Magnesium removal induces paroxysmal neuronal firing and NMDA receptor-mediated neuronal degeneration in cortical cultures

Neuroscience Letters, 115 (1990) 313-317 Elsevier Scientific Publishers Ireland Ltd.

313

NSL 07023

Magnesium removal induces paroxysmal neuronal firing and NMDA receptor-mediated neuronal degeneration in cortical cultures K. Rose, C.W. Christine and D.W. Choi Department of Neurology, H-3160, Stanford University Medical Center, Stanford, CA 94305 (U.S.A.) (Received 19 February 1990; Revised version received 4 April 1990; Accepted 6 April 1990)

Key words: Glutamate; Epilepsy; Seizure; Neurotoxicity; Excitatory amino acid; Cortex; Cell culture Removal of extracellular Mg 2+ triggered the onset of repetitive excitatory discharges in cultured murine cortical neurons, detected by recording with patch electrodes in the whole cell configuration. The discharges were suppressed by i00/tM D-2-amino-5-phosphonovalerate. Over the next 24-72 h substantial numbers of neurons, but not glia, degenerated, releasing lactate dehydrogenase to the bathing medium. The neuronal death induced by removal of extracellular Mg 2÷ could be attenuated by either 3/~M tetrodotoxin or 50/tM dextrorphan, and thus likely reflects excessive activation of N-methyl-o-aspartate receptors triggered by excitatory discharges. This Mg 2÷ removal model may be a useful model in which to study certain aspects of epileptic neocortical injury.

Much of the morbidity and even mortality of prolonged seizures may be due to neuronal degeneration, which can be extensive even if hypoxia or systemic acidosis are prevented [11]. While several factors may contribute to this neuronal degeneration, excess release of glutamate or related excitatory amino acids may be a key pathogenetic mechanism. Glutamate neurotoxicity, in large part mediated by the Nmethyl-D-aspartate (NMDA) subtype of glutamate receptors, has been hypothesized to contribute to several types of acute brain injury, including that produced by hypoxia, ischemia, hypoglycemia, and trauma [3,15,19]. Prolonged seizures are a likely setting for excessive release of transmitter glutamate, and NMDA antagonists have been reported to reduce thalamic neuronal loss in rats with cortical seizures [7]. Study of the mechanisms underlying epileptic brain injury in intact systems can be hampered by limited experimental access to neurons, or by the occurrence of associated metabolic derangements. Development of a culture model for such injury might provide a method for isolating some relevant cell-cell interactions, and identi-

Correspondence: D.W. Choi, Department of Neurology, H-3160, Stanford University Medical Center, Stanford, CA 94305, U.S.A. 0304-3940/90/$ 03.50 © 1990 Elsevier Scientific Publishers Ireland Ltd.

314 fying promising approaches for reducing that portion of neuronal damage directly attributable to the excessive excitatory discharges themselves. One effective method for inducing paroxysmal discharges in brain slices is removal of extracellular Mg 2+ [2, 16, 20]. Recently, Miller and colleagues found that removal of Mg 2+ from the extracellular medium induced discharges and neuronal cell degeneration in hippocampal cell cultures [1]. Furthermore, Furshpan and Potter have reported the induction of similar hippocampal neuronal discharges and N M D A receptor-mediated neuronal degeneration by the abrupt termination of chronic glutamate receptor blockade [8]. The purpose of the present study was to examine the effects of Mg 2+ removal on electrical activity and neuronal survival in neocortical cell cultures. An abstract has appeared [18]. Cultures of mixed neocortical neurons and glia were prepared as previously described from El5 El7 fetal mice [4]. Dissociated neocortical cells were plated on modified polystyrene (Falcon Primaria) 15 mm culture wells and 35 mm dishes in Eagle's minimal essential medium (MEM, Earle's salts, supplied glutamine-free) supplemented with glucose (total 21 mM), glutamine (2 mM), 10% equine serum, and 10% fetal bovine serum; this medium contained about 0.8 mM Mg 2+. Non-neuronal cell division was halted after 10-14 days in vitro by 24-72 h exposure to 10/~M cytosine arabinoside. Cultures were selected for study after 14 days in vitro. Intracellular recording at room temperature, using patch electrodes in the whole cell current clamp configuration [10] was used to examine the electrophysiological effects of Mg 2+ removal. The pipette solution contained (in mM) 140 KCI, 10 HEPES, 5 Mg-ATP, 5 EGTA, and 0.5 CaC12 ( p C A = 7.3), titrated to pH 7.3 with NaOH. Large bore pipettes fed by selector valves were used to locally perfuse neurons with the desired experimental solution. Recordings were obtained in a series of 6 cells, with the average membrane potential just after patch rupture - 6 5 . 6 4-4 mV (S.D.), and the cells were perfused with a control solution with the following compo-

Control

Mg + + free

,

APV

~

.~

i

Mg + + free

, 4 0 mV i

1 rnin Fig. 1. M g 2 ~ removaltriggers neuronal cell firing. Penwritertracing of a whole cell current clamp (voltage sensing) recording from a cortical neuron perfused sequentiallywith a physiologicalsalt solution (Control), the same solution without Mg2+ (Mg free), or 50 ,uM D-2-amino-5-phosphonovalerate(o-APV). Holding potential was -65 mV. The penwriter does not faithfullyreflect the full vertical extent of the action potential.

315 sition (in mM): 140 NaCI, 2.8 KC1, 1 CaC12, 10 HEPES, 1 MgC12, and 5 glucose (pH adjusted to 7.3 with NaOH). Five gM glycine was also included in the perfusion solution to maintain N M D A receptor function [12]. Sporadic ongoing synaptic potentials and action potentials were seen (Fig. 1), consistent with previous sharp electrode recordings [4]. When perfusion was changed to one lacking MgC12 (otherwise identical to the control solution), the abrupt onset of repetitive excitatory discharges was observed (6/6 cells) (Fig. 1), often settling into a periodic pattern. Reintroduction of the Mg2+-containing control solution, or another solution lacking Mg 2+ but containing 50/~M o-2-amino-5-phosphonovalerate (D-APV) reversibly attenuated neuronal firing (Fig. 1); in 3/5 cells o-APV completely blocked firing. These experiments suggested that Mg 2+ removal was effective in eliciting repetitive excitatory discharges in our cultures, as found in other in vitro systems. An important basis for this effect was likely unblocking of the N-methyl-D-aspartate (NMDA) receptor-activated channel [14, 17] leading to enhancement of synaptic excitation. However, Mg 2+ removal may also have other important pro-excitatory effects, including augmentation of quantal transmitter release, or general facilitation of inward current flow due to reduced membrane surface screening charge. The second part of the present study examined the effects of Mg 2+ removal on neuronal cell survival. Mg 2+ removal was accomplished by thorough exchange of the growth medium for a defined serum-free solution with the following composition (mM); NaC1 116, KCI 5.4, CaC12 1.8, glucose 21, NaHCO3 26, NaHPO4 1; procedure control cultures were handled the same, but 0.8 mM MgSO4 was added back (yielding the salt composition of Earle's Balanced Salt Solution). Cultures were then returned to the 37°C incubator. Twenty-four h later, morphological examination by phase-contrast microscopy

Fig. 2. Mg2+ deprivationneuronalinjury:morphology.Photomicrographsof representativefieldsin cortical cell cultures, taken with phase contrast, following72 h incubationin these solutions:(A) physiological salt solution (procedure control); (B) the same but with removalof Mg2+;(C) Mg2+ removedbut 50/aM dextrorphan added; (D) Mg2+ removedbut 3/zM tetrodotoxinadded. Bar = 100/zm.

316

and Trypan blue dye exclusion showed some minor neuronal damage in the Mg 2+deprived cultures that was not found in the procedure control cultures. By 48-72 h, substantial numbers of neurons but not gtia in the MgZ+-deprived cultures had degenerated, whereas control cultures appeared largely intact (Fig. 2). Media lactate dehydrogenase (LDH), an index previously found to be useful in quantitating cortical cell injury [9, 13], was consistent with the morphological appearance; at 48 h, the LDH signal was about half of the near maximal level found in cultures exposed for the same time to 40/iM NMDA, an exposure which selectively destroyed almost all the neurons (3 experiments) (Fig. 3). The remaining media glucose at 72 h was measured in one experiment and found to be > 10 mM, excluding glucose deprivation as a cause of injury. However, if either 3/~M tetrodotoxin, or a 50/~M concentration of the NMDA antagonist dextrorphan [5, 6] was included in the MgZ+-deprivation solution, then both morphological (Fig. 2) and chemical (Fig. 3) indications of neuronal damage were substantially attenuated. Other experiments verified that the tested exposures to dextrorphan or tetrodotoxin lacked intrinsic neurotoxicity. The present study thus suggests that the removal of extracellular Mg 2+ is sufficient to induce both repetitive excitatory discharges and NMDA receptor-mediated degeneration in cultured cortical neuronal cells. These findings are consistent with observations of Miller and colleagues [1], and Furshpan and Potter [8], on cultured hippocampal neurons. The increased neuronal firing triggered by Mg 2+ removal appears to have been critically involved in resultant neuronal death, as tetrodotoxin markedly reduced this death. Most likely, the discharges were directly responsible for the excesMagnesium deprivation injury-48 Hr 120 100

p

0.05

vs -Mg

__L_

I:D

z

80

w

~ .<

60

w

40

a: T

9

2o I

NMDA

-Mg

r

-Mg/DX

I

"

I

-Mg/TTX

Fig. 3. Mg 2+ deprivation neuronal injury: lactate dehydrogenase (LDH) ettlux. Bars show LDH + SEM (n = 445 cultures at each condition) present in the bathing medium of sister cultures, following 48 h incubation in these conditions: 40 pM N-methyl-D-aspartate (NMDA), physiological salt solution with Mg 2+ removed ( - M g ) , Mg ~÷ removed but 50/tM dextrorphan added, Mg2+ removed but 3 pM tetrodotoxin added. The relatively small amount of LDH found in procedure control cultures was subtracted from all values to yield the signal specifically associated with Mg 2÷ removal, and values are expressed relative to that found in the NMDA-exposed condition (corresponding to near complete neuronal cell loss). Indicated statistical comparisons are by two-tailed t-test with Bonferroni correction for two comparisons.

317 sive synaptic release of t r a n s m i t t e r g l u t a m a t e from excitatory nerve terminals, inducing overactivation of postsynaptic N M D A receptors. This overactivation was likely e n h a n c e d by reduced M g 2+ blockade o f the N M D A receptor-activated channel. However, present observations c a n n o t exclude the alternative possibility that increased n e u r o n a l firing was necessary for n e u r o n a l death n o t because o f linkage to g l u t a m a t e release from presynaptic terminals, b u t rather because o f u n f a v o r a b l e postsynaptic implications, such as o n energy metabolism. F u r t h e r study of the m e c h a n i s m s u n d e r l y i n g n e u r o n a l loss in the present model m a y facilitate the developm e n t o f therapeutic approaches effective against epileptic b r a i n d a m a g e in vivo. This study was supported by N I H G r a n t NS26907 a n d by a g r a n t from the American Paralysis Association. 1 Abele, A.E., Scholz, W.K., Scholz, K.P. and Miller, R.J., Increased excitatory neurotransmission resuits in the death of cultured rat hippocampal pyramidal neurons, Soc. Neurosci. Abstr., (1989) 480. 2 Anderson, W. W., Lewis D.V., Swartzwelder, H.S. and Wilson, W.A., Magnesium-free medium activates seizure-likeevents in the rat hippocampal slice, Brain Res., 398 (1986) 215-219. 3 Choi, D.W., Glutamate neurotoxicity and diseases of the nervous system, Neuron, I (1988) 623~24. 4 Choi, D.W., Maulucci-Gedde, M.A. and Kriegstein, A.R., Glutamate neurotoxicity in cortical cell culture, J. Neurosci., 7 (1987) 357-368. 5 Choi, D.W., Peters, S., Viseskul,V., Dextrorphan and levorphanol selectivelyblock N-methyl-o-aspartate receptor mediated neurotoxicity on cortical neurons, J. Pharmacol. Exp. Ther., 242 (1987) 713720. 6 Church, J., Lodge, D. and Berry, S.C., Differential effects of dextrorphan and levorphanol on the excitation of rat spinal neurons by amino acids, Eur. J. Pharmacol., 111 (1985) 185-190. 7 Clifford DB, Zorumski CF, Olney JW, Ketamine and MK-801 prevent degeneration of thalamic neurons induced by focal cortical seizures, Exp. Neurol., 105 (1989) 272-279. 8 Furshpan, E.J. and Potter, D.D., Seizure-likeactivity and cellular damage in rat hippocampal neurons in cell culture, Neuron, 3 (1989) 199-207. 9 Goldberg, M.P., Weiss, J.W., Pham, P.C. and Choi, D.W., N-methyl-D-aspartate receptors mediate hypoxic neuronal injury in cortical culture, J. Pharmacol. Exp. Ther., 243 (1987) 784-791. 10 Hamill, O.P., Marty, A., Neher, E., Sakmann, B. and Sigworth, F.J., Improved patch-clamp techniques for high-resolution current recording from cellsand cell-freemembrane patches, Pflugers Arch., 391 (1981) 85-100. 11 Hauser, W.A., Status epilepticus: frequency, etiology, and neuropathological sequelae, Adv. Neurol., 34 (1983) 3-14. 12 Johnson, J.W. and Ascher, P., Glycine potentiates the NMDA response in cultured mouse brain neurons, Nature, 325 (1987) 529-531. 13 Koh, J. and Choi, D.W., Quantitative determination of glutamate mediated cortical neuronal injury in cell culture by lactate dehydrogenase efflux assay, J. Neurosci. Methods, 20 (1987) 83-90. 14 Mayer M.L., Westbrook G.L. and Guthrie, P.B., Voltage dependent-block by Mg÷+ of NMDA responses in spinal cord neurons, Nature, 309 (1984) 261-263. 15 Meldrum, B., Possible therapeutic applications of antagonists of excitatory amino acid neurotransmitters, Clin. Sci., 68 (1985) 113-122. 16 Modi, I., Lambert, J.D.C. and Heinemann, U., Low extracellular magnesium induces epileptiform activity and spreading depression in rat hippocampal slices, J. Neurophysiol., 57 (1987) 869-888. 17 Nowak, L., Bregestovski, P., Ascher, P., Herbert, A. and Prochiantz, A., Magnesium gates glutamate activated channels in mouse central neurons, Nature, 307 (1984) 462-465. 18 Rose, K., Christine, C.W. and Choi, D.W., Magnesium removal induces hyperactivity and neuronal degeneration in vitro, Soc. Neurosci. Abstr., (1989) 701. 19 Rothman, S.M. and Olney, J.W., Excitotoxicityand the NMDA receptor, Trends Neurosci., 10 (1987) 299-302. 20 Thompson, A.M. and West, D.C., N-Methylaspartate receptors mediate epileptiform activity evoked in some but not all conditions in rat neocortical slices, Neuroscience, 19 (1986) 1161-1177.