Pharmacological properties of the postsynaptic inhibition by Purkinje cell axons and the action of γ-aminobutyric acid on Deiters neurones

Summary 1.Various drugs were applied intravenously or into the fourth ventricle and their effects upon the inhibition exerted by Purkinje cell axons were appraised by both extra- and intracellular recording from Deiters neurones. Strychnine, picrotoxin, pentamethylenetetrazol, -methyl--ethylglutarimide, noradrenaline, dopamine, dibenamine and nethalide did not affect this inhibition. 2.-aminobutyric acid (GABA) and inhibitors of GABA transaminase were applied iontophoretically into the vicinity of Deiters neurones through an outer barrel of coaxial electrodes, the effects being observed either intra- or extracellularly through an inner barrel. 3. GABA depressed both inhibitory and excitatory postsynaptic potentials and often blocked the spike potentials, while it increased the membrane conductance. 4. GABA also produced a membrane hyperpolarization of 3–8 mV. Concomitantly both the spike potential and after-depolarization increased in amplitude and the after-hyperpolarization decreased. 5. In a few cases hydroxylamine but not amino-oxyacetic acid potentiated the inhibition, there being an increase in the inhibitory postsynaptic potentials thereby induced. 6. These effects were considered in connection with the possibility that GABA acts as a natural transmitter.     

A pharmacological study of the depression of spinal neurones by glycine and related amino acids

Summary An analysis has been made in anaesthetised cats of the depression by glycine and related amino acids of the firing of spinal dorsal horn interneurones, Renshaw cells and cortical neurones. In general, electrophoretically administered glycine was a more potent depressant of interneurones than GABA. The reverse was true for cortical neurones, whereas these two amino acids were approximately equally effective upon Renshaw cells. Strychnine blocked the depressant action of - and -amino acids, but not that of - and higher -amino acids. Only convulsants having a strychnine-like effect on spinal post-synaptic inhibition blocked the action of glycine. The depression of spinal neurones produced by glycine or GABA was not affected by structural analogues of glycine and GABA that were not depressants, or by substances influencing amino acid transport systems. Some evidence was obtained for the enzymic inactivation of electrophoretically administered glycine in spinal tissue. The results are discussed in terms of the involvement of a glycine-like amino acid as a major spinal inhibitory transmitter.Supported by a grant from the Swiss Academy of Medical Sciences.     

The inhibition of hypoglossal motoneurones by impulses in the glossopharyngeal nerve of the rat

Summary Stimulation of the glossopharyngeal nerve produced inhibitory post synaptic potentials in neurones of the hypoglossal nucleus of the rat. The corresponding pause in the amino acid induced firing of these cells was antagonized by electrophoretically administered strychnine but not by bicuculline. Strychnine and bicuculline antagonized the actions of glycine and GABA respectively on these cells. The findings suggest that the transmitter for this synaptic inhibition is a glycine-like amino acid.C. J. Martin Travelling Fellow.Wellcome Trust/Animal Health Trust Research Fellow.     

Bicuculline and thalamic inhibition

Summary A study was made of the effects of electrophoretically administered bicuculline on the recurrent inhibition of thalamo-cortical relay neurones of the ventrobasal complex and the dorsal lateral geniculate nucleus of decerebrate and anaesthetised cats (pentobarbitone sodium). Electrophoretically administered glycine, GABA and taurine depressed the firing of thalamic neurones. GABA was usually the most potent depressant, glycine the least. The effects of GABA and taurine were blocked by bicuculline, those of taurine and glycine were blocked by strychnine. The inhibition of the firing of lateral geniculate neurones by electrical stimulation of the visual cortex was reduced by bicuculline but not by strychnine. The inhibition of these cells following an optic nerve volley, and the inhibition of ventrobasal neurones by stimulation of the primary somatosensory cortex, were also reduced by bicuculline.These findings suggest that intrathalamic inhibitory neurones release GABA as a transmitter at synapses on thalamo-cortical relay neurones. There is insufficient evidence to assess the significance of glycine and taurine as inhibitory transmitters in these thalamic nuclei.     

The inactivation of extracellularly administered amino acids in the feline spinal cord

Summary The mercurials, p-hydroxymercuribenzoate and p-chloromercuriphenylsulphonate, administered electrophoretically from multi-barelled micropipettes, potentiate the depressant action of similarly administered glycine on feline spinal neurones. In addition, these mercurials inhibit the transport of glycine into rat brain slices. Neither action is very specific for glycine, since slightly higher concentrations of p-chloromercuriphenylsulphonate than those required to potentiate glycine-induced depression also potentiate depression induced by GABA, -alanine, Land D--alanine, and even higher concentrations enhance the excitant action of acidic amino acids. p-Hydroxymercuribenzoate also inhibits the uptake of GABA, DL-aspartate and L-lysine by brain slices. The potentiation by the mercurials of amino acid-induced effects is considered likely to be the result of inhibition of transport processes rather than enzymic activities.Thiosemicarbazide, administered electrophoretically and intravenously, does not enhance the effects of amino acid excitants or depressants on spinal interneurones. Hydrazinopropionic acid, a potent inhibitor of GABA transaminase, does not enhance GABA-induced depression of spinal interneurones when administered electrophoretically.These findings suggest the importance of transport processes in the removal of amino acids from the synaptic environment, and evidence is discussed that these processes are likely to differ in detail from the observed gross transport of amino acids into tissue slices.     

Glycine and GABA as inhibitory transmitters in the medullary reticular formation stStudies involving intra- and extra-cellular recording

Summary Studies involving intracellular recording during extracellular drug administration have revealed that both glycine and GABA hyperpolarize and increase the membrane conductance of reticulospinal neurones, glycine usually being more effective than GABA. The inhibition of firing of some medullary reticular neurones produced by stimulating the ventrolateral area of the spinal cord (where reticulospinal axons lie) is blocked by strychnine, which blocks the depression by glycine without affecting that by GABA. Such inhibition is not reduced by bicuculline or picrotoxin in concentrations which block the action of GABA, but do not affect that of glycine. The synaptic inhibition of other reticular neurones is blocked by bicuculline or picrotoxin, but not by strychnine. Thus, both glycine and GABA are likely to be inhibitory transmitters in the medullary reticular formation, possibly released from axon collaterals of two different populations of inhibitory reticulospinal neurones.     

Actions of amino acids and convulsants on bulbar reticular neurones

Summary A study has been made of the actions of microelectrophoretically administered amino acids and convulsants on spontaneous and glutamate- (or DLH-) induced firing of bulbar reticular neurones in unanaesthetized, decerebrate, cats. DLH was a more potent excitant than glutamate and aspartate on almost all the neurones tested. Although glutamate was usually more effective than aspartate, their relative potencies were often similar. Glycine, -alanine and GABA depressed the majority of neurones tested (93%, 89% and 75%, respectively), and had no action on the remainder. Glycine was invariably more potent than -alanine, which was usually more potent than GABA. Strychnine reversibly blocked the depressant actions of glycine and -alanine but not that of GABA. Electrophoretically administered picrotoxin slightly reduced the depression caused by glycine and GABA on less than half the neurones tested. Intravenously injected picrotoxin (0.3–5 mg/kg) did not block the effects of these amino acids. A comparison of the results with those obtained in the spinal cord provides some evidence that glycine may be an inhibitory transmitter substance released on bulbar reticular neurones.Supported by a grant from the Swiss National Foundation of Scientific Research.     

The hyperpolarization of spinal motoneurones by glycine and related amino acids

Summary Electrophoretically administered glycine, -alanine and GABA hyperpolarize spinal motoneurones in cats anaesthetized with pentobarbitone. The reversal potential for these hyperpolarizations is similar to that of inhibitory postsynaptic potentials. Alterations in intracellular K⁺ and Cl^– ion concentrations, and intracellular injection of a series of anions of different hydrated ion size, affect inhibitory and amino acid potentials in the same fashion. Hence it is probable that glycine, -alanine and GABA produce an alteration in membrane permeability similar to that produced by spinal inhibitory synaptic transmitters. Strychnine reversibly blocks the action of inhibitory transmitters, glycine and -alanine, but is without effect on the hyperpolarizing action of GABA.These results indicate that glycine may be a major spinal inhibitory transmitter, in which case strychnine affects spinal postsynaptic inhibition by limiting the action of glycine upon subsynaptic inhibitory receptors.Supported by a grant from the Swiss Academy of Medical Sciences.     

Inhibitory control of intracerebellar nuclei by the Purkinje cell axons

Summary In anaesthetized cats, synaptic events in cerebellar nuclei neurones were investigated with intracellular microelectrode techniques. These cells were identified by their antidromic activation along their axons and/or by their location in histological sections. In the cells of lateral nucleus IPSPs were induced monosynaptically during stimulation of the overlying hemispheral cortex of the cerebellum. In the cells of nuclei interpositus and fastigii, similar IPSPs were produced from the paravermal and vermal cortices, respectively. The postulate that the Purkinje cells exert an inhibitory action upon their target neurones thus applies not only to Deiters neurones, as previously proposed, but also to cells in the cerebellar nuclei. Stimulation of the cerebellar afferents at the inferior olive, the pontine nucleus and the lateral reticular nucleus produced EPSPs in cerebellar nuclei cells with relatively brief latencies, probably through axon collaterals of these afferents. The EPSPs were followed by IPSPs and slow depolarizations of disinhibitory nature, which, as studied previously in Deiters neurones, might be caused respectively by activation and subsequent depression of Purkinje cells through the cerebellar intracortical mechanisms.     

The in vivo inactivation of GABA and other inhibitory amino acids in the cat nervous system

Summary In cats anaesthetised with pentobarbitone, the effect of inhibitors of the in vitro cellular uptake of GABA were tested on the responses of single central neurones to GABA and other depressant amino acids. (+)- And (-)-nipecotic acid, (+)-2,4-diaminobutyric acid (DABA) and 2,2-dimethyl--alanine, enhanced the action of GABA on spinal, cerebellar and cerebral cortical neurones.In the spinal cord DABA, and to a less extent (-)-nipecotic acid, enhanced the action of -alanine, whereas the actions of glycine and taurine were unaffected by DABA and reduced by (-)-nipecotic acid. In the cerebellum and cerebral cortex, these two substances enhanced the action of GABA, usually to a greater extent than that of -alanine and taurine, although this specificity was not marked.The GABA-mediated basket cell inhibition of Purkinje cells in the cerebellum was unaffected by DABA and (-)-nipecotic acid, and neither substance appears suitable for determining the role of uptake processes in the inactivation of synaptically released GABA.Quantitatively these in vivo results agree more closely with recent in vitro uptake studies in cat tissue than the previously published data on rat cerebral cortex and dorsal root ganglia, and the observations provide further evidence for the importance of cellular uptake in maintaining low extraneuronal concentrations of inhibitory amino acid transmitters.     

Differential strength of action of glycine and GABA in hypoglossus nucleus

Summary Glycine and GABA were applied iontophoretically in the hypoglossus nucleus and their effects on the antidromic field potential compared. Preliminary results indicate that GABA may act relatively stronger than glycine in the dorsal dendritic region of the nucleus whilst glycine is more effective in the center of the nucleus, where the somata of the neurones are located. Furthermore, records could be obtained in which GABA preferentially reduced a dendritic component of the antidromic field potential.There is a possibility that actions of glycine and GABA could be unequally distributed over the cell membrane.Supported by grant Br 242/7, Deutsche Forschungsgemeinschaft.     

Excitatory and inhibitory responses of Purkinje cells, in the rat cerebellum in vivo, induced by excitatory amino acids

Responses of rat cerebellar Purkinje cells to iontophoretically administered excitatory amino acids have been studied in vivo. Responses to N-methyl-D-aspartate (NMDA) were either biphasic (excitation followed by inhibition) or purely inhibitory and were antagonized by the selective NMDA-receptor antagonist, 2-amino-5-phosphonovalerate. Quisqualate and kainate either excited or induced biphasic responses, in these neurones, which were only reduced by amino acid antagonists that acted at non-NMDA receptors. The excitatory amino acid-induced inhibitions were also antagonized by the selective gamma-aminobutyric acid (GABA) antagonist, picrotoxin, suggesting that they were indirectly mediated via GABAergic inhibitory interneurones, which could be excited via NMDA and non-NMDA receptors.     

Axon reflex activation of Deiters neurones from the cerebellar cortex through collaterals of the cerebellar afferents

Summary When recording intracellularly from cat's Deiters neurones, stimulation of the anterior lobe of the cerebellar cortex produced excitatory postsynaptic potentials (EPSPs) monosynaptically, in addition to the inhibitory ones (IPSPs) that were identified previously as being produced via Purkinje cell axons. The EPSPs were induced bilaterally from a wide area of the anterior and posterior lobes of the cerebellum, in contrast to the IPSPs that were evoked only ipsilaterally, mainly from the vermal cortex. The latency of the EPSPs was slightly, but significantly, shorter than that of the IPSPs. The presynaptic impulses responsible for these EPSPs were represented by the discrete field potentials and also by unit spikes of individual fibres. The pathway for these EPSPs and presynaptic impulses was pursued by testing their interference, in the manner of impulse collision and refractoriness, with those induced from various spots within or outside the cerebellum. It is found that the excitatory fibres for Deiters neurones extend transversely, and probably longitudinally too, over the culmen and pass out of the cerebellum through cerebellar peduncles. The major portion of them appears to originate from the medulla and a minority from the spinal cord. It is postulated that cerebellar afferents from these structures have synapses with Deiters neurones via their collateral branches, through which a kind of axon reflex occurs to Deiters neurones during stimulation of the cerebellar cortex.     

Caudate stimulation and substantia nigra activity in the rat.

1. The responses of spontaneously active single neurones in the substantia nigra and overlying mesencephalic reticular formation have been analysed during the electrical stimulation of the ipsilateral caudate nucleus. Experiments were performed in rats anaesthetized with urethane or pentobarbitone. All recordings were made extracellularly with multi-barrelled glass micropipettes which were also used to test neuronal responsiveness to electrophoretically administered substances. The micropipette tip position was marked and the distribution of neurones studied has been analysed. 2. Single shock stimulation of the caudate nucleus inhibited neuronal activity in the substantia nigra (270/320 cells: mean latency 5-4 msec) and in the mesencephalic reticular formation (62/72 cells: mean latency 16-6 msec). However, these effects were often accompanied by periods of excitation. In pentobarbitone anaesthetized animals the latency and duration of these substantia nigra inhibitions was increased. 3. Compared with the zona reticulata, fewer neurones in the zona compacta of the substantia nigra responded to caudate stimulation in both urethane or pentobarbitone anaesthetized animals. 4. The activity of most cells was depressed by electrophoretically administered GABA or glycine and increased by acetylcholine or glutamate. Neurones of the mesencephalic reticular formation were less sensitive to GABA and glycine than substantia nigra neurones. Within the substantia nigra, both zona compacta and zona reticulata neurones were more sensitive to GABA than to glycine. Over-all, glutamate was a more potent excitant than acetylcholine (ACh). 5. Electrophoretic bicuculline methochloride (BMC) consistently reduced GABA but not glycine depression of substantia nigra neurones. Approximately twice as much BMC was required to reduce the endogenous inhibition of the same substantia nigra neurones and the amplitude of concomitantly evoked positive field potential as was required to abolish exogenous GABA responses. Some evoked substantia nigra inhibitions were resistant to BMC. 6. Electrophoretic strychnine consistently reduced glycine but not GABA depression of substantia nigra neurones, and did not modify caudate evoked inhibition of these neurones or the accompanying field potential. 7. The results support the concept of a slowly conducting caudato-nigral pathway which has both facilitatory and inhibitory components. The inhibitory pathway uses GABA as the neurotransmitter. The identity of the possible excitatory transmitter is unknown. The monosynaptic nature of this pathway is uncertain and the possible contribution of other bicuculline insensitive nigral inhibitory processes is discussed.     

The origin of cerebellar-induced inhibition of Deiters neurones III. Localization of the inhibitory zone

Summary By recording intracellularly from Deiters neurones of cats, there was a survey of those cerebellar areas that, when stimulated, produced inhibitory postsynaptic potentials (IPSPs) monosynaptically in Deiters neurones. The monosynaptic inhibitory area expanded longitudinally mainly along the ipsilateral vermal cortex of the anterior lobe. The ipsilateral cortex of the posterior lobe was also effective in inhibiting Deiters neurones though less prominently than the anterior lobe. The inhibitory fibers could be stimulated in the white matter of the cerebellum, predominantly in the ipsilateral side at rostral regions of nuclei fastigii and interpositus. It was further shown that the monosynaptic inhibition from the anterior and posterior lobes occurs chiefly in the dorsal portion of Deiters nucleus. Since in both the cerebellum and Deiters nucleus the spatial pattern of distribution of the inhibitory fibers conforms to that of the corticovestibular fibers as histologically defined, the experimental findings are in accord with the hypothesis that the cerebellar Purkinje cells are inhibitory in nature.     

Inhibition and facilitation of antidromically identified tubero-infundibular neurones following stimulation of the median eminence in the rat.

1. Stimulation of the median eminence of female rats inhibited the spontaneous firing of antidromically identified tubero-infundibular units. Some units could be inhibited by stimuli subthreshold for the antidromic spike. A conditioning stimulus of subthreshold intensity for the antidromic spike also delayed or abolished the invasion of the soma and dendrites of the same unit by antidromic action potentials evoked by a suprathreshold stimulus given 2-10 msec later. 2. Although strychnine (0-2 mg/kg, i.v.) did not significantly alter the inhibition evoked by stimulation of the median eminence, it was abolished by picrotoxin (2-6 mg/kg, iv.). 3. In seven of the fifty-seven identified units examined stimulation of the median eminence facilitated the spontaneous firing. However, after an i.v. injection of picrotoxin the facilitory response was observed in thirty-seven of the forty-six units tested. Post-stimulus time histograms obtained from the same unit before and after an injection of picrotoxin demonstrated that the latency and duration of the facilitation did not always coincide with that of the inhibition. 4. After an injection of picrotoxin the field potential evoked by antidromic stimulation of the median eminence consisted of a small positive wave followed by a negative wave. Frequently the negative wave of the field potential was accompanied by a convulsive discharge. The latency of the negative wave appears to be identical to that of the facilitation seen in nearby single units. 5. The facilitation evoked by antidromic stimulation in the presence of picrotoxin was blocked by an i.v. injection of alpha-methyl-p-tyrosine (250 or 375 mg/kg). None of the nine units sampled from rats pre-treated with alpha-methyl-p-tyrosine injected twice I.P. (250 mg/kg for each) were facilitated by stimulation of the median eminence following I.V. picrotoxin, while eight of the eleven units sampled from control rats pre-treated with L-tyrosine could be facilitated by antidromic stimulation. 6. These results suggest that tuber-infundibular neurosecretory neurones are inhibited and facilitated by neural pathways which could involve the axon collaterals of the neruosecretory neurones which project to the external layer of the median eminence. It is also suggested that GABA-releasing neurones mediate the inhibition and catecholaminergic neurones are involved in the facilitory pathways. Presumably the facilitation is normally masked by the activity of the GABA-releasing neurones.     

Acetylcholine sensitivity of cerebellar neurones in the cat

1. Cholinomimetics, acetylcholine antagonists and some other compounds of pharmacological interest were administered electrophoretically near neurones within the vermal cerebellar cortex of anaesthetized (pentobarbitone) and unanaesthetized (cerveau isolé) cats.2. The neurones were identified by position within the cortex, spontaneous activity, and the responses to afferent and antidromic stimulation.3. Purkinje cells, but neither granule nor basket cells, were excited by cholinomimetics, and the acetylcholine receptors had muscarinic properties. Excitation was often preceded by depression of the spontaneous firing.4. Intravenously administered atropine and dihydro-beta-erythroidine did not depress the synaptic excitation of cerebellar neurones evoked by impulses in mossy, climbing or parallel fibres.5. Acetylcholine is thus unlikely to be an excitatory transmitter within the feline cerebellum, particularly at mossy fibre-granule cell synapses, despite the presence of relatively high levels of acetylcholinesterase within mossy fibre terminals.     

Contribution of mossy fiber-granule cell pathway to the cerebellar-induced delayed inhibition in Deiters neurones

Summary In anaesthetized cats, electric pulse stimuli were applied at various lateralities to the anterior lobe of the cerebellum. In dorsal Deiters neurones delayed IPSPs with latencies of 3–6 msec were evoked from the entire area of the culmen including the paravermis bilaterally. The delayed IPSPs had a summit time of about 2 msec and a duration of about 7 msec. They showed a marked temporal facilitation and subsequent depression with double shock stimulation. Corticovestibular fibers were penetrated within the nucleus of Deiters and showed delayed, labile responses to cortical stimulation, corresponding to the delayed IPSPs in Deiters neurones. During stimulation of the anterior lobe at any laterality, field potentials recorded in the cerebellar cortex further revealed that there was activation, presumably through axon collaterals of mossy fibers, of granule cells and subsequently of Purkinje cells in the vermal cortex. Cortical events exhibited a prominent temporal facilitation and subsequent depression, in parallel with that observed for the delayed IPSPs in Deiters neurones. The delayed IPSPs in Deiters neurones arising from a wide area of the cerebellar cortex thus were attributed to activation through mossy fiber-granule cell pathway of Purkinje cells of the corticovestibular projection.     

Cerebellar-evoked disinhibition in dorsal deiters neurones

Summary Following the stimulation of cerebellar cortex, a slow depolarization developed in the neurones which were impaled with microelectrodes in the dorsal portion of the nucleus of Deiters. Characteristically, it was produced bilaterally from a wide area of the culmen and, with double shock stimulation at brief intervals, showed a marked potentiation, often in association with a later depression. After repetitive stimulation of the cerebellar cortex the slow depolarization was prolonged for a period of many seconds. Even stimulation of the spinal cord caused similar depolarization. By intracellular injection of currents and ions, the depolarization was shown to be disinhibition, i. e., removal of background inhibition. Accordingly, it was confirmed that there was a steady production of IPSPs in dorsal Deiters neurones, which diminished during the phase of disinhibition. As the possible source of these background IPSPs, the Purkinje cell axons within the nucleus of Deiters were found to be discharging rhythmically at a rate of 20–90/sec, and in fact they were depressed very effectively after cerebellar stimulation. At the same time, volleys along Purkinje cell axons produced by a testing cerebellar stimulation also were diminished, indicating a depression in the excitability of Purkinje cells.     

Electrophysiological and pharmacological studies of the inhibitory projection from the cerebellar cortex to the deep cerebellar nuclei in tissue culture.

In explant cultures of mouse cerebellum, short latency inhibitions can be recorded from neurones in the deep cerebellar nuclei after electrical stimulation in the cortical region. The responsible projection appears to originate from Purkinje cells and to be monosynaptic to cells in deep nuclei. In many cases, the inhibition is followed by a rebound excitation which has been attributed to a disinhibition. Known blockers of synaptic inhibition (bicuculline, bicuculline methiodide, picrotoxin and strychnine) were added to the perfusing solution, for study of their effects on the evoked inhibition and on depressions of activity by iontophoretically applied glycine and γ-aminobutyrate. The results indicate that γ-aminobutyrate is probably the transmitter responsible for the synaptic inhibition.The experiments demonstrate the reproducibility of a central nervous system pathway in tissue culture and the ease of pharmacological manipulation of such a model system. The model has promising applicability for the study of synaptic mechanisms and as a test substrate for the interaction of drugs with a γ-aminobutyrate mediated pathway, particularly where the study of such substances is difficult in vivo.