Excitatory amino acids and synaptic transmission in embryonic rat brainstem motoneurons in organotypic culture

N-methyl-D-aspartate triggers neonatal rat hypoglossal motoneurons in vitro to express rhythmic bursting with unusual Mg2+ sensitivity

The brainstem nucleus hypoglossus innervates the tongue which must contract rhythmically during respiration, chewing and swallowing. Such rhythmic discharges are due to network bursting mediated by AMPA receptor-dependent glutamatergic transmission. The contribution by hypoglossal motoneurons themselves to rhythmicity remains, however, unclear as they might simply express cyclic patterns produced by premotoneurons or, in analogy to spinal motoneurons, might participate to bursting due to activation of their N-methyl-D-aspartate (NMDA) receptors. Using patch clamp recording from hypoglossal motoneurons in slice preparations of neonatal rat brainstem, we observed that NMDA directly depolarized motoneurons to generate various discharge patterns. Most motoneurons produced transient bursts which were consistently restored by repolarizing membrane potential to rest. Fewer motoneurons generated either sustained bursting or random firing. Rhythmic bursts were recorded from XII nerve rootlets even when single motoneuron bursting required hyperpolarization. NMDA evoked bursts were blocked by the Ca2+ antagonist Cd2+, the gap junction blocker carbenoxolone, or Mg2+ free solution, and partially inhibited by tetrodotoxin or nifedipine. Under voltage clamp, NMDA-induced bursting persisted at negative or positive potentials and was resistant to high extracellular Mg2+ in accordance with the observation of widespread motoneuron expression of NMDA 2D receptor subunits that confer poor Mg2+ sensitivity. It is proposed that NMDA depolarized motoneurons with the contribution of Mg2+ insensitive channels, and triggered bursting via cyclic activation/deactivation of voltage-dependent Na+, Ca2+ and K+ currents spread through gap junctions. The NMDA-evoked bursting pattern was similar to the rhythmic discharges previously recorded from the XII nerve during milk sucking by neonatal rats.

Oral-motor patterns of rhythmic trigeminal activity generated in fetal rat brainstem in vitro

Development of neural circuits generating fetal oral-motor activity was characterized in an in vitro isolated brainstem block preparation. Rhythmical trigeminal activity (RTA) at E20-E21 resembled either the pattern or rhythm of neonatal RTA. Conversely, at E18-E19, RTA displayed a different pattern of discharge from neonatal RTA, and output was not regular but intermittent with another slow rhythm.

Negative allosteric modulation of AMPA-preferring receptors by the selective isomer GYKI 53784 (LY303070), a specific non-competitive AMPA antagonist

GYKI 53784 or LY303070 [(-)1-(4-aminophenyl)-4-methyl-7,8-methylenedioxy-4,5-dihydro-3-methylcarbamoyl-2,3-benzodiazepine] belongs to a new family of 2,3-benzodiazepine compounds (also called homophtalazines) selective and non-competitive antagonists at alpha-amino-3-hydroxy-5-methyl-4-isoxazole-propionate (AMPA) receptors. These compounds include the original GYKI-52466, its more potent derivative GYKI 53655 and the active isomer of the latter, GYKI 53784. This review summarizes current knowledge of this novel AMPA antagonist: GYKI 53784. GYKI 53784 is the most potent of the compounds in the 2,3-benzodiazepine class, blocking AMPA receptor-mediated responses. In contrast to the compounds of the quinoxalinedione family, that block AMPA as well as kainate receptors, GYKI 53784 does not block the activation of kainate receptors. Furthermore, GYKI 53784 does not act at the same receptor site as positive AMPA modulators (i.e., cyclothiazide, BDP-12, 1-BCP or aniracetam). GYKI 53784 is a powerful neuroprotective agent in both in vitro and in vivo models of AMPA receptor-mediated excitotoxicity. In contrast to NMDA receptor antagonists, whose favorable clinical actions are compromised by important side effects such as the impairment of memory functions, the selective AMPA antagonist, GYKI 53784, may be of potential clinical value, both in acute (stroke and trauma) and chronic (Alzheimer's disease, epilepsy) neurological disorders.

AMPA-evoked acetylcholine release from cultured spinal cord motoneurons and its inhibition by GABA and glycine

The release of [(3)H]acetylcholine evoked by alpha-amino-3-hydroxy-5-methyl-4-isoxazole propionate (AMPA) and its inhibition mediated by GABA(A) and glycine receptors were studied in superfused cultured rat embryo spinal cord motoneurons prelabeled with [(3)H]choline. AMPA elicited tritium release, possibly representing [(3)H]acetylcholine release in a concentration-dependent manner. The release was external Ca(2+)-dependent and was sensitive to Cd(2+) ions, omega-conotoxin GVIA and omega-conotoxin MVIIC, but not to nifedipine, suggesting the involvement of N-, P/Q-, but not L-type Ca(2+) channels. The AMPA effect was insensitive to tetrodotoxin. The glutamate receptors involved are AMPA type since the AMPA-evoked [(3)H]acetylcholine release was blocked by LY303070 and was potentiated by the antidesensitizing agent cyclothiazide. Muscimol inhibited completely the AMPA effect on [(3)H]acetylcholine release; muscimol was potentiated by diazepam and antagonized by SR95531, indicating the involvement of benzodiazepine-sensitive GABA(A) receptors. Glycine, acting at strychnine-sensitive receptors, also inhibited the effect of AMPA, but only in part. The inhibitory effects of muscimol and glycine are additive. We conclude that glutamate can act at AMPA receptors sited on spinal motoneurons to evoke release of acetylcholine. GABA and glycine, possibly released as cotransmitters from spinal interneurons, inhibit glutamate-evoked acetylcholine release by activating GABA(A) and glycine receptors on motoneurons.

Properties and interconnections of trigeminal interneurons of the lateral pontine reticular formation in the rat

Numerous evidence suggests that interneurons located in the lateral tegmentum at the level of the trigeminal motor nucleus contribute importantly to the circuitry involved in mastication. However, the question of whether these neurons participate actively to genesis of the rhythmic motor pattern or simply relay it to trigeminal motoneurons remains open. To answer this question, intracellular recordings were performed in an in vitro slice preparation comprising interneurons of the peritrigeminal area (PeriV) surrounding the trigeminal motor nucleus (NVmt) and the parvocellular reticular formation ventral and caudal to it (PCRt). Intracellular and extracellular injections of anterograde tracers were also used to examine the local connections established by these neurons. In 97% of recordings, electrical stimulation of adjacent areas evoked a postsynaptic potential (PSP). These PSPs were primarily excitatory, but inhibitory and biphasic responses were also induced. Most occurred at latencies longer than those required for monosynaptic transmission and were considered to involve oligosynaptic pathways. Both the anatomical and physiological findings show that all divisions of PeriV and PCRt are extensively interconnected. Most responses followed high-frequency stimulation (50 Hz) and showed little variability in latency indicating that the network reliably distributes inputs across all areas. In all neurons but one, excitatory postsynaptic potentials (EPSPs) or inhibitory postsynaptic potentials (IPSPs) were also elicited by stimulation of NVmt, suggesting the existence of excitatory and inhibitory interneurons within the motor nucleus. In a number of cases, these PSPs were reproduced by local injection of glutamate in lieu of the electrical stimulation. All EPSPs induced by stimulation of PeriV, PCRt, or NVmt were sensitive to ionotropic glutamate receptor antagonists 6-cyano-7-dinitroquinoxaline and D,L-2-amino-5-phosphonovaleric acid, while IPSPs were blocked by bicuculline and strychnine, antagonists of GABA(A) and glycine receptors. Examination of PeriV and PCRt intrinsic properties indicate that they form a fairly uniform network. Three types of neurons were identified on the basis of their firing adaptation properties. These types were not associated with particular regions. Only 5% of all neurons showed bursting behavior. Our results do not support the hypothesis that neurons of PeriV and PCRt participate actively to rhythm generation, but suggest instead that they are driven by rhythmical synaptic inputs. The organization of the network allows for rapid distribution of this rhythmic input across premotoneuron groups.