WO1991002523A1

WO1991002523A1 - Glutamatergic amino acid agonists and antagonists

Info

Publication number: WO1991002523A1
Application number: PCT/US1990/004647
Authority: WO
Inventors: Paul A. Rosenberg; Elias Aizenman
Original assignee: Boston Childrens Hospital
Current assignee: Boston Childrens Hospital
Priority date: 1989-08-17
Filing date: 1990-08-17
Publication date: 1991-03-07
Anticipated expiration: 1992-02-17

Abstract

A nerve cell medium composition containing an effective dose of an agonist or antagonist of a glutamate receptor, effective to act as an agonist or antagonist on a nerve cell, chosen from an amino acid having chemical structure (I), wherein A is chosen from (a), (b), (c) or (d), wherein each R1, R2, R3, R4, R5, R6 R8, R9, R10, R11, and R12 is independently an electronegative group, H, CH3, SCH3, NH2 or OCH3; and R7 is an electronegative group.

Description

GLUTAMATERGIC AMINO ACID

AGONISTS AND ANTAGONISTS

Background of the Invention

This invention concerns amino acids, and methods of their use, which protect a neuron of a human patient from injury, especially injury caused by the presence of excess glutamate or related compounds.

Glutamate is known to be an excitatory amino acid (causing depolarization and action potential generation when applied to neurons of the central nervous system) with efficacy at at least three subtypes of excitatory amino acid receptors, namely kainate, quisqualate, and N-methyl-D-aspartate (NMDA). It is present at high

concentrations in mammalian central nervous systems (CNS) and is toxic to central neurons. Glutamate or compounds that act like glutamate (glutamate agonists) are thought to play a role in neuronal injury, and to mediate a variety of brain insults, including ischemia, hypoxia, and physical trauma. A role for glutamate or glutamate agonists has been postulated in Alzheimer's disease, Parkinson's disease, and Huntington's disease.

Certain glutamate antagonists (compounds that block the action of glutamate or glutamate agonists) can attenuate the acute neuronal injury produced by hypoxia, ischemia, and hypoglycemia. This protective effect on central neurons indicates that the antagonists may have clinical therapeutic utility, e.g., in treatment of hypoxic brain injury. By antagonist is meant a compound which counteracts the effect of an agonist, i.e., it opposes the action on a nerve cell of an agonist. Implicit in the concept of agonist and antagonist action is interaction of these substances at a specific

receptor. Agonists bind to specific receptor sites and this interaction results in a measurable response.

Antagonists bind to the same site, and are capable of displacing agonists and preventing the occurrence of the associated response. Choi, U.S. Patent 4,806,543, describes a method for reducing adverse effects of neurotoxic injury by administering an enantiomer of an analgesic opioid agonist or antagonist. Such compounds are said to be useful for treatment of any animal species having NMDA receptors.

Hahn et al. (Proceedings National Academy of Science USA, 85:6556, 1988; not admitted to be prior art to the present invention) state that, in the mammalian central nervous system, glutamate is thought to be the major excitatory neurotransmitter and acts at the three receptor subtypes. Excessive stimulation of the NMDA subtype has been implicated in the pathophysiology of neuronal degeneration caused by a variety of conditions; these include anoxia, ischemia, hypoglycemia, seizures, and several neurodegenerative diseases, such as

Huntington's disease, and the amyotrophic lateral

sclerosis-Parkinsonism-dementia complex found on Guam.

Honore et al. Science 241:701, 1988, describe quinoxalinediones as competitive antagonists of a non-NMDA glutamate receptor. Administration of these

antagonists to cells having glutamate receptors reduces the response to quisqualate and kainate, but has little effect on response to NMDA. The authors state that these compounds will be useful probes to study the role of non-NMDA receptors in synaptic transmission in the mammalian brain, and may offer important leads to the therapeutic potential of non-NMDA antagonists. Biscoe et al., Br. J. Pharmac. 58.:373, 1976 describe 2,4,5-Trihydroxyphenylalanine (6.0H.DOPA, or TOPA) as a weaker excitant than L-glutamate when perfused onto frog spinal motoneurones. In contrast, it has no activity on electrophoretic administration. They suggest that this difference may be due to the oxidation of topa to produce an inactive form.

Karschin et al., J. Neurosci. 8_.:2895, 1988 describe the effects of two non competitive NMDA

antagonists phencyclidine (PCP) and MK-801. Both

completely block responses to NMDA but not to kainate.

Murphy et al. Br. J. Pharmacol. 95:932, 1988 describe the NMDA antagonist [³H]CGS-19755 as one of the most potent competitive NMDA antagonists.

Ross et al., Brain Research 425:120, 1987 describe excitant amino acids found in seeds of plants, namely, ß-N-methylamino-L-alanine and ß-N-oxalylamino-L-alanine. These amino acids are thought to be associated with Guam amyotrophic lateral sclerosis and lathyrism. The actions of the amino acids can be blocked by specific glutamate receptor antagonists, e.g., cis-2, 3-piperidine

dicarboxylic acid, and 2-amino-7-phosphonoheptanoic acid but not by the antagonists glutamine diethyl-ester and streptomycin.

Bridges et al., J. Neurosci 9:2073, 1989, (not admitted to be prior to the present application) describe ß-N-oxalyl-L-α,ß-diaminopropionic acid as a potent excitotoxic amino acid, and an agonist at the non-NMDA receptor. The authors state that this compound will be useful as a probe of non-NMDA mechanisms, and that future therapeutic strategies attempting to block excitotoxic damage must take into account the contribution of nonNMDA receptors to the pathologies. Summary of the Invention

In a first aspect, the invention features a nerve cell medium composition containing an effective dose of an agonist or antagonist of a glutamate receptor, effective to act as an agonist or antagonist on a nerve cell or of neuronal injury, chosen from an amino acid having the following chemical structure:

wherein A is chosen from

wherein each R₁, R₂, R₃ R ₄, R₅, R₆, R₈, R ₉, R₁₀, R₁₁, and R₁₂ is independently an electronegative group, H, CH₃, SCH₃, NH₂ or OCH₃; and R₇ is an electronegative group.

By nerve cell medium composition is meant a buffer or vehicle which is compatible with nerve cell survival in vitro, e.g., physiological saline. In preferred embodiments, each electronegative group is separately selected from OH, N₂O, CN, SO₃H, Cl, F, Br, I, and CO₂H; A is:

and each R group is independently CH₂, SCH₃ , OCH₃ or H; A is

and R₁ and R₃ are independently H, CH₃, OCH₃ or SCH₃ and R is H, CH₃, SCH₃, OCH₃ or an electronegative group; A is:

and R₄ and R₆ group are independently SCH₃, CH₃, OCH₃ or H, and R₅ is SCH₃, CH₃, OCH₃, H or an electronegative group; most preferably R₄ and R₆ are each H and R₅ is H or a halogen; and A is:

and R₇ is an electronegative group, most preferably CN or a halogen.

In a related aspect, the invention features a pharmaceutically acceptable composition including an effective dose of an antagonist of a glutamate receptor, effective to act as an antagonist of neuronal injury, e.g., toxicity, or death, chosen from the above described amino acids. The antagonist is admixed with a suitable buffer.

In another aspect, the invention features a method for identifying an amino acid antagonist useful for protection of a neuron of an organism from injury. The method includes providing a cell having a glutamate receptor, selecting an amino acid from those described above, treating the cell with the amino acid, and

determining whether the amino acid acts as an antagonist at the receptor, e.g., coapplying the amino acid with a known agonist.

By protection of a neuron from neuronal injury is meant that the amino acid is able to either totally reverse, or at least. partially reverse, the effect of a toxic agonist on that neuron, e.g., kainate or

quisqualate. Alternatively, the amino acid is able oo r.educe the effect of glutamate on the neuron, and thereby significantly increase the chances of that neuron

surviving in the presence of glutamate or a related substance.

The term organism is intended to include any animal to which an amino acid of the invention can be administered for the indicated purpose, including both medicinal and veterinary purposes. Use in mammals and birds of all types is preferred, with use in humans being a primary utility.

In preferred embodiments, the step of determining includes determining the electrophysiological response of the cell to the amino acid, with or without the amino acid; or determining the survival of the cell after exposure to the amino acid; the survival may be

determined in the presence of the amino acid and a known agonist. Most preferably, the receptor is chosen from a non-NMDA receptor.

In another related aspect, the invention features a method for protecting a neuron of a human patient from injury. The method includes identifying a patient susceptible to neuronal injury, providing a

pharmacologically acceptable composition including an antagonist of a glutamate receptor, chosen from those listed above, and administering the antagonist to the patient in an amount effective to protect a neuron from injury.

Organisms, e.g., human patients, susceptible to neuronal injury are identified by any of a number

standard techniques. Such patients will include those discussed above which are susceptible to, or suffer from, strokes, anoxia and certain degenerative diseases. They will also include those patients which have no symptoms but are found to have abnormally high levels of glutamate or related compounds in the CNS, and also those who have a genetic predisposition to the development of disease, e.g., Huntington's disease. Those skilled in the art will recognize how to determine, by routine

experimentation, the amount of amino acid necessary to provide sufficient protection of a neuron without causing significant deleterious or side effects to the patient. Generally, this amount will be a balance between a level of amino acid where the potential of causing such

deleterious effects is significant and a level where the amino acid provides complete protection against injury to the neuron.

We have identified a class of simple amino acids, which are derivatives of topa quinone, a potent glutamate agonist, and are generally active at non-NMDA receptors. These amino acids are readily synthesized and serve as useful glutamate agonists and antagonists. Since they are amino acids related to dopa they are expected to be readily transported into the central nervous system through the blood brain barrier. The antagonists are useful for the treatment of strokes, or any equivalent insults to the brain, e.g., hypoxia, and physical trauma, and for treatment of Alzheimer's disease. The

antagonists are also useful in the treatment of diseases which seem to involve central dopamine projections and their target areas, including schizophrenia,

complications of anti-phychotic drug therapy, Parkinson's disease, and Huntington's disease.

Other features and advantages of the invention will be apparent from the following description of the preferred embodiments thereof, and from the claims.

Description of the Preferred Embodiments The figure will first briefly be described. Drawing

The figure is a schematic representation of chemicals produced by oxidation. of topa. The numbers represent the following compounds: 1, tyrosine; 2, dopa; 3, dopamine; 4, dopa quinone; 5, leuko-dopachrome; 6, dopachrome; 7, topa; 8, ortho-quinone of topa; 9, paraquinone of topa.

Amino Acids

Amino acids useful in this invention are described above. They may be synthesized by any standard

technique, well known to those of ordinary skill in the art. There follows an example of one such amino acid, topa quinone, which acts as an agonist at non-NMDA sites. Topa quinone is a simple and novel prototype for non-NMDA receptor ligands, and may be an endogenous neurotoxin. The above enumerated amino acids are

variations on the structure of topa quinone. This example is not meant to be limiting to the present invention.

Application of solutions of topa to central neurons results in glutamatergic responses mediated predominantly by non-NMDA receptors. In addition, exposure to solutions of topa results in killing of greater than 97% of the neurons in cortical cultures.

This neurotoxicity is partially prevented by the non-NMDA antagonist CNQX. Topa itself is unstable in aqueous solutions and oxidizes to form the amino acid topa quinone. This compound appears to be the active

compound, rather than topa, at glutamatergic receptors. Details of the effect of topa quinone are now presented.

Example 1: Effect of Application of Topa

to Cortical Neuron

Experiments on cortical neurons used rat cerebral cortex in dissociated cell culture prepared as follows. Tissue was derived from E16 fetal rats, dissociated using trypsin, and plated on collagen and poly-L-lysine coated glass coverslips. Growth medium was Dulbecco's modified Eagle's medium/Ham's F-12/calf serum 8:1:1 (DHS). Cultures were mitotically inhibited by exposure to 5 μM cytosine arabinoside for 48 hours starting at 15 days in vitro. and medium was changed three times per week.

Approximately 6% of the cells in these cultures were neurons, as identified by tetanus immunochemistry.

Electrophysiological studies were performed on neurons from 3-5 week old cultures, using patch electrodes (3-5 Mohms) to investigate the membrane responses of the neurons to application of topa.

The extracellular solution for the physiological experiments contained (in mM) : NaCl, 137; NaHCO₃, 1;

NaHPO₄, 0.34; KCl, 5.36; KH₂PO₄, 0.44; CaCl₂, 2.5; HEPES, 5; and dextrose, 22.2, as well as phenol red, 0.011 g/L; and glycine, 1 μM, adjusted to pH 7.2 with 0.3 N NaOH (no added magnesium). One μM TTX was routinely added to block synaptic activity. The intracellular pipette solution contained (in mM) CsCl, 120; TEA-Cl, 20; MgCl₂, 2; CaCl₂, 1; EGTA, 1.5-2.25; and HEPES 10, adjusted to pH 7.2 with concentrated NaOH. Drugs were applied by pressure ejection from micropipettes (5 μM aperture) placed in close proximity (20 μM) to the cell under study. For these and all other experiments, 10 mM stock solutions of topa (Sigma) were made in 1 mM HC1 and were kept frozen at -80°C.

A List EPC-7 patch-clamp amplifier was used, and signals were digitized with a 12-bit 125 kHz analog-to-digital converter (Model DT2782 DMA: Data Translation), and viewed both on an analog oscilloscope and a Hewlett-Packard digital display. The sampling rate was set at 1 to 1.6 kHz and the signals were filtered at 500 Hz. The indifferent Ag/AgCl electrode was connected to the extracellular solution by a 2 M KCl-agarose bridge.

Cortical neurons had an input resistance of 0.3 to 0.7 Gohms and a cell capacitance of 20-40 pF. Recordings were performed at 33-35°C. Application of 20-100μM topa elicited responses in all neurons tested which reversed in polarity near OmV. Currents activated by application of 50μM topa were substantially (>90%) and reversibly blocked by 10-20 μM 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX) , a

competitive non-NMDA glutamatergic antagonist. Any remaining current after CNQX block was only partially antagonized by the NMDA blocker DL-2-amino-5-phosphonovaleric acid (APV) at 200μM. APV (200 μM) added alone blocked approximately 25% of 50 μM topa-induced responses in cortical neurons.

Example 2 : Effect of Application of Topa on

Chick Eyecup

Futher pharmacological studies were performed on the chick eyecup, which is a convenient preparation for the study of glutamatergic substances. Eyecups obtained from 5-17 day old chicks were continously superfused at room temperature, and drugs applied directly via the perfusate. Responses were recorded as a change in the potential difference between two electrodes, one placed in the solution inside the eyecup, and one in electrical contact with the optic nerve. Perfusion was at a rate of 8ml/min with a physiological saline (130 mM NaCl, 3 mM KCl, 7mM MgCl₂, 0.1 mM CaCl₂, 17 mM glucose, 20 mM NaHCO₃, 0.011 g/L phenol red) bubbled with a gas mixture of 95% O₂ and 5% CO₂. The concentrations of calcium and

magnesium were chosen to reduce synaptic activity.

The DC potential difference between a unipolar suction electrode placed on the cut optic nerve and a unipolar Ag/AgCl electrode placed inside the eyecup solution was amplified 1000X and displayed on an

oscilloscope. Data were transferred and stored in a microcomputer via an A to D converter. Traces were triggered by the opening of two solenoid valves which shunted approximately 200 μl of agonist into the perfusion line. Agonist reached the preparation

approximately 7 seconds later. The actual concentration of drugs reaching tissue is not known.

Addition of 300μM topa produced a large response, similar in polarity to that produced by either kainate, quisqualate, or NMDA. Responses to application of 300μM topa were 73-100% blocked by 5 μM CNQX but were not affected by 100-200 μM APV. NMDA responses can be detected in this preparation, even in high magnesium, low calcium solution.

Example 3 : Topa Quinone as an Agonist

Topa rapidly oxidizes in aqueous solution at physiological pH (as seen from the development of

orange-red color with time). To determine whether the activity of topa was due to topa itself or an oxidation product, the effect of DTT was tested. Dithiothreitol (DTT) is a reducing agent whose effects on glutamatergic responses in the eyecup preparation have been

characterized. DTT appears to affect NMDA but not non-NMDA glutamatergic responses. DTT was found to be an effective agent for preventing the oxidation of topa at a 10:1 molar ratio. The addition of 3mM DTT eliminated the response of the above described eyecup preparation to 300μM topa.

Three other lines of evidence indicate that the glutamatergic activity of topa is related to its

oxidation product: 1) Solutions of topa in physiological saline became colored in 2-3 minutes. If solutions of topa were applied to the eyecup immediately after

preparation, and prior to color development, no responses were elicited. Several minutes later, after the topa solution had become colored, its application produces a typical response. 2) Superoxide dismutase (SOD) retards the oxidation of catecholamines. Inclusion of 10μg/ml SOD with 300μM topa briefly retards the oxidation of topa in solution and, in parallel, also retards the

development of glutamatergic activity. 3) In binding experiments in washed rat cortical membranes, using the high affinity NMDA antagonist ³H-CGS-19755 as a ligand, the addition of topa produces a dose-dependent decrease in binding, with 39+6% specific counts remaining in the presence of 50 μM topa. DTT was effective in preventing the oxidation of topa under the conditions of the binding assay. In the presence of 2 mM DTT, in contrast to the result obtained with topa alone, 94+17% specific counts remained in the presence of 50 μM topa. The addition of SOD plus catalase, which under the conditions of assay did not prevent the oxidation of topa, but would be expected to eliminate superoxide and peroxide formed, did not significantly affect the number of specific counts remaining in the presence of 50 μM topa (44+6%,). These results all indicate that the active component of

solutions of topa is not topa itself, but rather an oxidation product. Therefore, the active species appears to be one of the oxidation products of topa, shown in the Figure as compounds 6, dopachrome; 8, topa ortho-quinone; or 9, topa para-quinone.

Example 4 : Dopachrome is Not an Agonist

Dopachrome was synthesized from dopa using silver oxide (Ag₂O) Occording to published methods. A solution of 0.5 mg/ml L-dopa in sodium phosphate buffer (50 mM, pH 6.8, passed through a Chelex-100 column) was reacted with Ag₂O for 3 minutes at 0ºC, filtered through a 0.22 μ Millex filter, and batch treated with Chelex-100 to remove Ag ions. The resulting product had no activity in the eyecup preparation experiments described above. In contrast, when topa was used after exposure to silver oxide, using the same method, the product is as potent in the eyecup preparation as spontaneously oxidized topa itself. Unlike solutions of dopachrome, which are stable for about 30 minutes on ice and thereafter convert from a deep orange-red to a brown color (suggestive of further oxidation), solutions of the oxidation product of topa (also orange-red) are much more stable and persist on ice without color change for several hours. The absorbance spectrum of dopachrome, and oxidized topa, in 50 mM sodium phosphate buffer pH 6.8 display a broad absorbance peak centered at approximately 475 nm, and display absorbance peaks in the ultraviolet range, at 306 nm for dopachrome (in agreement with the value previously reported) and at 272 nm for oxidized topa. Topa quinone reacts with ninhydrin, whereas dopachrome did not.

It appears the oxidation of topa in solution at physiological pH produces a relatively stable product which acts as an agonist at non-NMDA receptors, which is distinct from dopachrome, and is the amino acid topa quinone (tautomeric compounds 8 and 9 in the Figure).

Example 5: Toxicity of Topa Quinone

Toxicity experiments were performed on cultures 3-8 weeks in vitro. Rat cerebral cortex in dissociated cell culture was exposed to 500 μM topa, alone, or in conjunction with 20 μM CNQX, 20 μM MK-801, or CNQX plus MK-801. Coverslip cultures were washed once in a

physiological saline (PS, in mM) : NaCl, 145; KCl, 3;

CaCl₂, 1.8; MgCl₂, 1.0; glucose, 8; NaH₂PO₄, 2.4; and Na₂HPO₄, 0.42, using 2 ml/35 mm dish containing 5

coverslips. The coverslips were then placed in wells containing 0.5 ml MEM (no glutamine) with either 500 μM topa or a vehicle. After 3 hours, experiments were terminated by replacing medium with trypan blue (1:1 dilution with PS), washing once in PS containing 0.01% BSA, and fixing with 2.5% glutaraldehyde in PS.

Surviving neurons were identified as those which had excluded trypan blue. Cultures were counted following 24 hours in fixative, at 125X. Neurons were identified as phase bright cells 10-30 μm in diameter. The accuracy of morphological identification was confirmed both by tetanus immunochemistry, and by electrophysiology on live cultures. CNQX (Tocris Neuramin) was added from a 5mM stock in 0.3 M NaOH. MK-801 HCl was added from a 2 mM stock in water. Topa was added from a 10 mM stock in 1 mM HCl kept at -80°C.

In some experiments, medium containing topa was relaced with MEM after three hours, and cultures were returned to the incubator for 15-20 hours, and were then trypan blue stained and fixed. Cultures treated in this way showed obvious glial toxicity-trypan blue staining of the glial layer starting at the periphery and extending variably toward the center. Few neurons were left in these trypan blue stained areas, even in the presence of CNQX, whereas in areas where the glial layer was not stained, many surviving neurons were present.

A three hour exposure to 500 μM topa produced a large loss of neurons (a mean of 2.7+1.1% neurons

survived in 4 experiments) from the cortical cultures. In order to determine the pharmacology of the toxicity. of topa, experiments were performed in which cultures were exposed either to topa (500 μM) alone or in conjunction with 20 μM CNQX, 20 μM MK-801 HCl [(+)-10,11-dihydro-5-methyl-5H-dibenzo-

-cycloheptene], or CNQX plus MK-801 together. MK-801 is an NMDA channel blocker and has been shown to block NMDA receptor mediated toxicity in central neurons in culture. CNQX, but not MK-801, produced a significant sparing of neurons exposed to topa. The fact that the majority of neurons could not be saved by glutamate antagonists in these experiments indicates that 500 μM topa is toxic to neurons by both glutamatergic and non-glutamatergic mechanism. Because of the glutamatergic properties of a dopa oxidation product, dopaminergic nuclei or their target areas may be at particular risk for glutamate receptor mediated neurotoxicity. The demonstration of the

glutamatergic activity of topa indicates that it is useful to determine the presence of this substance, and its breakdown products, in disease states involving dopaminergic areas of the brain, such as Parkinson's disease (both in untreated patients as well as those being treated with L-dopa), Huntington's disease, brain ischemia, as well as schizophrenia. In Huntington's disease and brain ischemia an endogenous NMDA agonist has been implicated, whereas the glutamatergic toxicity of topa appears to be mediated via non-NMDA receptors.

Given the voltage dependence of NMDA activated channels in the presence of magnesium, it is conceivable that a non-NMDA agonist might predispose the cell to NMDA mediated toxicity by causing sufficient depolarization to result in excessive activation of NMDA channels by the normal endogenous transmitter.

Useful Amino Acids

Not all of the above listed amino acids are useful in vivo. In general, amino acids having a chemical structure similar to topa quinone are potentially useful agonists (for in vitro use) or antagonists (for in vivo and in vitro use). These amino acids can be synthesized by standard procedure and screened as described in this application to determine their utility. The tests described above can be used to determine agonist and antagonists activity in vitro, and those described below to determine activity in vivo. Other equivalent tests are well known to those of ordinary skill in this art. Generally, those amino acids which are antagonists at non-NMDA receptors are useful. Amino acids which are antagonists at NMDA receptors are also useful. The 4 vessel occlusion model of ischemia in rat described by Pulsinelli, et al. Stroke, 10:2678, 1979 is used. Generally, those compounds which show antagonist activity at the glutamate receptors are administered to animals prior to induction of ischemia by bilateral temporary ligation of the common carotid arteries in rats with prior permanent occlusion of the vertebral

arteries. Animals are subject to ischemia for 20

minutes, then perfusion is reestablished. After 72 hours the animals are fixed, and the brains embedded in

paraffin. Thick sections are cut and stained with H & E, and examined for ischemic changes in several regions, including hippocampus, neocortex, and striatum. Special attention is given to the striatum because it is in this structure that a role for dopamine in ischemic injury has been established.

Careful attention is also given to the timing of the administration of drugs in relation to the ischemic event because of the possibility that drugs given after the ischemic event may still reduce injuury and death.

Alternatively, the methamphetamine-induced nigrostriatal dopaminergic toxicity model of Sonsalla et al. Science 243 :398 can be used. Generally, mice receive 4 injections of methamphetamine at 2 hr intervals (1-10 mg/kg). For testing antagonist potency of drugs, drugs are given 15 minutes before and 3 hours after the first injection of methamphetamine. Animals are killed 3 days after treatment. Dopamine and tyrosine hydrόxylase activity are measured in striata of test animals.

Use

The above methods, and other methods well known to those in the art, are useful for identification of amino acids useful in treatment of central neuronal injury, such as the acute and chronic neurological diseases of ischemia, hypoxia, hypoglycemia, epilepsy, Parkinson's, Huntington's disease, and Alzheimer's disease. These amino acids generally act to selectably block the

neurotoxicity of glutamate at only one of the glutamate receptors, namely a non-NMDA receptor, and thus allow accomplishment of protection against glutamate with a low number of side effects. Thus, there is less disruption of normal brain function by these amino acids than by compounds effective at more than one site.

It is, of course, possible to use the amino acids identified above in conjunction with other compounds which act at other sites of glutamate receptors. In fact, such use will provide synergistic results, in that the level of protection of the neuron from neuronal injury will be greater than the protection provided by either agent alone. This means that lower levels of amino acids, which are identified as useful in the invention, can be used in combination with agents which act at other sites. Thus, advantageous compositions useful for treatment of the above diseases can be formed by combinations of existing agents, and those amino acids identified by the method of the present invention.

Useful amino acids identified by the above methods can be used by standard procedures in treatment of the above mentioned diseases, and related diseases or

symptoms. These amino acids are administered to patients susceptible to neuronal injury in an amount of amino acid sufficient to reduce the neuronal injury. Such

administration can be performed on any animal having neuronal glutamate receptors and includes mammals, birds and, in particular, humans.

Administration can be by any technique capable of introducing the amino acid into the blood stream of the patient. Once introduced, the amino acids are expected to penetrate the blood-brain barrier. These techniques include oral administration, and intravenous,

intramuscular, and subcutaneous injections.

The amino acids of the invention can be formulated into orally administerable forms or pills by standard procedure. Typical doses of the amino acids in

pharmaceutically acceptable carriers are from 50 mg to 2 g, and preferably from 100 mg to 1 g. These doses are. suitable for administration to a typical 70 kg human.

Administration can, be adjusted to provide the same relative dose per unit of body weight. Typically useful concentrations of the amino acids in the blood stream is the order of 1 to 1000 micromolar, preferably from 1 to

100 micromolar, even more preferably 2.5 to 25

micromolar.

The above amino acids are also useful for in vitro tests, such as for binding studies on rat cortical membranes, study of physiology of rat cortical neurons and chick eyecup preparations (these tests are described above and can be used to discriminate between agonist and antagonist activity), and for toxicity studies.

Other embodiments are within the following claims.

Claims

3. The pharmaceutically acceptable composition of claim 9 wherein said A is:

and each said R group is independently CH₃, SCH₃, OCH₃ or H.

4. The pharmaceutically acceptable composition of claim 9 wherein said A is:

and said R₁ and R₃ groups are independently H, CH₃, OCH₃, or SCH₃, and said R₂ is H, CH₃, SCH₃, OCH₃, or an

electronegative group.

5. The pharmaceutically acceptable composition of claim 9 wherein said A is:

and said R₄ and R₆ groups are independently SCH₃, CH₃, OCH₃, or H, and R₅ is SCH₃, CH₃ , OCH₃, H, or an

electronegative group.

6. The pharmaceutically acceptable composition of claim 9 wherein said A is:

and R₇ is an electronegative group.

7. The pharmaceutically acceptable composition of claim 9 wherein R₄ and R₆ are each H and R₅ is H or a halogen.

8. The pharmaceutically acceptable composition of claim 15 wherein R₇ is CN, or a halogen.

9. A method for identifying an amino acid antagonist useful for protection of a human neuron from injury, comprising the steps of:

providing a cell comprising a glutamate receptor, selecting an amino acid having the following chemical structure:

wherein A is chosen from

wherein each R₁ , R₂ , R₃ , R₄, R₅, R₆ R₈, R₉ , R ₁₀, R ₁₁, and

R ₁₂ is independently an electronegative group, H, CH₃, NH₂, OCH₃ or SCH₃; and R₇, is an electronegative group and wherein R₁ or R₂ is an electronegative group,

treating said cell with said amino acid, and determining whether said amino acid acts as an antagonist at said receptor.

10. The method of claim 17 wherein said step of determining includes determining the electrophysiological response of said cell. to said amino acid.

11. The method of claim 17 wherein said step of determining includes determining the survival of said cell after exposure to said amino acid.

12. The method of claim 19 wherein said survival is determined in the presence of said amino acid and a chemical known to agonize said receptor

13. The method of claim 17 wherein said receptor is chosen from a non-NMDA receptor.

14. A method for protecting a neuron of a human patient from injury; comprising the steps of: identifying a patient susceptible to neuronal injury,

providing a pharmacologically acceptable

composition comprising an antagonist of a glutamate receptor, said antagonist being chosen from an amino acid having the following chemical structure:

wherein A is chosen from:

wherein each R₁, R₂, R ₃, R₄ , R₅, R₆, R₈, R₉, R₁₀, R_{1 1}, and

R₁₂ is independently an electronegative group, H, CH₃, NH₂, OCH₃, or SCH₃; and R₇ is an electronegative group, and wherein R₁ or R₂ electronegative group, and

administering said antagonist to said patient in an amount effective to protect a neuron from injury.

15. The method of claim 22 wherein each R is independently selected from the group consisting of OH, N₂, CN, SO₃H, Cl, F, Br, I, and CO₂H.

16. The method of claim 22 wherein said A is

and each said R group is independently CH₃, SCH₃, OCH₃, or H.

17. The method of claim 22 wherein said A is:

and said R₁ and R₃ groups are independently H, CH₃, OCH₃ or SCH₃ and said R₂ is H, CH₃, SCH₃, OCH₃, or an

electronegative group.

18. The method of claim 22 wherein said A is:

and said R₄ and R₆ groups are independently SCH₃, CH₃. OCH₃, or H, and R₅ is SCH₃, CH₃, OCH₃, H, or an

electronegative group.

19. The method of claim 22 wherein said A is

and R₇ is an electronegative group,

20. The method of claim 22 wherein R₄ and R₆ are each H, and R₅ is H, or a halogen.

21. The method of claim 28 wherein R₇ is CN, or a halogen.

22. A nerve medium composition comprising an amino acid present in an amount effective to modify glutamate receptor-related neuronal injury, said amino acid having the following chemical structure:

wherein each R₁, R₂, R₃, R₄, R₅, R₆, R₈, R₉, R₁₀, R₁₁, and R₁₂ is independently an electronegative group, H, CH₃, NH₂, OCH₃ or SCH₃, and R₇ is an electronegative group and wherein R₁ or R₂ is an electronegative group.