morphological changes associated with seizures [6, 81. Enhanced excitatory synaptic neurotransmission can also damage hippocampal neurons in culture via activation of N-methyl-D-aspartate (NMDA) receptors [ 1, 111. The use of EAA receptor antagonists as a potential therapy in these disorders may be limited, owing to their adverse effects on EAA-dependent synaptic plasticity [17] and induction of acute neuropathological changes [21]. A novel pharmacotherapeutic approach to limit EAA neurotoxicity derives from the recent observation that gangliosides, a class of naturally occurring, sialoglycosphingolipids [29], limit both exogenous [lo, 271 and endogenous [9,26] EAA cytotoxicity in cultured cerebel-

Via Ponte

4 February

1991)

Neuroprotection

30 min to Mg *+-free, glycine-supplemented

injury appeared

been described

1991; Accepted

synaptic and NMDA, damage,

medium

transmission

undergo

a selective (about

and subsequent

but not non-NMDA, also prevented

tool for brain injury associated

activation

receptor

35%) of N-

antagonists,

the death of hippocampal with epileptic-like

neu-

activity.

lar granule cells without affecting glutamate binding [lo], EAA receptor-operated cation channel opening [lo, 181 or second messenger events [lo]. The ability of these compounds to act solely against the consequences of paroxysmal EAA receptor occupation has been termed receptor abuse-dependent antagonism (RADA) [16]. The present series of experiments demonstrate that gangliosides also block cytotoxicity associated with enhanced excitatory neurotransmission in cultured hippocampal neurons. Hippocampi were removed from embryonic day 17.518 rat brains (Sprague-Dawley), incubated in Ca2+, salt solution for 10 min Mg 2+-free Hank’s balanced (37”C), then in 0.08% trypsin (1:250 hog pancreas, ICN) in this same solution for 30 min (37°C). The tissue was washed twice in complete culture medium and dissociated in this same medium. Culture medium consisted of Dulbecco’s minimal essential medium and Ham’s F 12 (1:l) supplemented with 2 mM glutamine, 100 III/ml penicillin, 27.5 mM glucose, 25 mM KCl, 10% heat-inactivated fetal calf serum (Seromed) and the hormones described by Clos and Gabrion [7]. A suspension containing 2.5-3 x IO5 cells in 1 ml was plated in Falcon multiwell dishes (16 mm diameter), previously treated with 10 ,ug/ml poly-r_-lysine (mol.wt. 70-100.000, Sigma). The cultures were maintained at 37°C in a humidified 5% C@-95% air atmosphere. Cytosine arabinoside (5 PM) was added after 5-7 days to inhibit non-neuronal cell growth. Medium glucose was renewed every 7 days by adding fresh glucose to 5 mM. Cultures were used at 1417 days without medium changes.

99

For neurotoxicity assays, cultures were washed 3 times with Locke’s solution [lo, 271 with or without 1.0 mM MgC12. The Mg *+-free buffer contained 1 ,uM glytine. The various drug treatments were then initiated and incubation continued for 30 min at 23-25°C. Following treatments the cells were washed 3 times with complete Locke’s solution and returned to their original culture medium. Twenty-four hours later, cultures were fixed with 2% glutaraldehyde. Neuronal death was assessed before and after treatment by microscopic examination of representative (but not necessarily the same) fields under phase contrast optics at 200 x . Cytotoxicity was evidenced by neurons which underwent disintegration in the 24-h interval following the initial treatment (Fig. 1). Cell counts were made blindly from at least 6 such fields in triplicate wells in 3 independent experiments (50&600 cells in each condition). Statistical analyses were performed by ANOVA followed by Duncan’s post hoc test. Neurons prepared from this embryonic age of the rat hippocampal area are predominantly pyramidal neurons, with granule cells of the dentate gyrus generated mostly postnatally [4]. At 2 weeks in vitro, large, tripolar or multipolar neurons (Fig. la,a’) comprised approximately 85% of the neurons present, the remaining 15% being smaller, multipolar phase-bright refractile cells. Nonneuronal elements (mainly polygonal-shaped astroglia) accounted for about 20% of the total cellular population.

Fig. 1. Hippocampal pyramidal neurons in culture (a,a’). Following incubation in a Mg*+-free/glycine-supplemented solution for 30 min (24”C), the neurons became dark and swollen (b,b’). Twenty-four hours later, such neurons had disintegrated (c,c’). Bar = 25 pm.

When cells were treated with Mg*+-free, glycine-supplemented solution for 30 min, a selective neuronal death was observed 24 h later. About one in 3 pyramidal neurons developed a darkened and swollen cell body immediately after the 30 min incubation (Fig. lb,b’). These neurons began to undergo degeneration over the next several hours, with cells taking on a granular appearance. By the next day, such neurons had undergone a complete disintegration (Fig. lc,c’). This cell loss is presented quantitatively in Fig. 2, and shows that 33 f 9% of the hippocampal neurons were lost in Mg2+free, glycine-supplemented medium. A comparable level of cell death was reported in another study [l] using similar culture conditions. The minor population of smaller neurons was apparently not affected (not shown). Neuronal death under these conditions very likely is a consequence of enhanced excitatory neurotransmission in a Mg*+-free environment [19], as tetrodotoxin (1 ,uM) protected the pyramidal cells (Fig. 2) [l]. Activation of NMDA receptors appeared to be a key ingredient of this injury, as the NMDA antagonist 3-( f )-2-carboxypiperazine-4-yl)-propyl-1-phosphonic acid (CPP) abolished cell death, whereas the quisqualate/kainate receptor antagonist 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX) had no effect (Fig. 2). Monosialogangliosides, like GM 1, have been reported to reduce EAA neurotoxicity in vitro [9, 10, 26, 271 and in vivo [13, 141. It was therefore of interest to examine if GM1 could protect against excitotoxicity in the present paradigm. Hippocampal cultures were first treated with l&200 PM GM1 ganglioside for 2 h, followed by serum washes and exposure to Mg*+-free, glycine-supplemented medium. This protocol provides for maximal protection against EAA neurotoxicity in vitro [9, 10, 26, 271. As Fig. 3 shows, GM1 attenuated cell death in a concentration-dependent fashion (EDSo x 40 ,uM). Asialo GM 1, which lacks sialic acid, was inactive. Treatment of the hippocampal neurons with GM1 atter the - Mg*+/ + glycine incubation period was not protective (not shown). In contrast to the selective neuronal loss of hippocampal neurons observed upon Mg2+ removal, acute application of glutamate to these same cells will destroy virtually all pyramidal neurons (not shown), suggesting that the fate of a neuron may be determined by its synaptic maturation. The apparent dependence of this former cell death on excitatory synaptic transmission and subsequent activation of NMDA receptors [l, 1l] may provide a useful model to study certain aspects of epileptic hippocampal injury and potential protective drugs. Ganglioside GM1 was able to reduce the cell death triggered by Mg*+-free, glycine-supplemented medium, consistent with previous reports describing the ability of

too

-Mg ?t/GLY

-Mg2+/GLY

GM1 (10)

-1 I

0

10

30

20

40

0

50

IO

20 30 % NEURONAL LOS.5

% NEURONAL LOSS

Fig. 2. Hippocampal

pyramidal

cell death 24 h after 30 min exposure

to a Mgr+-free/glycine-supplemented ments. TTX, tetrodotoxin represent

the mean

solution

with the indicated

(1 pM); CPP, 10 PM; CNQX,

treat-

5 PM. Data

pocampal

GM1 to protect against EAA neurotoxicity in vitro [9, 10, 26, 271 and in vivo [13, 141.The mechanism underlying these neuroprotective effects of the ganglioside is not completely understood, although GM 1-treated cerebellar granule cells show improved recovery from a destabilized Ca*+ homeostasis after glutamate removal [15, 181.This latter effect may be due, in part, to ganglioside inhibition of a glutamate-induced translocation of protein kinase C [30]. As perfusion of pyramidal neurons with Mg *+-free, glycine-supplemented medium also induces a destabilization of intracellular Ca*+ homeostasis [l] which may kill neurons [5] GM 1 could, conceivably, influence cytoplasmic Ca* + in this case. GM1 does not possess anti-convulsant activity (unpublished observations), in keeping with its lack of effect on EAA receptor ion channel opening [ 10, 181. Since gangliosides improve normal synaptic transmission in hippocampal slices [22], the ganglioside effects observed in the present study are not likely to result from a reduced membrane current or transmitter release. Rather, these data reinforce the concept of a ganglioside action targeted to the pathological consequences of EAA activity. Administered monosialogangliosides are beneficial in various animal models of cerebral ischemia [12, 231 and in acute human stroke [2]. The observations described here suggest that these compounds could offer a novel therapeutic perspective in the prevention of nervous tissue damage induced by epilepsy and associated disorders.

Numbers

GM1

by Mg2+-

Cultures

of hip-

for 2 h with the indicated

washed

and incubated

was prepared

of surviving

50

death induced

( -Mg2+/Gly).

cells were treated

conditions.

99% purity).

neuronal

conditions

(in PM) of GMI,

-Mg2+/Gly

alone.

GM1 prevents

pyramidal

centrations

*P < 0.01from Mgr+-free/+glycine

and SD.

Fig. 3. Ganglioside

free/glycine-supplemented

40

con-

30 min (24°C) in

from

bovine

brain

(>

cells were assessed 24 h later. Data

the mean and S.D. *P i 0.05or ** P < 0.01from -Mg’+!

represent

Gly alone.

Senin,

U. and

Fieschi,

C., GM1

ganglioside

therapy

in acute

ischemic stroke, Stroke, 20 (1989) 1143--l 149. 3 Auer, R.N. and Siesjo, B.K., Biological ia, hypoglycemia, 4 Banker,

and epilepsy,

G.A. and Cowan,

persed cell culture,

between ischem-

24 (1988) 6999707.

W.M., Rat hippocampal

neurons

in dis-

Brain Res., I76 (1977) 397425.

5 Choi, D.W., Calcium-mediated fic channel

differences

Ann. Neural.,

neurotoxicity:

types and role in ischemic

relationship

damage,

to speci-

Trends

Neurosci..

I1 (1988) 465469. 6 Clifford,

D.B.,

MK-801

Zorumski,

prevent

focal cortical tion promotes dissociated

and Olney,

postnatally,

action

of excitatory

Neurochem.

899-90

1.

in vitro: involvement uation

B.S., Anticonvulsant Science,

S.D., Hypoglycemic

of excitatory

by monosialoganglioside

neurons

Res., 14 (1989) 919-925.

acid antagonists,

9 Facci, L., Leon, A. and Skaper,

by

interac-

of hippocampal

J.F. and Meldrum,

amino

and

induced

hormone-vasopressin

and maturation

M.J., Collins,

Ketamine

neurons

105 (1989) 272-279.

J., A thyroid

survival

J.W..

of thalamic

seizures, Exp. Neural.,

7 Clos, J. and Gabrion,

8 Croucher,

C.F.

degeneration

216 (1982)

neurotoxicity

amino acid receptors

and atten-

GM 1, Neuroscience,

37

(I 990)

709-716. 10 Favaron,

M., Manev,

dotti, A. and Costa, nate neurotoxicity

H., Alho, H., Bertolino, E., Gangliosides

in primary

neuronal

M., Ferret,

B., Gui-

prevent

glutamate

and kai-

cultures

of neonatal

ebellum and cortex, Proc. Natl. Acad. Sci. U.S.A., 7355. II Furshpan, damage

E.J. and Potter,

D.D., Seizure-like

in rat hippocampal

neurons

rat cer-

85 (1988) 7351.m

activity

and cellular

in cell culture,

Neuron,

3

(1989) 199-207. 12 Karpiak, reduction

SE.,

Mahadik,

of ischemic

S.P. and

injury,

Wakade,

C.G.,

Ganghoside

CRC Crit. Rev. Neurobiol.,

5 (1990)

221-237. Abele, A.E., Scholz, K.P., Scholz, W.K. and Miller, R.J., Excitotoxicity

induced

by enhanced

tured hippocampal Argentino,

C., Sacchetti,

gelo, E., Erminio, Gambi,

pyramidal

D., Mamoli,

excitatory neurons,

M.L.,

F., Federico,

Toni,

neurotransmission Neuron,

A., Ottonello,

4 (1990) 4134

D., Savoini,

F., Ferro

Milone.

G.A., Ponari,

in cul19.

G., D’Arcan-

13 Lipartiti,

M., Mazzari,

S., Lazzaro,

and Leon, A., Monosialogangliosides in neonatal 14 Lombardi, GM

rats, Sot. Neurosci. G., Zanoni,

1 ganglioside

A., Zanoni,

Abstr.,

F., Gallai,

V.,

with

O., Rebucci,

G.,

lesions in the rat. Eur. J. Pharmacal.,

neurotoxicity

15(1989) 764.

R. and Moroni, reduce

R., Seren, M.S.

reduce NMDA

quinolinic

F., Systemic

treatments

acid-induced

174 (1989) 1233125.

striatal

101 15 Manev, H., Favaron, M., Guidotti, A. and Costa, E., Delayed increase of Ca2+ influx elicited by glutamate: role in neuronal death, Mol. Pharmacol., 36 (1989) 106112. 16 Manev, H., Costa, E., Wroblewski, J.T. and Guidotti, A., Abusive stimulation of excitatory amino acid receptors: a strategy to limit neurotoxicity, FASEB J., 4 (1990) 2789-2797. 17 McDonald, J.W. and Johnston, M.V., Physiological and pathophysiological roles of excitatory amino acids during central nervous system development, Brain Res. Rev., 15 (1990) 41-70. 18 Milani, D., Guidolin, D., Facci, L., Pozzan, T., Buso, M., Leon, A. and Skaper, SD., Excitatory amino acid-induced alterations of cytoplasmic free Ca2+ in individual cerebellar granule neurons: role in neurotoxicity, J. Neurosci. Res., in press. 19 Mody, I., Lambert, J.D.C. and Heinemann, U., Low extracellular magnesium induces epileptiforrn activity and spreading depression in rat hippocampal slices, J. Neurophysiol., 57 (1987) 869-888. 20 Olney, J.W., Fuller, T. and De Gubareff, T., Acute dendrotoxic changes in the hippocampus of kainate-treated rats, Brain Res., 176 (1979) 91-100: 21 Olney, J.W., Labruyere, J. and Price, M.T., Pathological changes induced in cerebrocortical neurons by phencyclidine and related drugs, Science, 244 (1989) 136&1362. 22 Ramirez, O.A., Gomez, R.A. and Carrer, H.F., Gangliosides improve synaptic transmission in dentate gyrus of hippocampal rat slices, Brain Res., 506 (1990) 291.-293. 23 Scheibel, M.E., Crandall, P.H. and Scheibel, A.B., The hippocam-

pal-dentate complex in temporal lobe epilepsy; a Golgi study, Epilepsia, 15 (1974) 55-80. 24 Seren, M.S., Rubini, R., Lazzaro, A., Zanoni, R., Fiori, M.G. and Leon, A., Protective effects of a monosialoganglioside derivative following transitory forebrain ischemia in rats, Stroke, 21 (1990) 1607-1612. 25 Siesjii, B.K. and Bengtsson, F., Calcium fluxes, calcium antagonists, and calcium-related pathology in brain ischemia, hypoglycemia and spreading depression: a unifying hypothesis, J. Cereb. Blood Flow Metab., 9 (1989) 127-140. 26 Skaper, S.D., Facci, L., Milani, D. and Leon, A., Monosialoganglioside GM1 protects against anoxia-induced neuronal death in vitro, Exp. Neural., 106 (1989) 297-305. 27 Skaper, S.D., Facci, L. and Leon, A., Gangliosides attenuate the delayed neurotoxicity of aspartic acid in vitro, Neurosci. Lett., 117 (1990) 154159. 28 Sloviter, R.S., ‘Epileptic’ brain damage in rats induced by sustained electrical stimulation of the perforant path. I. Acute electrophysiological and light microscopic studies, Brain Res. Bull., 10 (1983) 675697. 29 Svennerholm, L., Chromatographic separation of human brain gangliosides, J. Neurochem., IO (1963) 613623. 30 Vaccarino, F., Guidotti, A. and Costa, E., Ganglioside inhibition of glutamate-mediated protein kinase C translocation in primary cultures of cerebellar neurons, Proc. Natl. Acad. Sci. U.S.A., 84 (1987) 8707-8711.