Excitotoxicity in the Embryonic Chck Spinal Cord Gregory R. Stewart, PhD,* John W. Olney, MD,? Maya Pathikonda, MD,S and William D. Snider, MDS

Recent evidence implicates excitatory amino acids (EAAs), acting as excitotoxic agents, in the pathogenesis of neurological disorders involving the spinal cord. In this study, we used the chick embryo spinal cord as an in vitro model for studying the sensitivity of spinal neurons to the excitotoxic effects of EAA agonists. Compounds tested include the prototypic receptor-specific agonists, N-methyla-aspartate (NMDA), quisqualic acid (Quis), and kainic acid (KA), and the plant-derived excitotoxic food poisons, P-N-oxalylamino-L-alanine, P-N-methylamino-L-alanine, and domoic acid. Each agonist induced concentration-dependent acute degeneration of neurons distributed throughout the spinal cord. These cytopathological changes consisted of acute edematous degeneration of dendrosomal structures in the dorsal horn and intermediate zone, and dark cell changes with intracytoplasmic vacuolization of motor neurons; this damage is identical to that induced by excitotoxin agonists in other regions of the central nervous system. The NMDA receptor-specific antagonist MK-801 completely blocked toxicity of NMDA, and the nonNMDA antagonist CNQX preferentially blocked the toxicity of Quis- and KA-type agonists in the spinal cord. Our findings suggest that (1)the majority of spinal neurons have all three subtypes of EAA receptors, making them acutely vulnerable to excitotoxin exposure; and (2) EAA antagonists are effective in preventing excitotoxin-induced damage of the spinal cord. Stewart GR, Olney JW, Pathikonda M, Snider WD. Excitotoxicity in the embryonic chick spinal cord. Ann Neurol 1991;30:758-766

The excitatory amino acids (EAAs) glutamate (Glu) and aspartate, with their structural analogues, are of considerable neurobiological interest both as excitatory neurotransmitter candidates and as neurotoxins (excitotoxins). To date, studies exploring the neurotoxic actions of EAAs in the central nervous system (CNS) have focused primarily on their effects in the brain or retina El, 21. Comparatively little attention has been directed toward the excitotoxic potential of EAAs within the spinal cord, despite three decades of electrophysiological experiments identifying EAAs as potent excitants of spinal neurons C3, 41. More recent studies have shown that direct or intrathecal infusion of certain EAA agonists destroys spinal neurons near the site of injection [S-81 and that EAA antagonists may protect against the neurobehavioral deficits associated with spinal cord trauma [9, lo]. Several studies have implicated excitotoxins in a variety of neurological disorders, including human motor neuron disease { 11- 151. Neurolathyrism, a degenerative disorder of the corticospinal tracts that results in spastic paraparesis, occurs endemically in certain parts of the world where the legume Lutbyrus ~atzvuris ingested in excess during periods of famine [lb]. The causative agent in L. sativzls has been identified as (3-Noxalylamino-L-alanine (BOAA o r ODAP), an acidic

amino acid that has excitotoxic properties 117) and produces neurolathyric symptoms when fed chronically to monkeys [IS]. Another plant-derived excitotoxin, (3N-methylamino-L-alanine (BMAA), a constituent of Cyca crrcinalis, has been implicated in an amyotrophic lateral sclerosis (ALS)-like disease in certain Pacific island cultures where these cycad seeds are used both medicinally and in foods [ 191. Following chronic ingestion of a diet enriched in BMAA, monkeys developed spasticity accompanied by pathomorphological changes in corticospinal and motor neurons [203. Finally, there is evidence of altered blood o r cerebrospinal fluid levels of Glu and aspartate in patients suffering from ALS {21-23. Thus, clinical and experimental evidence points to a potentially significant role for endogenous and exogenous EAAs in motor neuron diseases although little is known about the excitotoxic properties of such agents in the spinal cord. In particular, almost nothing is known about the relative sensitivity of different cell types in the spinal cord to various excitotoxic agonists. In this study, we describe the use of the embryonic chick spinal cord in an acute, in vitro preparation to assay excitotoxic activity. We have found the chick embryo to be a very suitable tissue source for studies of this nature because the spinal cord develops preco-

From the "Laboratory of Neurophysiology, National Institute of Mental Health, Poolesvillc, MD; and the Departments of tPsychiaand $Neurology, Washingcon University School of Medicine, St Louis, MO.

Received Feb 7 , 1991, and in revised form May 6. Accepted for publication May 14, 1991. Address correspondence to D~ stewart, NIMH/NIH ~ ~ten- i ter, PO Box 289, Poolesville, MD 20837.

758 Copyright Q 1991 by the American Neurological Association

~

d

ciously, bears a close homology to mammals, and may be removed at a relatively mature state with Iitde artifactual damage due to preparative procedures. Using this model, we have evaluated the prototypic agonists, N - m e t h y b a s p a r t a t e (NMDA), quisqualate (Quis), and kainic acid (KA), each of which interacts preferentially with a subclass of EAA receptor bearing its name. In addition, we have examined the putative environmental excitotoxins BOAA, BMAA [19], and domoic acid (Dom), the amnestic shellfish toxin {24-261 and potent kainate agonist {27]. Here we describe the pattern and type of cytopathological reaction observed in the chick spinal cord after incubation with these agents, their relative potency, and the protective action afforded by receptor-specific antagonists.

Materials and Methods Fertilized chicken eggs were obtained from Spafas, Inc (Norwich, CT). Embryos were removed from their shells after 11 days of incubation (the first day of incubation counted as day 1). Each embryo was decapitated, then eviscerated with the body pinned out on its dorsal surface. The vertebral column and spinal cord were transected at the midthoracic and lower lumbosacral levels. Using fine forceps, the lateral aspect of the vertebral column was carefully incised on each side, and the ventral portion of the vertebral column was retracted. The underlying spinal cord was then gently removed and transferred to an incubation vial. Average dissection time was 4 minutes. Each vial contained 2 ml of a buffered balanced salt solution to which varying concentrations of EAA agonists, antagonists, or both had been added. The balanced salt solution contained (in mM): 1.20 calcium chloride, 5.37 potassium chloride, 0.44 potassium phosphate monobasic, 0.9 magnesium chloride, 123 sodium chloride, 24 sodium bicarbonate, and 5.55 glucose in 5 MOPS buffer (pH, 7.4). Vials were briefly oxygenated (95% 0 2 / 5 % CO,), tightly capped, then placed into an agitating water bath at 37°C and incubated under these conditions for 90 minutes. The p H of test solutions was measured for each vial prior to addition of tissue and after the 90-minute incubation period. Across several experiments, the average initial p H was 7.44 2 0.05 (mean k standard deviation), and the average final p H was 7.45 -t 0.06; this change was not statistically significant when evaluated with a paired Student's t test: t(84) = - 1.439; p > 0.05. At the end of the 90-minute incubation period the medium was removed and the tissue fixed overnight in phosphate-buffered 1.5% glutaraldehyde and 1% paraformaldehyde at 4°C. O n the following day, spinal cords were washed in buffer, placed in 1% buffered osmium for 2 hours, washed in distilled water, dehydrated through an ascending ethanol series, cleared in toluene, then polymerized in Araldite resin (EM Sciences, Fort Washington, PA) for 18 hours at 60°C. For light microscopic examination, semithin (1 pm) sections were cut from spinal cord blocks with the aid of an ultramicrotome, then stained with methylene blue/azure 11. Thin sections (silver refraction color) were cut from the same block face to permit ultrastructural examination of areas of special interest. These sections were floated onto formvar-

coated slot grids, stained with lead citrate and uranyl wetate, then examined with aJEOL JEM- lOOB transmission electron microscope. Pathomorphology resulting from incubation with an EAA was rated on a %point graded scale of damage: 1 = normal intact structure; 2 = pericellular edema, but no distinct cellular damage; 3 = dilations present in neuropil (neuronal somata largely unaffected); 4 = dilations in neuropil, degenerative changes in neuronal somata; and 5 = same damage as grade 4, but more severe and widespread. For each EAA agonist tested, the excitotoxic threshold was defined as the lowest concentration necessary to produce consistently a rating of 4 to 5. Agonists used in the present study were NMDA, Quis, amino-3-hydroxy-5methylisoxazole-4-proprionic acid (AMPA), and KA (Sigma Chemical Co, St Louis, MO); BOAA and BMAA (Cambridge Research Biochemicals, Valley Stream, NY); and Dom (DiagnosticChemicals Limited, Monroe, CT). For each EAA antagonist, a protective threshold was defined as the lowest concentration needed to reduce damage of an EAA (coincubated at its excitotoxic threshold) to a rating of 2 or less. Antagonists used in the present study were MK-801 (Research,Biochemicals Inc, Natick, MA), a specific NMDA receptor antagonist 1287; and 6-cyano-2,3-dihydroxy-7nitro-quinoxaline (CNQX; Tocris, Essex, England), an antagonist of nonNMDA receptors 1291. Each experiment included at least two control spinal cords that were incubated in balanced salt solution alone. In some cases, treated and control cords were immersionfixed overnight in phosphate-buffered 2% paraformaldehyde (pH, 8.0), then prepared for fluorescent cell labeling using the lipid-soluble carbocyanine dye 1' , l-dioctadecyl-3,3,3 ',3 'tetramethylindocarbocyanine perchlorate (DiI; Molecular Probes, Eugene, OR). For this procedure, the dorsal portion of the vertebral column and lateral tissues (including dorsal root ganglia [DRG]) were left attached to the spinal cord. A small crystal of DiI was placed on the DRG, and the preparation was incubated at 37°C for several days. The ventral roots, which pass through the DRG, were stained, and diffusion of the dye in the retrograde direction completely labeled the motor neuron pool. The spinal cords were then sectioned with a vibratome (50 km), and motor neuron cytopathology was observed with a Nikon Microphot AFX microscope equipped with rhodamine epifluorescence. To observe the morphology of motor neurons fixed in vivo without ,any prior exposure to excitotoxins or manipulative procedures, some embryos were anesthetized and perfused with paraformaldehyde fixative through the heart; the vertebral column was then prepared as described, and DiI was applied to the DRG. Results Tissue incubated under control conditions (buffer only) resulted in excellent preservation of neuronal morphology within all three cellular zones of the spinal cord (i.e., dorsal horn, intermediate zone, and ventral horn) (Fig 1A). I n contrast, coincubation with EAAs resulted in acute degeneration of neurons throughout the spinal cord in a concentration-dependent manner (Fig 1B-D), but the nature of excitotoxin-induced

Stewart et al: Excitotoxicity in Spinal Cord

759

Fig 1 . Comparison of spinal cord morphology following a 90-minute incubation in (A) bufler only, (B) N M D A (25 (C) Quis (25 or (0)KA (50 In (A),the three major cellular divisions of the spinal cord (dorsal horn, in-

m),

m),

m).

is severe edematous damage to neurons throughout the spinal cord. The pattern of damage was essentially the Jame for all excitotoxins tested, but the threshold concentration required to product damage oaried wid&. (Bar = 100 m.)

temediate zone, and ventral horn) are evident. In (B-D), there 760 Annals of Neurology Vol 30 N o 6 December 1991

Fig 2. Motor neurons within the ventral horn incubated under control (bufler only) conditions (A) or incubated with 3 pM Dom (B) (Barin {A) = 50 Although motor neurons in control tissue are well preserved, motor neurons in the Domtreated spinal cord are already in an advanced state of degeneration; motor neurons in the lateraE aspect of the ventral born are undergoing a dark-cell type of degeneration with intrarytoplasmic vacuolization compared with motor neurons within the interior, which stuin pallidly. (C) These changes are shown in greater detail in this electron micrograph taken from a spinal cord treuted with 25 NMDA ( x 3,500). (D)Electron micrograph of a dendritic dilation taken from a cord incubated with 40 pM NMDA ( X 14,300);a normal-appearing presynaptic terminal is visible (arrow)and is shown at higher mgnijkation (inset, X 35,870).

a).

damage varied between neuronal populations. Dorsal horn and intermediate zone neurons underwent an acute edematous response consisting of cytoplasmic swelling accompanied by clumping and condensation of nuclear chromatin. Within the ventral horn, motor neurons along the outer margin of the cord typically assumed a dark and condensed appearance that was offset by the presence of many intracytoplasmic vacuoles (dark cell degeneration), whereas more deeply situated motor neurons often stained pallidly (Fig 2A-C).

Numerous dilations were present throughout the neuropil and surrounding white matter. At the electron microscopic level, many dilated structures were identified as dendrites based on the presence of axodendritic contacts o n their surfaces (Fig 2D). Consistent with the nature of an excitotoxic process, the pathological changes were confined to postsynaptic structures, whereas presynaptic axonal boutons appeared normal. The DiI fluorescent cell-labeling approach provided an excellent means of studying excitotoxic cytopathology in motor neurons (Fig 3). Damage induced by EAA agonists appeared as a series of large focal swellings, or varicosities, spread out along the length of each dendrite as it radiated away from the cell body. Tissue incubated in buffer alone possessed a few dilations as well. In perfused tissue, however, dendritic morphology was smooth and dilations were rarely encountered. These differences were quantified by counting the number of varicosities (diameter 2 2 p.m) and by expressing the result as the mean (? standard deviation) number of swellings per 100 p m of dendrite for each of the three conditions: perfused (15 motor neurons) = 0.01 ? 0.13; control/buffer (24 motor neurons) = 1.07 0.57; and NMDA-incubated (18 motor neuStewart et 4: Excitotoxicity in Spinal Cord 761

Fi g 3. Motor neuron pathology visualized using Dil. Top panel are camera hcidu drawings of representatiue motorneurons fmm spinal cords that were taken from an animal that was (1) pedused with parafomzak&hydeefixative prior to dissection (left),(2) dissected and incubated under control conditions (middle), and (3) dissected and incubated with 50 pM NMDA (right). The contours of dendrites taken fmm pdused tissue or spinal cords incubated in buffer alone were generally

smooth (A) compared with numerous mcitotoxin-induced varicosities present along the dendritic shafts of motor neurons incubated with N M D A (B) (Bar = 5 pn).

762 Annals of Neurology

rons) = 4.09 k 1.38. Results for the latter two groups were significantly different when evaluated with an unpaired Student’s t test: t(40) = -9.467; p < 0.0005. In time-course experiments, dorsal horn and intermediate zone neurons underwent edematous changes rapidly (4 pM >4 pM >4pM >4pM 2 4 pMC

aFor each EAA shown, the excitotoxic threshold represents the lowest concentration of agonist necessary to produce a full lesion pattern (damage rating, 4-5). Under each antagonist is listed the lowest concentration required to provide substantial protection (damage rating, 1-2) when coincubared with the threshold concentration of a given EAA. ‘Toxicity of BMAA was greatly attenuated when incubated in low bicarbonate (4 mM) buffer. ‘Effective protection against Glu was only provided by high concentrations of MK-801 (4 pM) and CNQX (500 kM). NMDA :A’-methyl-~-aspartate;Quis = quisqualic acid; AMPA = amino-3-hydroxy-5-methylisoxazole-4-proprionic acid; BOAA = P-N-oxalylamino-L-aine; BMAA = P-N-methylamino-r-alanine; KA = kainic acid; Dom = dornoic acid; Glu = 1-glutamate; NT = not tested.

action selectively at a different subtype of EAA receptor. Our principal findings are that neurons at all levels of the spinal cord undergo acute degeneration when exposed to either NMDA, KA, or AMPAIQuis-type agonists; the neurotoxic effect in each case is dosedependent and preventable by coapplication of an appropriate receptor-specific antagonist. These results suggest that neurons throughout the chick embryo spinal cord possess each of the three major subtypes of EAA receptors known to mediate excitotoxic events. Autoradiographic studies of mammalian spinal cord (human and rat) have shown a higher binding density of EAA ligands in the substantia gelatinosa of the dorsal horn than in the ventral horn [31-341. Although we observed that dorsal horn neurons degenerated more rapidly than motor neurons, it would be premature to attribute the apparent increased sensitivity of these cells to an increased density of EAA receptors in view of age, species, and other differences among the studies being compared. We elected to use the chicken spinal cord in this study because of its close homologous relationship to mammalian spinal cord and its precocious developmental schedule. At the embryonic age employed here (11th embryonic day, stages 36-37 of Hmburger and Hamilton 1353, the period of motor neuron cell death has passed 1361, and descending and sensorimotor connections, many of which may use EAA transmitters, Stewart et al: Excitotoxicity in Spinal Cord

763

are well developed {37, 38). It is unclear, however, how closely this developmental age compares with the adult in expression of EAA receptors, and it is probable that the density of EAA receptors within the spinal cord will increase with further maturation. Our decision to use 1l-day-old embryos was based partially on the observation that it was not possible to incubate more mature spinal cords under control conditions (buffer only) without encountering excessive dendritic edematous changes. These changes may have been due to an excitotoxic mechanism because they could be suppressed by EAA antagonists. The cytopathological changes induced in the isolated chick spinal cord by EAA agonists are identical with those that occur in other CNS regions as a result of excitotoxin exposure C39-417 or that occur as a feature of acute neurodegenerative processes in which endogenous excitotoxins are thought to have a role [42-451. It is noteworthy that certain chick spinal neurons (e.g., small neurons in the dorsal horn and intermediate zone), regardless of the EAA agonist used, tended to display acute dendrosomal swelling and nuclear pyknosis, whereas others (most notably large motor neurons) exhibited the vacuolar condensation type of reaction. A similar dichotomy has been observed between parvicellular and large pyramidal neurons in the brain under excitotoxic-related conditions [27, 4 1, 42, 46). This finding suggests that factors intrinsic to a particular cell type dictate the type of cytopathological reaction to be expected rather than the type of excitotoxic molecule or EAA receptor that mediates the reaction. The use of DiI to label motor neurons following excitotoxic damage provides a novel means of visualizing varicosities along the entire length of the dendritic shaft, thus corroborating in three-dimensional panoramic perspective the massive swelling of individual postsynaptic dendritic segments previously identified in electron microscopic studies as a pathognomonic feature of excitotoxicity [40). That a few dilations also occurred in control tissue incubated in buffer, but did not occur if the tissue was perfused in vivo with h a tive, attests to the susceptibility of the cord to mild excitotoxic-type damage as a function of the isolation/ incubation procedures. The concentration of EAAs required to produce damage varied over a wide range. The order of potencies based on the threshold concentration required to produce a full lesion pattern was Dom ( 3 pM), AMPA ( 1 5 pM), Quis (25 pM), NMDA (25 pM), KA (50 pM), BOAA (100 pM), Glu (3 mM), and BMAA (3 mM). This order is very similar to the order of potencies and excitotoxic threshold concentrations we have found for these same compounds in the isolated chick embryo retina 127); (Stewart, unpublished observations, 1990): Dom (3 pM), AMPA (15 pM), Quis (15 FM), KA (25 pM), NMDA (40 pM), BOAA (100

KM), Glu (1 mM), and BMAA (3 mM). Dom, the shellfish contaminant that caused a serious outbreak of food poisoning in Canada in 1987 [25), was the most potent compound tested in this study, which is consistent with previous observations pertaining to the excitatory-excitotoxic activities of Dom [4, 27, 47). The least potent of the EAA agonists tested was the cycad derivative, BMAA, an environmental agent putatively linked to ALS/parlunsonism/dementia complex. Consistent with the observations of Weiss and Choi [30], we found that physiological concentrations of bicarbonate markedly enhanced the excitotoxic potency of BMAA. Recently, the ability of bicarbonate to confer enhanced excitotoxic potency was described for another simple monocarboxylic amino acid, L-cysteine C48). Within the CNS, in the presence of bicarbonate, these agents may undergo conversion to a carbamate with the dicarboxylic structure of EAAs 1491. Heafield and associates [SO) described abnormally high cysteine-sulfate ratios in the blood of patients with motor neuron disease, parkinsonism, or Alzheimer’s dementia (the neurological disease triad linked with BMAA). Thus, despite the relative weakness of BMAA as a spinal cord excitotoxin in this study, like cysteine it warrants further study as a potential pathogen in human neurological disorders. The known selectivity of MK-801 as an NMDA antagonist and C N Q X as a nonNMDA antagonist provides a means of confirming the receptor specrficity of the excitotoxic actions of EAA agonists. Because Glu acts at both NMDA and nonNMDA receptors, it is not surprising that both C N Q X and MK-801 were required to protect against Glu-induced damage. MK-801 effectively blocked damage by NMDA, whereas the protective action of C N Q X against the other EAA agonists is consistent with available evidence identifying them as nonNMDA receptor agonists, except for BMAA, which has also been described as an NMDA agonist [20, 51). Mixed receptor specificity for BMAA would be consistent with the observations of Weiss and colleagues, who first described BMAA as an NMDA agonist [30], then reported that it behaves as a nonNMDA agonist at lower, presumably more physiologically relevant concentrations [52]. Our findings show that the chick spinal cord is remarkably sensitive to excitotoxic injury, which lends credence to the possible complicity of excitotoxins, either endogenous or exogenous, in neurological disease involving the spinal cord. It might be maintained that our findings argue against the involvement of excitotoxins in motor neuron disease because motor neurons were not more vulnerable than other spinal neurons to excitotoxic damage. However, it is an inherent weakness of acute, in vitro assay systems that they cannot assess whether slightly elevated levels of Glu or aspartate, released in a regionally specific manner and

764 Annals of Neurology Vol 30 No 6 December 1991

acting over a protracted time course (years as opposed to minutes), might lead to the selective degeneration of specific neuronal populations, as suggested by Plaitakis [l5}. Despite this shortcoming, the in vitro assay system described herein does provide a simple and reliable method for (1)further evaluating the vulnerability of specific neurons within the intact isolated spinal cord to acute excitotoxin-induced degeneration, ( 2 ) identifying the EAA receptor subtypes involved in mediating such degeneration, and ( 3 ) developing drugs that may be effective in protecting spinal neurons against excitotoxin-induced degeneration. This research was supported by T32 ES 07066 (G.R.S.), the Center for Cellular and Molecular Neurobiology (small grants program), Washington University School of Medicine (M.P. and W.D.S.), AG 05681, and Research Scientist Award M H 38894 (J.W.0.).

Tht authors wish to thank Barbara C. Fry for expert technical assistance in the histological preparation of tissue for this study.

References 1. Olney JW. Neurotoxicity of excitatory amino acids. In: McGeer EG, OlneyJW, McGeer P, eds. Kainic acid as a tool in neurobiology. New York: Raven, 1978:95-121 2. Olney JW. The toxic effects of glutamate and related compounds in the retina and the brain. Retina 1982;2:341-359 3. Curtis DR, Phillis JW,Watkins JC. The chemical excitation of spinal neurones by certain acidic amino acids. J Physiol 1960; 1501656-682 4. Biscoe TJ, Evans RH, Headley PM, et al. Structure-activity relations of excitatory amino acids on frog and rat spinal neurones. Br J Pharmacol 1976;58:373-382 5. Chase RA, Pearson S, Nunn P, Lantos PL. Comparative toxicities of a- and P-N-oxalyl-L-a$-diaminoproprionic acids to rat spinal cord. Neurosci Lett 1985;55:89-94 6. Pisharodi M, Nauta HJ. An animal model for neuron-specific spinal cord lesions by the microinjection of N-methylaspartate, kainic acid and quisqualic acid. Appl Neurophysiol 1985;48: 2 26-2 33 7. Hugon J, V d a t JM, Spencer PS, et al. Kainic acid induces early and delayed degenerative neuronal changes in rat spinal cord. Neurosci Lett 1989;104:258-262 8. Nag S, Riopelle RJ. Spinal neuronal pathology associated with continuous infusion of N-methyh-aspartate in the rat. Acta Neuropathol 1990;81:7-13 9. Faden AI, Simon EP. A potential role for excitotoxins in the pathophysiology of spinal cord injury. Ann Neurol 1988;23: 623-626 10. Martinez-Arizala A, Rigamonti DD, Long JB, et al. Effects of NMDA receptor antagonists following spinal ischemia in the rabbit. Exp Neurol 1990;108:232-240 11. Olney JW,Misra CH, deGubareff T. Cysteine-S-sulfate: brain damaging metabolite in sulfite oxidase deficiency.J Neuropathol Exp Neurol 1975;34:167-176 12. Plaitakis A, Berl S, Yahr MD. Abnormal glutamate metabolism in an adult-onset degenerative neurological disorder. Science 1982;2 16:193-196 13. Choi DW. Glutamate neurotoxicity and diseases of the nervous sysrem. Neuron 1988;1:623-634 14. Olney JW. Excitatory amino acids and neuropsychiatric disorders. Biol Psychiatry 1989;26:505-525 15. Plaitakis A. Glutamate dysfunction and selective motor neuron

degeneration in amyotrophic lateral sclerosis: a hypothesis. Ann Neurol 1990;28:3-8 16. Ludolph AC, Hugon J, Dwivedi MP, et al. Studies on the aetiology and pathogenesis of motor neuron diseases. Brain 1987; 110:149-165 Misra CH, Rhee V. Brain and retinal damage from 17. Olney JW, the lathyrus excitotoxin, p-N-oxalyl-L-c*,P-diaminoproprionic acid (ODAP). Nature 1976;264:659-661 18. Spencer PS, Ludolph A, Dwivedi MP, et al. Lathyrism: evidence for role of the neuroexcitatory amino acid BOAA. Lancet 1986;II:1066-1067 19. Spencer PS. Guam ALS/parlunsonism-dementia: a long-latency neurotoxic disorder caused by “slow toxin(s)” in food? Can J Neurol Sci 1987;14:347-357 20. Spencer PS, Nunn PB, Hugon J, et al. Guam amyotrophic ]ateral sclerosis-parkinsonism-dementia linked to a plant excitant neurotoxin. Science 1987;237:517-522 21. Plaitakis A, Caroscio JT. Abnormal glutamate metabolism in amyotrophic lateral sclerosis. Ann Neurol 1987;22:575-579 22. Plaitakis A, Constantakakis E, Smith J. The neuroexcitotoxic amino acids glutamate and aspartate are altered in the spinal cord and brain in amyotrophic lateral sclerosis. Ann Neurol 1988;24:446-449 23. Rothstein JD, Tsai G, Kuncl RW, et al. Abnormal excitatory amino acid metabolism in amyotrophic lateral sclerosis. Ann Neurol 1990;28:18-25 24. Quilliam MA, Wright JLC. The amnestic shellfish poisoning mystery. Anal Chem 1989;61:1053-1059 25. Per1 TM, Bedard L, Kosatsky T, et al. An outbreak of toxic encephalopathy caused by eating mussels contaminated with domoic acid. N Engl J Med 1990;322:1775-1780 26. Teitelbaum JS, Zatorre RJ, Carpenter S, et al. Neurologic sequelae of damoic acid intoxication due to the ingestion of contaminated mussels. N Engl J Med 1990;322:1781-1787 27. Stewart GR, Zorumski CF, Price MT, Olney JW.Domoic acid: a dementia-inducing excitotoxic food poison with kainic acid receptor spccificity. Exp Neurol 1990;110:127-138 28. Wong EHF, Kemp JA, Priestley T, et al. The anti-convulsant MK-801 is a potent A’-methyl+-aspartate antagonist. Proc Natl Acad Sci USA 1986;83:7104-7108 29. Blake JF, Brown MW, Collingridge GL. CNQX blocks acidic amino acid induced depolarizations and synaptic components mediated by non-NMDA receptors in rat hippocampal slices. Neurosci Lett 1988;89:182-186 30. Weiss JH, Choi DW. Beta-N-methylamino-L-alanineneurotoxicity: requirement for bicarbonate as a cofactor. Science 1988; 241:973-975 31. Monaghan DT, Cotman CW. The distribution of I3H] kainic acid binding sites in rat CNS as determined by autoradiography. Brain Res 1982;252:91-100 32. Greenamyre JT, Young AB, Perry JB. Quantitative autoradiographic distribution of L-[~H] glutamate-binding sites in rat central nervous system. J Ncurosci 1984;4:2133-2144 33. Monaghan DT, Cotman CW. Distribution of N-methyl+aspartate-sensitive ~-[~H]gIutamate-bindingsites in rat brain. J Neurosci 1985;5:2909-2919 34. Jansen KLR, Faull RLM, Dragunow M, et al. Autoradiographic localisation of NMDA, quisqualate and kainic acid receptors in human spinal cord. Neurosci Lett 1990;108:53-57 35. Hamburger V, Hamilton HL. A series of normal stages in the development of the chick embryo. J Morphol 1951;88:49-92 36. Hamburger V. Cell death in the development of the lateral motor column of the chick embryo. J Comp Neurol 1975;160: 535-546 37. Okado N, Oppenheim RW. The onset and development of descending pathways to the spinal cord in the chick embryo. J Comp Neurol 1985:232:143-161

Stewart et al: Excitotoxicity in Spinal Cord

765

38. Lee MT, Koebbe MJ, ODonovan MJ. The development of sensorimotor synaptic connections in the chick embryo. J Neurosci 1988;8:2530-2543 39. Olney JW.Brain lesions, obesity and other disturbances in mice treated with monosodium glutamate. Science 1969;164: 719-721 Glutamate-induced neuronal necrosis in the infant 40. Olney JW. mouse hypothalamus: an electron microscopic study. J Neuropathol Exp Neurol 1971;30:75-90 41. Olney JW.Kainic acid and other excitotoxins: a comparative analysis. In: DiChiara G, Gessa GL, eds. Glutamate as a neurotransmitter. New York Raven, 1981:375-384 42. Olney JW, deGubareff T, Sloviter RS. “Epileptic”brain damage in rats induced by sustained electrical stimulauon of the perforant path. 11. Ultrastructural analysis of acute hippocampal pathology. Brain Res Bull 1983;10:699-712 43. Auer RN, Kalimo H, Olsson Y ,Siesjo BK. The temporal evolution of hypoglycemic brain damage. 11. Lght- and electronmicroscopic findings in the hippocampal gyms and subiculum of the rat. Acta Neuropathol (Berl) 1985;67:25-36 44. Olney JW,Collins RC, Sloviter RS. Excitotoxic mechanisms of epileptic brain damage. In: Delgado-Escueta AV, Ward AA. Jr, Woodbury DM, Porter RJ, eds. Basic mechanisms of the epilepsies: molecular and cellular approaches. New York: Raven, 19862357-878

45. Ikonomidou C, Price MT, Mosinger JL, et al. Hypobaricischemic conditions produce glutamate-like cytopathology in infant rat brain. J Neurosci 1989;9:1693-1700 46. Clifford DB, Olney JW, Maniotis A, et al. The functional anafomy and pathology of lithium-pilocarpine and high dose pilocarpine seizures. Neuroscience 1987;23:953-968 47. Debonnel G , Beauchesne L, De Montigny C . Domoic acid, the alleged “mussel toxin,” might produce its neurotoxic effect through kainate receptor activation: an electrophysiological study in the rat dorsal hippocampus. Can J Physiol Pharmacol 1989;67:29-33 48. Olney JW, Zorumski C, Price MT, Labruyere J. L-cysteine, a bicarbonate-sensitive endogenous excitotoxin. Science 1990; 248:596-599 49. Nunn PB, Davis AJ, OBrien P. Carbamate formation and the neurotoxicity of L-a-amino acids. Science 1991;251:1619-1620 50. Heafield MT, Fearn S, Steventon GB, et al. Plasma cysteine and sulphate levels in patients with motor neurone, Parkinson’s and Alzheimer’s disease. Neurosci Lett 1990;110:216-220 51. Zeevalk GD, Nicklas WJ. Acute excitotoxicity in chick retina caused by the unusual amino acids BOAA and BMAA: effects of MK-801 and kynurenate. Neuosci Lett 1989;102:284-290 52. Weiss JH, Koh J, Choi, DW. Neurotoxicity of P-N-methylamino-L-alanine (BMAA) and p-N-oxalylamino-L-alanine(BOAA) on cultured cortical neurons. Brain Res 1989;497:64-71

766 Annals of Neurology Vol 30 No 6 December I991