Changes in the state of the excitatory synaptic system in the hippocampus on prolonged exposure to excitatory amino acids and antagonists

Prolonged exposure of hippocampal slices to L-glutamate was found to produce a biphasic effect on excitatory synaptic transmission from Schaffer collateral-commissural fibers to CA1 pyramidal neurons: an early blockade of postsynaptic responses was followed by a progressive recovery. Antagonists of N-methyl-D-aspartate (NMDA) receptors, which had no effect on the initial responses, blocked the responses which reappeared during continued exposure to L-glutamate. Furthermore, when NMDA antagonists were maintained in the presence of glutamate, excitatory transmission was again restored and the new responses were now insensitive both to 'NMDA' and 'non-NMDA' antagonists. Once elicited, the changes in synaptic transmission appear to be 'memorized' by the slice for at least tens of minutes. These phenomena suggest that the well-known plasticity of the hippocampal synapses may involve a sequence of distinct 'states', and that transitions between these states can be induced by certain pharmacological stimuli.

A glial toxin reduces effects of gold thioglucose on the hypothalamus and area postrema

To see if DL-alpha amino adipate (DL-AA) an agent known to cause transient swelling and dysfunction of hypothalamic glia, would affect the ability of goldthioglucose (GTG) to cause hypothalamic lesions, groups of male, 100 g rats were given saline, 0.5 mg/g of DL-AA, or 0.75 mg/g of DL-AA 30 min before injection of 0.4 mg/g of GTG. The incidence of GTG lesions was significantly less in the DL-AA treated groups than in the saline treated group. Histochemical staining of brain tissue revealed the existence of a population of specialized iron-containing glia with an anatomical distribution similar to the specific areas in the hypothalamus area postrema most sensitive to GTG. The results suggest that GTG lesion formation may result from an effect of GTG on glia.

Putative gliotoxin, alpha-aminoadipic acid, fails to kill hippocampal astrocytes in vivo

The putative gliotoxic compound, alpha-aminoadipic acid (AAA), has previously been reported to be a glial-selective toxin exerting its effects both in vivo and in vitro. This study sought to examine the effects of this compound on adult rat hippocampal astrocytes in vivo. Fourteen adult male Sprague-Dawley rats were injected with high concentrations of either the L-isomer or the racemic mixture of AAA into the dorsal hippocampal formation or corpus striatum. The animals were allowed to survive for times ranging from 4 to 48 h and the possible toxicity of the acid on astrocytes was monitored by glial fibrillary acidic protein (GFAP) immunocytochemistry. In every case, the GFAP-positive cells at the level of the injection site appeared identical, both in morphology and cell density, to control preparations. Based on our results, we conclude that AAA is not toxic to adult rat hippocampal astrocytes in vivo, nor could we confirm previous reports of toxicity to striatal astrocytes.

Pharmacological dissociation of the b-wave and pattern electroretinogram

Electroretinographic responses to modulation of either luminance (focal ERG) or spatial contrast (pattern ERG) were recorded from the pigeon eye before and after intravitreal injection of glutamate analogues DL alpha amino adipic acid (DL alpha AA) and 2-amino-4 phosphonobutyric acid (APB). Both toxins reversibly abolished the b-wave. The pattern ERG was still present, however, when the b-wave had been abolished by the toxins. This result demonstrates that the b-wave and the pattern ERG can be pharmacologically dissociated and suggests the possibility that in pigeons the b-wave and pattern ERG reflect the activity of different generators.

Descending and segmental spinal pathways differently regulate contractions of antagonistic muscles of frogs in vitro

A comparison of fibres descending through the ventral column and segmental afferent fibres was carried out using an isolated frog spinal cord-nerve-muscle preparation with a pair of antagonistic muscles. The m. tibialis anterior (TIB) was contracted with stimulation of the ventral column much stronger than with that of the dorsal root (DR). Contraction of the m. gastrocnemius (GAS) was selectively induced with DR stimulation. These results suggest that the descending fibres and segmental afferent fibres preferentially regulate contractions of the flexor (the TIB) and the extensor (the GAS), respectively. Effects of alpha-aminoadipate, methysergide and LSD were also examined.

Activation of excitatory amino acid receptors may mediate the folate-induced stimulation of locomotor activity after bilateral injection into the rat nucleus accumbens

Folic acid (FA) and 5-formyltetrahydrofolic acid (FTHF) have been shown previously to produce a marked stimulation of locomotor activity after bilateral injection into the rat nucleus accumbens. This study was designed to determine whether the hypermotility response produced by the folates is mediated through the activation of excitatory amino acid receptors in the nucleus accumbens. Although FA stimulated locomotor activity, pteroic acid, a congener of FA that lacks the glutamate moiety, was ineffective, suggesting that the glutamate portion of the molecule is essential for the hypermotility response. The N-methyl-D-aspartic acid (NMDA) receptor antagonists, D-alpha-aminoadipic acid, DL-alpha-epsilon-diaminopimelic acid and MgCl2, at doses that attenuated NMDA-induced hypermotility, were ineffective in decreasing the folate-induced hypermotility response. This behavioral observation is consistent with the biochemical observation that the folates, at a 1 mM concentration, were unable to stimulate the release of [3H]acetylcholine from striatal slices, a model system that is sensitive to the activation of NMDA receptors. In contrast to the ineffectiveness of the NMDA antagonists in inhibiting the response to the folates, the antagonist, glutamic acid diethylester, which inhibited the response to quisqualic acid, but not NMDA, also inhibited the response to both FA and FTHF. Two recently characterized dipeptides, gamma-D-glutamylaminomethylsulfonic acid and gamma-D-glutamyltaurine, antagonized the stimulation of locomotor activity produced by quisqualic acid, FA and FTHF. However, these dipeptides also inhibited the response to NMDA, suggesting that these compounds are not able to distinguish between quisqualate and NMDA receptors in the nucleus accumbens.(ABSTRACT TRUNCATED AT 250 WORDS)

L-[3H]lysine binding to rat retinal membrane: II. Effect of kainic acid, D,L-alpha-aminoadipic acid, iodoacetic acid, and modification by dark-exposure

The rat retina and the different brain regions contain membranes sites that bind L-lysine in the nanomolar range. These binding sites undergo changes in different experimental conditions, thus: intraocular injection of kainic acid induces a reduction of the density of L-lysine binding sites, D,L-alpha-aminoadipic acid injected into the eye enhances both kinetic parameters (Bmax and Kd) of L-[3H]lysine binding sites, the intraperitoneal injection of iodoacetic acid decreases the sensitivity for its ligand binding sites, and the exposure to darkness of the rats reduces L-[3H]lysine binding in the retina, thalamus, hypothalamus and superior colliculus, but not in the occipital cortex; such a decrease appears to be characterized, at least in the retina, by a lower sensitivity of the binding sites for L-lysine after the exposure to darkness. The results show that L-lysine binding sites are located on kainic acid-sensitive cells and can be involved in the physiological mechanism of vision.

Glucose represses formation of delta-(L-alpha-aminoadipyl)-L-cysteinyl-D-valine and isopenicillin N synthase but not penicillin acyltransferase in Penicillium chrysogenum

The content of alpha-aminoadipyl-cysteinyl-valine, the first intermediate of the penicillin biosynthetic pathway, decreased when Penicillium chrysogenum was grown in a high concentration of glucose. Glucose repressed the incorporation of [14C]valine into alpha-aminoadipyl-cysteinyl-[14C]valine in vivo. The pool of alpha-aminoadipic acid increased sevenfold in control (lactose-grown) penicillin-producing cultures, coinciding with the phase of rapid penicillin biosynthesis, but this increase was very small in glucose-grown cultures. Glucose stimulated homocitrate synthase and saccharopine dehydrogenase activities in vivo and increased the incorporation of lysine into proteins. These results suggest that glucose stimulates the flux through the lysine biosynthetic pathway, thus preventing alpha-aminoadipic acid accumulation. The repression of alpha-aminoadipyl-cysteinyl-valine synthesis by glucose was not reversed by the addition of alpha-aminoadipic acid, cysteine, or valine. Glucose also repressed isopenicillin N synthase, which converts alpha-aminoadipyl-cysteinyl-valine into isopenicillin N, but did not affect penicillin acyltransferase, the last enzyme of the penicillin biosynthetic pathway.

Fine structural changes in the rat brain after local injections of gliotoxin, alpha-aminoadipic acid

Toxic effects of locally injected gliotoxin, L-alpha-aminoadipic acid, were examined in the adult rat brain at the fine structural level. Astrocytes were the only cell type in the corpus striatum that showed clear structural degeneration 1 to 3 days after the treatment. With a 7 day survival period, two major structural changes were detectable, including tremendous invasion of astrocytes and their processes, and an appearance of a large number of microglia around injection sites. Both the stereospecificity and cellular specificity were firmly established on the toxic effects of this chemical.

Pharmacologically distinct glutamate receptors on cerebellar granule cells

Cultured cerebellar granule cells were found to exhibit calcium-dependent release of 3H-D-aspartate when stimulated with excitatory amino acids. L-glutamate and L-aspartate were found to be potent stimulators of 3H-D-aspartate release, D-aspartate was weaker and only minor effects were seen with D-glutamate, quisqualate, kainate, N-methyl-D-aspartate (NMDA) and L-alpha-aminoadipate (L-alpha AA). It was also found that only L-glutamate and L-aspartate showed high affinity for the 3H-L-glutamate binding sites on granule cell membranes. Stimulation by L-glutamate of 3H-D-aspartate release could be blocked by various excitatory amino acid antagonists. From the relative potencies of agonists and antagonists on D-aspartate release it is suggested that cerebellar granule cells express functionally active glutamate receptors with pharmacological characteristics different from all known excitatory amino acid receptors.

A microiontophoretic study on the nature of the putative synaptic neurotransmitter involved in the pedunculopontine-substantia nigra pars compacta excitatory pathway of the rat

The nature of the synaptic transmitter involved in the excitatory fibers linking the nucleus tegmenti pedunculopontinus (PPN) to the pars compacta of the substantia nigra (SNPC) was investigated using microiontophoretic techniques in rats anesthetized with ketamine. Among the SNPC cells activated orthodromically by PPN electrical stimulation, only a few cells were weakly excited by iontophoretically administered acetylcholine (Ach) while most were not affected. Conversely all cells were promptly and powerfully excited by short pulses of glutamate (GLU). The administration of the GLU antagonists glutamic acid diethyl ester (GDEE) and D-alpha-aminoadipic acid (DAA) reversibly and simultaneously suppressed both the PPN-evoked orthodromic response and the GLU-induced excitation of SNPC cells without affecting their response to iontophoretic Ach. GDEE was more effective than DAA in counteracting the synaptically evoked excitation. On the other hand, atropine, while antagonizing the Ach response in those cells which were cholinoceptive, did not affect either the PPN-evoked or the GLU-induced excitation. Hence, despite the presence of cholinergic cells in the PPN region, Ach does not appear to be involved in the excitatory PPN-SNPC pathway. The present findings suggest that the excitatory PPN fibers innervating the SNPC may utilize GLU or a closely related amino acid as a neurotransmitter.

Induction of glutamate binding sites in hippocampal membranes by transient exposure to high concentrations of glutamate or glutamate analogs

The number of Na+-independent, Cl--dependent glutamate binding sites in rat hippocampal membranes is increased two- to fourfold after pre-exposing isolated membranes or hippocampal slices to high concentrations (0.1-10 mM) of L-glutamate or of glutamate analogs with high affinity for this binding site, such as quisqualate, homocysteate, or aminoadipate. N-Methylaspartate and kainate are ineffective. A similar binding increase is induced by transient exposure to the dipeptide tyrosylglutamate. The newly induced binding sites appear to be identical with pre-existing Cl--dependent binding sites by several criteria: They have a similar pharmacological profile, they are sensitive to low concentrations of Na+, and the number of sites can be further increased by transient exposure to micromolar calcium concentrations. Moreover, binding of [3H]APB, a ligand selective for the Cl--dependent glutamate binding sites, is also increased after glutamate preincubation. The induction of binding sites by high glutamate concentrations, described herein, is calcium-independent, not inhibited by leupeptin and, therefore, different from the previously described activation of binding sites by a calcium-sensitive protease. The high concentration of ligand needed to induce increased binding suggests the presence in hippocampal membranes of a binding site with low, millimolar affinity that is functionally related to the known high-affinity binding sites. Several interpretations of the observed effects and their implications for the possible relationship between the binding site and the synaptic receptor are discussed.

Specific binding of L-[3H]-glutamic acid to rat substantia nigra synaptic membranes

The specific binding of L-[3H]-glutamic acid (GLU) was investigated in synaptic membranes from rat substantia nigra. L-[3H]-GLU binding to the membrane preparations occurred in a reversible and saturable way. The specific binding was stimulated by the presence of CaCl2 and was reduced by freezing and thawing the membranes. Scatchard analysis of the saturation isotherms yielded a non-linear plot suggesting that the binding reaction does not occur through a simple bimolecular association. Assuming non-interacting binding sites, a high (KD1, 139 nM; Bmax1, 3.5 pmoles/mg protein) and a low (KD2, 667 nM; Bmax2, 15.1 pmoles/mg protein) affinity L-[3H]-GLU binding site were obtained. The kinetics of dissociation of bound L-[3H]-GLU was biphasic; the respective dissociation rate constant (k-1) being 0.20 min-1 and 0.013 min-1. A series of amino acid receptor agonists and antagonists were tested as inhibitors of L-[3H]-GLU specific binding. Quisqualic acid, L-GLU and D-alpha-aminoadipate (D-alpha-AA) were the most potent inhibitors. DL-2-amino-4-phosphonobutyrate (APB), N-Methyl-D-aspartate (NMDA) and D-GLU were moderate inhibitors, whereas diaminopimelic acid (DAPA) and glutamate diethyl ester (GDEE) exhibited the lowest relative potency. Kainic acid (KA), gamma-aminobutyric acid (GABA) and bicuculline were not able to modify at any concentration used the specific binding of L-[3H]-GLU. These data demonstrate the presence of specific GLU binding sites in synaptic structures at substantia nigra level and support the idea that excitatory amino acids may play a role in synaptic transmission in this brain region.

Pharmacological identification of retinal cells releasing taurine by light stimulation

The effect of drugs blocking synaptic activity at different retinal levels was examined in this study, in an attempt to identify the origin of the light-stimulated release of 3H-taurine from the chick retina. It was determined by autoradiography that the chick retina accumulates taurine in photoreceptors, in cells from the inner nuclear layer, and in processes of the inner plexiform layer. All these are possible sites for the release of taurine upon illumination. To discriminate among these possibilities, the effects of aspartate, tetrodotoxin, strychnine, picrotoxin, chlorpromazine, tubocurarine, atropine, glutamate diethyl esther, alpha-amino adipate and 2-amino-4-phosphonobutyrate were studied. Aspartate (10 mM), which is known to eliminate the light response of cells postsynaptic to photoreceptors, induced a marked increase of 150% in the resting efflux of 3H-taurine but did not decrease significantly the light-stimulated release. Tetrodotoxin, which blocks amacrine cell responses, decreased 3H-taurine release stimulated by light by less than 20%. The efflux of taurine was unaffected by strychnine, picrotoxin, tubocurarine, atropine, chlorpromazine, and 2-amino-4-phosphonobutyrate, whereas it was increased by glutamate diethyl esther and alpha-amino adipate. These results, all together, point to photoreceptors as the cells releasing 3H-taurine in response to light.

Effects of acidic amino acid antagonists upon the spectral properties of carp horizontal cells: circuitry of the outer retina

The acidic amino acid receptor antagonists, alpha-methylglutamate and alpha-aminoadipate, were applied to the carp retina to study their effects upon the spectral properties of horizontal cells and to elucidate the synaptic connections between horizontal cells and cones. Application of these antagonists strongly hyperpolarized the L-type cone horizontal cells and reduced the responses of these horizontal cells to red light more than to blue light. Application of Co2+ ions to the retina, a procedure which decreases transmitter release, also hyperpolarized the L-type cone horizontal cells but reduced the response of these horizontal cells to red and blue lights equally. These results suggest that red- or long wavelength-sensitive cones release a different transmitter onto L-type cone horizontal cells than do short wavelength-sensitive cones. Application of the acidic amino acid antagonists also revealed details of the feedback pathway from L-type cone horizontal cells to cones. Previous studies have shown that feedback varies directly with stimulus size and that the effects of feedback on the responses of cones are observed as a transient waveform at response onset (a large, hyperpolarizing potential that is quickly followed by a smaller plateau potential). Application of the acidic amino acid antagonists at a dose which partially hyperpolarized the horizontal cells selectively enhanced the response of the cells to blue lights, when full field, and not spot, stimuli were used. The antagonists also eliminated the transient at response onset. These findings are consistent with the presence of a feedback pathway from L-type cone horizontal cells to short wavelength cones but not to long (red-sensitive) cones.

Stereospecific effects of the alpha-aminoadipic acid on the retina: a morphological and electrophysiological study

In both frog and chicken an intravitreal injection of the dextrorotatory (D)-isomer of alpha-aminoadipic acid (alpha-aaa) leads to a progressive disappearance of the ERG b-wave without affecting a and c components. Tectal evoked potentials (TEP) are no longer recorded. These physiological effects are concomitant with a specific glial cell damage, without any apparent damage to neurons. The levorotatory (L)-isomer at low concentrations is more gliotoxic than the D-isomer, the ERG b-wave is suppressed, while the amplitude of both a and c components is increased. TEPs are always recorded, i.e., a visual message is still generated in the retina and transmitted to the optic tectum when the Müller cells have been damaged and the b-wave is abolished. At higher concentrations the L-isomer suppresses TEPs and damages both glial and neuronal cells. Thus alpha-aaa appears to be a good tool for analyzing ERG components, especially subcomponents of the c-wave.

Combined effects of DL-alpha-aminoadipic acid with sodium iodate, ethyl alcohol, or light stimulation on the ERG c-wave and on the standing potential of albino rabbit eyes

Albino rabbits were treated with intravitreal injections of DL-alpha-aminoadipic acid (alpha-AAA) into one eye (0.1 ml of a 0.15 M solution) and 0.1 ml of saline into the contralateral eye. Thirteen to fourteen hours later the DC electroretinogram (ERG) and/or the standing potential (SP) were recorded. (1) In eight of nine animals the c-wave amplitude of alpha-AAA injected eyes was increased compared with that of control eyes. Following intravenous injection of Sodium Iodate (40 mg/kg in 2% solution) the c-waves of both eyes were rapidly replaced by negative potentials. In 8 of 9 animals the amplitude reduction was more marked in alpha-AAA-treated eyes than in control eyes, but the final amplitude was higher in the former than in the latter. The SP was reduced with difference in curve form but not significantly in amplitude between the eyes. (2) In nine other rabbits iv.-injected ethyl alcohol (0.4 g/kg in 20% solution) provoked a transient increase of SP level and c-wave amplitude in control eyes and smaller but similar changes in alpha-AAA injected eyes. (3) In another five animals the SP was recorded following a step from darkness to continuous light stimulation. The light peak was less pronounced in alpha-AAA treated eyes than in control eyes.

Effect of glutamate analogues on brain tumor cell lines

Glutamate analogues have been used in many different experimental approaches in neurobiology. A small number of these analogues have been classified as gliotoxic. We have examined the effect of seven glutamate analogues (five gliotoxic and two neurotoxic) on the growth and viability of four human glioma cell lines, one human medulloblastoma cell line, and one human sarcoma cell line. Aminoadipic acid and homocysteic acid predominantly affected the growth of two glioma cell lines in the presence of 4 mM glutamine. Phosphonobutyric acid predominantly affected the other two glioma cell lines and the medulloblastoma cell line in the presence of 4 mM glutamine. In medium containing no glutamine, all three analogues had marked effects on all the cell lines except the sarcoma cell line. These effects were dose dependent. We postulate that these results can in part be explained on the basis of metabolic compartmentalization.

Influence of glutamine on the growth of human glioma and medulloblastoma in culture

Cellular supply of glutamine, an essential substrate for growth, is derived from extracellular fluid and de novo synthesis. We investigated the relative importance of these sources to the growth of six human anaplastic glioma- and one human medulloblastoma-derived permanent cell lines. Exogenous glutamine was limiting for the proliferation of glioma-derived lines D-54 MG, U-118 MG, and U-251 MG. In contrast, medulloblastoma-derived line TE-671 and glioma-derived lines U-373 MG, D-245 MG, and D-259 MG grew in the absence of supplemental glutamine. Two cell lines with contrasting glutamine requirements, D-54 MG and TE-671, were used to explore the pharmacological interference with glutamine metabolism. DL-alpha-Aminoadipic acid, a reported glutamic acid analogue with gliotoxic properties, significantly inhibited the growth of both lines. These effects were reversed by increasing glutamine, suggesting that the major action of DL-alpha-aminoadipic acid is as a glutamine antagonist. In contrast, the glutamine synthetase inhibitor delta-hydroxylysine demonstrated activity only against TE-671. Acivicin and 6-diazo-5-oxo-L-norleucine, glutamine analogues available for clinical use, reduced the proliferation of both cell lines at pharmacological concentrations. Methionine sulfoximine, a glutamine synthetase inhibitor previously used clinically, produced marked growth inhibition only against TE-671. These findings indicate that the synthesis and utilization of glutamine are potentially exploitable targets for the chemotherapy of some human gliomas and medulloblastomas.

Gangliosides in postmitotic retina of chick embryo: changes in vivo and in cell cultures

Developmental changes in ganglioside composition of postmitotic neural retina of chick embryo were analyzed by thin-layer chromatography. Gangliosides were identified by comparing their chromatographic mobilities with reference standards. The outstanding changes are decrease in the concentration of GD3L and increase in GD1a and GM1 concentrations. By depleting Müller glia cells from retina tissue of 13- and 16-day embryos (R13, R16) we determined that the bulk of the major gangliosides is associated with the neurons. Analysis of gangliosides in monolayer cultures of R13 and R16 cells highly enriched for Müller cell-derived gliocytes indicated that these cells express the same types of gangliosides as neurons, but in somewhat different concentrations and relative proportions; however, after time in culture these cells showed ganglioside types and changes in ganglioside profile that are not characteristic of normal retina. The latter observation is consistent with other evidence that the phenotype of Müller glia cells becomes altered in monolayer culture. In contrast to cultures of early embryonic retina, in organ cultures of later postmitotic retina, ganglioside composition did not continue to change as in normal development. This suggests that in postmitotic retina, normal developmental progression of ganglioside changes requires systemic and/or other conditions which are missing or altered when this tissue is isolated and cultured in vitro.

Effects of alpha-aminoadipic acid on the glutamate-isolated P III of the rabbit electroretinogram

alpha-Aminoadipic acid was intravitreally applied to adult rabbits. After 5 h, the retinae of these animals were examined by electroretinography and histochemistry. The retinal Müller cells were extremely swollen, and the electroretinographic slow P III was extinguished. The mass receptor potential was somewhat diminished. The results are consistent with the opinion that the slow P III is the reaction of the Müller cells to the changed external potassium ion concentration caused by the activity of the photoreceptors.

Mg2+ reduces N-methyl-D-aspartate neurotoxicity in embryonic chick neural retina in vitro

N-Methyl-D-aspartate (NMDA) is a potent neurotoxin that affects cells in the inner layers of the embryonic chick retina exposed in vitro. After exposure of the embryonic day 12 neural retina to 0.5-10.0 mM NMDA for 30 min, 50-80% of the cells in the inner region of the inner nuclear layer and 50-100% of the cells in the ganglion cell layer were hypochromatic. When retinas were incubated with Mg2+ (0.5-10.0 mM) for 15 min and then incubated with Mg2+ and NMDA (0.5 mM) for 30 min, the NMDA effect in the inner layers was dramatically reduced but not abolished. Removal of Mg2+ before NMDA exposure produced retinas as seriously affected as retinas not exposed to Mg2+. Studying the effects of NMDA inhibitors, such as Mg2+, may help elucidate the mechanism of the cytotoxic events that occur in the retina in response to certain excitatory acidic amino acids.

Blockade of hyperbaric oxygen induced seizures by excitatory amino acid antagonists

The effectiveness of several excitatory amino acid antagonists to delay or block seizures induced by oxygen at high pressure was examined in mice. Of the antagonists tested, namely, L-proline, DL-alpha-aminoadipate, DL-2-amino-5-phosphonovalerate, and L-glutamic acid diethyl ester, DL-2-amino-5-phosphonovalerate was the most effective in delaying or preventing seizures. L-Glutamic acid diethylester was also effective but at significantly higher doses, which were also associated with marked sedation.

Pharmacological activation of locomotor patterns in larval and adult frog spinal cords

The effects of amino acids, catecholamines, and their agonists shown to elicit locomotor activity in several vertebrate species were examined in spinal animals and isolated nervous systems of developing tadpoles (Rana catesbiana) and adult frogs (R. catesbiana and pipiens). Elicited activity was correlated in spinal animals by video and electromyographic analysis, and in in vitro spinal cords by recordings of tail and hindlimb motor activity. Of the agents tested, only N-methyl-DL-aspartate (NMA), an amino acid agonist, was effective in eliciting motor activity in spinal animals. In isolated nervous systems, both NMA and D-glutamate added to the bath activated locomotor activity. NMA injected i.p. into tadpoles with high spinal cord transections elicited coordinated swimming motor activity in axial and hindlimb muscles that was roughly typical for the stage of development of the animal. In late stage tadpoles (st. XX), NMA also elicited wiping and alternating or synchronous (i.e. kicking or jumping) hindlimb movements. Addition of NMA or glutamate to a bath containing an in vitro tadpole spinal cord preparation elicited ventral root motor activity characteristic of swimming, but without a rostrocaudal phase lag. Rhythmic activity thought to underlie stepping and kicking was seen in lateral ventral rootlets innervating the hindlimbs. In adult frogs with high spinal cord transections, injection of NMA elicited a general sequence of spontaneous hindlimb motor functions: reflex wiping, stepping, and kicking or jumping. Isolated frog spinal cords were not responsive to bath applied NMA, under the present conditions. The activation by amino acids or their agonists of different motor functions in both larval and adult frogs, as well as in higher and lower vertebrates, suggests a general significance of amino acid-activated receptors in the neural networks controlling locomotor function.

The excitation of frog motoneurones in vitro by the glutamate analogue, DI-alpha-amino-3-hydroxy-5-methyl-4-isoxazole-propionic acid (AMPA), and the effect of amino acid antagonists

Intracellular recordings were made from motoneurones of the frog spinal cord in vitro and the excitatory effects of the glutamate analogue DL-alpha-amino-3-hydroxy-5-methyl-4-isoxazole-propionic acid (AMPA) examined. AMPA (25-100 microM) depolarized motoneurones with an increase in input conductance and an increase in spontaneous cell firing. The time-course of the AMPA response was prolonged compared to L-glutamate. D-Aminophosphonovalerate (1 microM) and D-aminoadipate (100 microM) did not antagonize AMPA-induced excitations while depressing L-glutamate responses. Glutamate diethylester up to 1 mM was not found to be a useful amino acid antagonist since it did not significantly reduce the depolarizations to quisqualate, L-glutamate and AMPA. The results are compatible with an action of AMPA on receptors distinct from those to N-methyl-D-aspartate and shared by quisqualate.