Glycine-immunoreactive terminals in the rat trigeminal motor nucleus: light- and electron-microscopic analysis of their relationships with motoneurones and with GABA-immunoreactive terminals

Carbachol increases locus coeruleus activation by targeting noradrenergic neurons, inhibitory interneurons and inhibitory synaptic transmission

The locus coeruleus (LC) consists of noradrenergic (NA) neurons and plays an important role in controlling behaviours. Although much of the knowledge regarding LC functions comes from studying behavioural outcomes upon administration of muscarinic acetylcholine receptor (mAChR) agonists into the nucleus, the exact mechanisms remain unclear. Here, we report that the application of carbachol (CCh), an mAChR agonist, increased the spontaneous action potentials (sAPs) of both LC-NA neurons and local inhibitory interneurons (LC I-INs) in acute brain slices by activating M1/M3 mAChRs (m_1/3 AChRs). Optogenetic activation of LC I-INs evoked inhibitory postsynaptic currents (IPSCs) in LC-NA neurons that were mediated by γ-aminobutyric acid type A (GABA_A ) and glycine receptors, and CCh application decreased the IPSC amplitude through a presynaptic mechanism by activating M4 mAChRs (m₄ AChRs). LC-NA neurons also exhibited spontaneous phasic-like activity (sPLA); CCh application increased the incidence of this activity. This effect of CCh application was not observed with blockade of GABA_A and glycine receptors, suggesting that the sPLA enhancement occurred likely because of the decreased synaptic transmission of LC I-INs onto LC-NA neurons by the m₄ AChR activation and/or increased spiking rate of LC I-INs by the m_1/3 AChR activation, which could lead to fatigue of the synaptic transmission. In conclusion, we report that CCh application, while inhibiting their synaptic transmission, increases sAP rates of LC-NA neurons and LC I-INs. Collectively, these effects provide insight into the cellular mechanisms underlying the behaviour modulations following the administration of muscarinic receptor agonists into the LC reported by the previous studies.

GABAB receptor-mediated tonic inhibition of noradrenergic A7 neurons in the rat

Noradrenergic (NAergic) A7 neurons that project axonal terminals to the dorsal horn of the spinal cord to modulate nociceptive signaling are suggested to receive tonic inhibition from local GABAergic interneurons, which are under the regulation of descending analgesic pathways. In support of this argument, we presently report GABA(B) receptor (GABA(B)R)-mediated tonic inhibition of NAergic A7 neurons. Bath application of baclofen induced an outward current (I(Bac)) in NAergic A7 neurons that was blocked by CGP 54626, a GABA(B)R blocker. The I(Bac) was reversed at about -99 mV, displayed inward rectification, and was blocked by Ba(2+) or Tertipian-Q, showing it was mediated by G protein-activated inward-rectifying K(+) (GIRK) channels. Single-cell RT-PCR results suggested that GIRK1/3 heterotetramers might dominate functional GIRK channels in NAergic A7 neurons. Under conditions in which GABA(A) and glycine receptors were blocked, bath application of GABA inhibited the spontaneous firing of NAergic A7 neurons in a dose-dependent manner. Interestingly, CGP 54626 application not only blocked the effect of GABA but also increased the firing rate to 126.9% of the control level, showing that GABA(B)Rs were constitutively active at an ambient GABA concentration of 2.8 μM and inhibited NAergic A7 neurons. GABA(B)Rs were also found at presynaptic excitatory and inhibitory axonal terminals in the A7 area. Pharmacological activation of these GABA(B)Rs inhibited the release of neurotransmitters. No physiological role was found for GABA(B)Rs on excitatory terminals, whereas those on the inhibitory terminals were found to exert autoregulatory control of GABA release.

Co-localisation of markers for glycinergic and GABAergic neurones in rat nucleus of the solitary tract: implications for co-transmission

Immunoreactive structures visualised with antibodies to glycine were prominent in areas of the nucleus of the solitary tract (NTS) surrounding the tractus solitarius, but scarcer in medial and ventral areas of the nucleus. This contrasted with a higher density, more homogenous distribution of structures labelled for gamma-aminobutyric acid (GABA). Immunolabelling of adjacent semi-thin sections nonetheless indicated a close correspondence between cells and puncta labelled by glycine and GABA antisera in certain NTS areas. With post-embedding electron microscopic immunolabelling, synaptic terminals with high, presumed transmitter levels of glycine were discriminated from terminals containing low, metabolic levels by quantitative analysis of gold particle labelling densities. In a random sample of terminals, 28.5% qualified on this basis as glycinergic (compared to 44.4% GABAergic); these glycinergic terminals targeted mainly dendritic structures and contained pleomorphic vesicles and symmetrical synapses. Serial section analysis revealed few terminals (5.2%) immunoreactive for glycine alone, with 82% of glycinergic terminals also containing high levels of GABA immunoreactivity. No evidence for co-localisation of glycine and glutamate was found. Light, confocal and electron microscopic labelling with antibodies to proteins specific for glycine and GABA synthesis, release and uptake confirmed that glycinergic terminals also containing GABA are found predominantly in more lateral areas of NTS, despite glycine receptors and the 'glial' glycine transporter (GLYT1) being expressed throughout all areas of the nucleus. The data suggest that synaptic terminals in certain functionally distinct areas of NTS co-release both inhibitory amino acids, which may account for the previously reported differential inhibitory effects of glycine and GABA on NTS neurones.

Physiological and morphological properties of, and effect of substance P on, neurons in the A7 catecholamine cell group in rats

The A7 catecholamine cell group consists of noradrenergic (NAergic) neurons that project to the dorsal horn of the spinal cord. Here, we characterized their morphology and physiology properties and tested the effect of substance P (Sub-P) on them, since the results of many morphological studies suggest that A7 neurons are densely innervated by Sub-P-releasing terminals from nuclei involved in the descending inhibitory system, such as the lateral hypothalamus and periaqueductal gray area. Whole cell recordings were made from neurons located approximately 200 microm rostral to the trigeminal motor nucleus (the presumed A7 area) in sagittal brainstem slices from rats aged 7-10 days. After recording, the neurons were injected with biocytin and immunostained with antibody against dopamine-beta-hydroxylase (DBH). DBH-immunoreactive (ir) cells were presumed to be NAergic neurons. They had a large somata diameter ( approximately 20 microm) and relatively simple dendritic branching patterns. They fired action potentials (AP) spontaneously with or without blockade of synaptic inputs, and had similar properties to those of NAergic neurons in other areas, including the existence of calcium channel-mediated APs and a voltage-dependent delay in initiation of the AP (an indicator of the existence of A-type potassium currents) and an ability to be hyperpolarized by norepinephrine. Furthermore, in all DBH-ir neurons tested, Sub-P caused depolarization of the membrane potential and an increase in neuronal firing rate by acting on neurokinin-1 receptors. Non-DBH-ir neurons with a smaller somata size were also found in the A7 area. These showed great diversity in firing patterns and about half were depolarized by Sub-P. Morphological examination suggested that the non-DBH-ir neurons form contacts with DBH-ir neurons. These results provide the first description of the intrinsic regulation of membrane properties of, and the excitatory effect of Sub-P on, A7 area neurons, which play an important role in pain regulation.

Neurobiology of orofacial proprioception

Primary sensory fibers innervating the head region derive from neurons of both the trigeminal ganglion (TG) and mesencephalic trigeminal nucleus (MTN). The trigeminal primary proprioceptors have their cell bodies in the MTN. Unlike the TG cells, MTN neuronal somata are centrally located within the brainstem and receive synaptic inputs that potentially modify their output. They are a crucial component of the neural circuitry responsible for the generation and control of oromotor activities. Gaining an insight into the chemical neuroanatomy of the MTN is, therefore, of fundamental importance for the understanding of neurobiology of the head proprioceptive system. This paper summarizes the recent advances in our knowledge of pre- and postsynaptic mechanisms related to orofacial proprioceptive signaling in mammals. It first briefly describes the neuroanatomy of the MTN, which is involved in the processing of proprioceptive information from the face and oral cavity, and then focuses on its neurochemistry. In order to solve the puzzle of the chemical coding of the mammalian MTN, we review the expression of classical neurotransmitters and their receptors in mesencephalic trigeminal neurons. Furthermore, we discuss the relationship of neuropeptides and their corresponding receptors in relaying of masticatory proprioception and also refer to the interactions with other atypical neuromessengers and neurotrophic factors. In extension of previous inferences, we provide conclusive evidence that the levels of transmitters vary according to the environmental conditions thus implying the neuroplasticity of mesencephalic trigeminal neurons. Finally, we have also tried to give an integrated functional account of the MTN neurochemical profiles.

Postnatal development of noradrenergic terminals in the rat trigeminal motor nucleus: A light and electron microscopic immunocytochemical analysis

The noradrenergic (NA) innervation in the trigeminal motor nucleus (Vmot) of postnatal and adult rats was examined by light and electron microscopic immunocytochemistry using antibodies against dopamine-beta-hydroxylase or tyrosine hydroxylase. NA fibers were identified in the Vmot as early as the day of birth (postnatal day 0; P0). A continuous increase in the density of labeled fibers was observed during development up to P20, with a slight decrease at P30 and in the adult. Electron microscopic analysis of serial ultrathin sections revealed that, at P5, nearly half (46%) of the examined NA terminals made synaptic contact with other neuronal elements with membrane specializations. The percentage of examined NA varicosities engaged in synaptic contacts increased at P15 (74%), then decreased in the adult (64%). At all developmental ages, the majority of contacts made by these boutons were symmetrical, the postsynaptic elements being mainly dendrites and occasionally somata. Interestingly, some of the NA terminals made axo-axon contacts with other unidentified boutons. These results show that, although the density of NA fibers increases during postnatal development, functional NA boutons are present in the Vmot at early postnatal ages. Some of these fibers might exert their effects via nonsynaptic release of noradrenaline, the so-called volume transmission, but, in the main, they form conventional synaptic contacts with dendrites, somata, and other axonal terminals in the Vmot. These results are consistent with previous electrophysiological studies that propose an important role for the NA system in modulating mastication.

Morphological and immunohistochemical characterization of interneurons within the rat trigeminal motor nucleus

Three series of experiments were carried out to characterize interneurons located within the trigeminal motor nucleus of young rats aged 5-24 days. Cholera toxin injections were made bilaterally into the masseter and, sometimes, digastric muscles to label motoneurons. In the first set of experiments, thick slices were taken from the pontine brainstem and cholera toxin-positive and cholera toxin-negative neurons located inside the trigeminal motor nucleus were filled with biocytin through whole-cell recording patch electrodes. Positively identified motoneurons (cholera toxin+) of various shapes and sizes always had a thick, unbranched axon that entered the motor root following a tight zigzag course. Many cholera toxin-negative neurons were also classified as motoneurons after biocytin filling based on this particularity of their axon. These are probably either fusimotor motoneurons or motoneurons supplying other jaw muscles. The cholera toxin-negative neurons classified as interneurons differed markedly from motoneurons in that they had thin, usually branched axons that supplied the ipsilateral reticular region surrounding the trigeminal motor nucleus (peritrigeminal area), the main trigeminal sensory nucleus, the trigeminal mesencephalic nucleus, the medial reticular formation of both sides, and the contralateral medial peritrigeminal area. Most often, their dendrites were arranged in bipolar arbors that extended beyond the borders of the trigeminal motor nucleus into the peritrigeminal area. Immunohistochemistry against glutamate, GABA and glycine was used to further document the nature and distribution of putative interneurons. Immunoreactive neurons were uniformly distributed throughout the rostro-caudal extent of the trigeminal motor nucleus. Their concentration seemed greater toward the edges of the nucleus and they were scarce in the digastric motoneuron pool. Glutamate- outnumbered GABA- and glycine-immunoreactive neurons. There was no clear segregation between the three populations. In the final experiment, 1,1'-dioctadecyl-3,3,3',3'-tetra-methylindocarbocyanine perchlorate crystals were inserted into one trigeminal motor nucleus in thick slices and allowed to diffuse for several weeks. This procedure marked commissural fibers and interneurons in the contralateral trigeminal motor nucleus. Together these results conclusively support the existence of interneurons in the trigeminal motor nucleus.

Neuron numbers in the sensory trigeminal nuclei of the rat: A GABA- and glycine-immunocytochemical and stereological analysis

The volume, total neuron number, and number of GABA- and glycine-expressing neurons in the sensory trigeminal nuclei of the adult rat were estimated by stereological methods. The mean volume is 1.38+/-0.13 mm3 (mean+/-SD) for the principal nucleus (Vp), 1.59+/-0.06 for the n. oralis (Vo), 2.63+/-0.34 for the n. interpolaris (Vip), and 3.73+/-0.11 for the n. caudalis (Vc). The total neuron numbers are 31,900+/-2,200 (Vp), 21,100+/-3,300 (Vo), 61,600+/-8,300 (Vip), and 159,100+/-25,300 (Vc). Immunoreactive (-ir) neurons were classified as strongly stained or weakly stained, depending on qualitative criteria, cross-checked by a densitometric analysis. GABA-ir cells are most abundant in Vc, in an increasing rostrocaudal gradient within the nucleus. Lower densities are found in Vip and Vp. The mean total number of strongly labeled GABA-ir neurons ranges between 1,800 in Vp to 7,800 in Vip and 22,900 in Vc, and varies notably between subjects. Glycine-ir neurons are more numerous and display more homogeneous densities in all nuclei. Strongly labeled Gly-ir cells predominate in all nuclei, their total number ranging between 9,400 in Vp to 24,300 in Vip and 34,200 in Vc. A substantial fraction of immunolabeled neurons in all nuclei coexpress GABA and glycine. In general, all neurons strongly immunoreactive for GABA are small, while weakly GABA-ir cells which coexpress Gly are larger. In Vc, one-third of all neurons are immunoreactive: 16.6% of them are single-labeled for GABA and 31.6% are single-labeled for glycine. The remaining 51.8% express GABA and glycine in different combinations, with those showing strong double labeling accounting for 22.6%.

Co-localization of glycine and gaba immunoreactivity in interneurons in Macaca monkey cerebellar cortex

Previous work demonstrates that the cerebellum uses glycine as a fast inhibitory neurotransmitter [Ottersen OP, Davanger S, Storm-Mathisen J (1987) Glycine-like immunoreactivity in the cerebellum of rat and Senegalese baboon, Papio papio: a comparison with the distribution of GABA-like immunoreactivity and with [3H]glycine and [3H]GABA uptake. Exp Brain Res 66(1):211-221; Ottersen OP, Storm-Mathisen J, Somogyi P (1988) Colocalization of glycine-like and GABA-like immunoreactivities in Golgi cell terminals in the rat cerebellum: a postembedding light and electron microscopic study. Brain Res 450(1-2):342-353; Dieudonne S (1995) Glycinergic synaptic currents in Golgi cells of the rat cerebellum. Proc Natl Acad Sci U S A 92:1441-1445; Dumoulin A, Triller A, Dieudonne S (2001) IPSC kinetics at identified GABAergic and mixed GABAergic and glycinergic synapses onto cerebellar Golgi cells. J Neurosci 21(16):6045-6057; Dugue GP, Dumoulin A, Triller A, Dieudonne S (2005) Target-dependent use of coreleased inhibitory transmitters at central synapses. J Neurosci 25(28):6490-6498; Zeilhofer HU, Studler B, Arabadzisz D, Schweizer C, Ahmadi S, Layh B, Bosl MR, Fritschy JM (2005) Glycinergic neurons expressing enhanced green fluorescent protein in bacterial artificial chromosome transgenic mice. J Comp Neurol 482(2):123-141]. In the rat cerebellum glycine is not released by itself but is released together with GABA by Lugaro cells onto Golgi cells [Dumoulin A, Triller A, Dieudonne S (2001) IPSC kinetics at identified GABAergic and mixed GABAergic and glycinergic synapses onto cerebellar Golgi cells. J Neurosci 21(16):6045-6057] and by Golgi cells onto unipolar brush and granule cells [Dugue GP, Dumoulin A, Triller A, Dieudonne S (2005) Target-dependent use of coreleased inhibitory transmitters at central synapses. J Neurosci 25(28):6490-6498]. Here we report, from immunolabeling evidence in Macaca cerebellum, that interneurons in the granular cell layer are glycine+ at a density of 120 cells/linear mm. Their morphology indicates that they include Golgi and Lugaro cell types with the majority containing both glycine and GABA or glutamic acid decarboxylase. These data are consistent with the proposal that, as in the rat cerebellum, these granular cell layer interneurons corelease glycine and GABA in the primate cerebellum. The patterns of labeling for glycine and GABA within Golgi and Lugaro cells also indicate that there are biochemical sub-types which are morphologically similar. Further, we find that glycine, GABA and glutamic acid decarboxylase identified candelabrum cells adjacent to the Purkinje cells which is the first time that this interneuron has been reported in primate cerebellar cortex. We propose that candelabrum cells, like the majority of Golgi and Lugaro cells, release both glycine and GABA.

Physiological characterization, localization and synaptic inputs of bursting and nonbursting neurons in the trigeminal principal sensory nucleus of the rat

A population of neurons in the trigeminal principal sensory nucleus (NVsnpr) fire rhythmically during fictive mastication induced in the in vivo rabbit. To elucidate whether these neurons form part of the central pattern generator (CPG) for mastication, we performed intracellular recordings in brainstem slices taken from young rats. Two cell types were defined, nonbursting (63%) and bursting (37%). In response to membrane depolarization, bursting cells, which dominated in the dorsal part of the NVsnpr, fired an initial burst followed by single spikes or recurring bursts. Non-bursting neurons, scattered throughout the nucleus, fired single action potentials. Microstimulation applied to the trigeminal motor nucleus (NVmt), the reticular border zone surrounding the NVmt, the parvocellular reticular formation or the nucleus reticularis pontis caudalis (NPontc) elicited a postsynaptic potential in 81% of the neurons tested for synaptic inputs. Responses obtained were predominately excitatory and sensitive to glutamatergic antagonists DNQX and/or APV. Some inhibitory and biphasic responses were also evoked. Bicuculline methiodide or strychnine blocked the IPSPs indicating that they were mediated by GABA(A) or glycinergic receptors. About one-third of the stimulations activated both types of neurons antidromically, mostly from the masseteric motoneuron pool of NVmt and dorsal part of NPontc. In conclusion, our new findings show that some neurons in the dorsal NVsnpr display both firing properties and axonal connections which support the hypothesis that they may participate in masticatory pattern generation. Thus, the present data provide an extended basis for further studies on the organization of the masticatory CPG network.

The distribution of inhibitory and excitatory synapses on single, reconstructed jaw-opening motoneurons in the cat

In a previous study, we reported that the distribution of inhibitory input, in contrast to excitatory input, decreased somatofugally along dendrites of cat jaw-closing alpha-motoneurons [J Comp Neurol 414 (1999) 454]. The present study examined the distribution of GABA, glycine, and glutamate immunopositive boutons covering horseradish peroxidase-labeled cat jaw-opening motoneurons. The motoneurons were divided into four compartments: the soma, and primary, intermediate, and distal dendrites. Ninety-seven percent of the total number of studied boutons had immunoreactivity for at least one of the three amino acids. The proportion of boutons immunoreactive for GABA and/or glycine was lower than the proportion of boutons immunoreactive for glutamate. Boutons immunoreactive to glycine alone were more numerous than boutons double-labeled for GABA and glycine, which, in turn, occurred more frequently than boutons immunoreactive to GABA alone. The percentage synaptic covering (proportion of membrane covered by synaptic boutons) of the putatively excitatory (glutamate containing) and putatively inhibitory (GABA and/or glycine containing) boutons decreased somatofugally along the dendrites. Such systematic variations were not seen in the packing density (number of boutons per 100 microm(2)); the packing density showed a distinct drop between the soma and primary dendrites but did not differ significantly among the three dendritic compartments. Overall, the packing density was slightly higher for the putatively excitatory boutons than for the inhibitory ones. When taken together with previous analyses of jaw-closing alpha-motoneurons the present data on jaw-opening alpha-motoneurons indicate that the two types of neuron differ in regard to the nature of synaptic integration in the dendritic tree.

Neurotransmitter phenotypes of intermediate zone reticular formation projections to the motor trigeminal and hypoglossal nuclei in the rat

Numerous studies suggest an essential role for the intermediate (IRt) and parvocellular (PCRt) reticular formation (RF) in consummatory ingestive responses. Although the IRt and PCRt contain a large proportion of neurons with projections to the oromotor nuclei, these areas of the RF are heterogeneous with respect to neurotransmitter phenotypes. Glutamatergic, GABAergic, cholinergic, and nitrergic neurons are all found in the PCRt and IRt, but the projections of neurons with these phenotypes to the motor trigeminal (mV) and hypoglossal nucleus (mXII) has not been fully evaluated. In the present study, after small injections of Fluorogold (FG) into mV and mXII, sections were processed immunohistochemically to detect retrogradely labeled FG neurons in combination with the synthetic enzymes for nitric oxide (nitric oxide synthase) or acetylcholine (choline acetyltransferase) or in situ hybridization for the synthetic enzyme for GABA (GAD65/67) or the brainstem vesicular transporter for glutamate (VGLUT2). In three additional cases, FG injections were made into one motor nucleus and cholera toxin (subunit b) injected in the other to determine the presence of dual projection neurons. Premotor neurons to mXII (pre-mXII) were highly concentrated in the IRt. In contrast, there were nearly equal proportions of premotor-trigeminal neurons (pre-mV) in the IRt and PCRt. A high proportion of pre-oromotor neurons were positive for VGLUT2 (pre-mXII: 68%; pre-mV: 53%) but GABAergic projections were differentially distributed with a greater projection to mV (25%) compared to mXII (8%). Significant populations of cholinergic and nitrergic neurons overlapped pre-oromotor neurons, but there was sparse double-labeling (<10%). The IRt also contained a high proportion of neurons that projected to both mV and MXII. These different classes of premotor neurons in the IRt and PCRt provide a substrate for the rhythmic activation of lingual and masticatory muscles.

Inhibition of afferent transmission in the feeding circuitry of aplysia: persistence can be as important as size

We are studying afferent transmission from a mechanoafferent, B21, to a follower, B8. During motor programs, afferent transmission is regulated so that it does not always occur. Afferent transmission is eliminated when spike propagation in B21 fails, i.e., when spike initiation is inhibited in one output region-B21's lateral process. Spike initiation in the lateral process is inhibited by the B52 and B4/5 cells. Individual B52 and B4/5-induced inhibitory postsynaptic potentials (IPSPs) in B21 differ. For example, the peak amplitude of a B4/5-induced IPSP is four times the amplitude of a B52 IPSP. Nevertheless, when interneurons fire in bursts at physiological (i.e., low) frequencies, afferent transmission is most effectively reduced by B52. Although individual B52-induced IPSPs are small, they have a long time constant and summate at low firing frequencies. Once IPSPs summate, they effectively block afferent transmission. In contrast, individual B4/5-induced IPSPs have a relatively short time constant and do not summate at low frequencies. B52 and B4/5 therefore differ in that once synaptic input from B52 becomes effective, afferent transmission is continuously inhibited. In contrast, periods of B4/5-induced inhibition are interspersed with relatively long intervals in which inhibition does not occur. Consequently, the probability that afferent transmission will be inhibited is low. In conclusion, it is widely recognized that afferent transmission can be regulated by synaptic input. Our experiments are, however, unusual in that they relate specific characteristics of postsynaptic potentials to functional inhibition. In particular we demonstrate the potential importance of the IPSP time constant.

Morphological characteristics of Lugaro cells in the cerebellar cortex

Two types of Lugaro cells were identified in the cat cerebellar cortex using sections impregnated with silver nitrate by the Golgi-Kopsch method; these cells were fusiform and triangular and their bodies were located at different levels in the granular layer. Their processes were directed horizontally, vertically, or obliquely to the axis of the leaf; axons never left the cerebellar cortex. These cells should therefore be regarded as interneurons. The processes of Lugaro cells were very extended, with the result that these cells formed numerous axosomatic and axodendritic contacts with all cerebellar cortical neurons and fibers. The structural and topographical characteristics of Lugaro cells and the features of their contacts with other cells in the cerebellar cortex, taken together with data on their neurotransmitter contents, show that they function as inhibitory interneurons.

Distinct profiles of refilling of inhibitory neurotransmitters into presynaptic terminals projecting to spinal neurones in immature rats

Corelease of glycine and GABA from the single synaptic terminal (synaptic bouton) is well accepted in immature rat spinal cord and brainstem. However, it raises the question of how glycine and GABA are accumulated in the same synaptic vesicles and coreleased. To address this issue, spontaneous miniature inhibitory postsynaptic currents (mIPSCs) and focally evoked IPSCs (eIPSCs) mediated via a single synapse were recorded from synaptic bouton preparations of the rat immature sacral dorsal commissural nucleus (SDCN) neurones by whole-cell patch recording. Focal stimulation of a single synaptic bouton revealed that three different quantal releases occur from a single synaptic bouton: i.e. pure glycine, pure GABA, and mixed. Prolonged treatment with bafilomycin A1, a vacuolar-type H+/ATPase inhibitor, to the SDCN neurone greatly suppressed frequency and amplitude of the mIPSCs. During washing out of bafilomycin A1, complete recovery in the amplitude of glycinergic mIPSCs was observed, while that of GABAergic and mixed mIPSCs was incomplete. These observations indicate that three types of vesicles coexist in single synaptic terminals, and that refilling of glycine into the synaptic vesicle predominantes over GABA after pretreatment with bafilomycin A1 in immature rats. This could be explained by the decrease in the cytosolic concentration of GABA, or by the presence of subtypes of vesicular inhibitory amino acid transporter in the synaptic vesicle membrane.

The physiological and morphological characteristics of interneurons caudal to the trigeminal motor nucleus in rats

In this study we have characterized the membrane properties and morphology of interneurons which lie between the caudal pole of the trigeminal motor nucleus and the rostral border of the facial motor nucleus. Previous studies suggest that many of these interneurons may participate in the genesis of rhythmical jaw movements. Saggital brainstem slices were taken from rats aged 5-8 days. Interneurons lying caudal to the trigeminal motor nucleus were visualized using near-infrared differential interference contrast (DIC) microscopy, and were recorded from using patch pipettes filled with a K-gluconate- and biocytin-based solution. The 127 neurons recorded could be categorized into three subtypes on the basis of their responses to injection of depolarizing current pulses, namely tonic firing (type I), burst firing (type II) and spike-adaptive (type III) neurons. Type I interneurons had a higher input resistance and a lower rheobase than type II neurons. All three neuron subtypes showed 'sag' of the voltage response to injection of large-amplitude hyperpolarizing current pulses, and, in addition, also showed rectification of the voltage response to injection of depolarizing current pulses, with type II neurons showing significantly greater rectification than type I neurons. The axonal arborizations were reconstructed for 44 of 63 neurons labelled with tracer. Neurons of each subtype were found to issue axon collaterals terminating in the brainstem nuclei, including the parvocellular reticular nucleus (PCRt), the trigeminal motor nucleus (Vmot), the supratrigeminal nucleus or the trigeminal mesencephalic nucleus. Twenty-five of the 43 neurons issued collaterals which terminated in the Vmot and the other brainstem nuclei. When viewed under 100x magnification, the collaterals of some interneurons were seen to give off varicosities and end-terminations which passed close to the somata of unidentified neurons in the trigeminal motor nucleus and in the area close to the interneuron soma itself. This suggests that the interneurons may make synaptic contacts both on motoneurons and also on nearby interneurons. These results provide data on the membrane properties of trigeminal interneurons and evidence for their synaptic connections both with nearby interneurons and also with motoneurons. Thus, the interneurons examined could play roles in the shaping, and possibly also in the generation, of rhythmical signals to trigeminal motoneurons.

Developmental switch from GABA to glycine release in single central synaptic terminals

Early in postnatal development, inhibitory inputs to rat lateral superior olive (LSO) neurons change from releasing predominantly GABA to releasing predominantly glycine into the synapse. Here we show that spontaneous miniature inhibitory postsynaptic currents (mIPSCs) also change from GABAergic to glycinergic over the first two postnatal weeks. Many 'mixed' mIPSCs, resulting from co-release of glycine and GABA from the same vesicles, are seen during this transition. Immunohistochemistry showed that a large number of terminals contained both GABA and glycine at postnatal day 8 (P8). By P14, both the content of GABA in these mixed terminals and the contribution of GABA to the mixed mIPSCs had decreased. The content of glycine in terminals increased over the same period. Our results indicate that switching from GABAergic to glycinergic inputs to the LSO may occur at the level of a single presynaptic terminal. This demonstrates a new form of developmental plasticity at the level of a single central synapse.

Effect of lidocaine and NMDA injections into the medial pontobulbar reticular formation on mastication evoked by cortical stimulation in anaesthetized rabbits

Neurons of the dorsal nucleus reticularis pontis caudalis (nPontc) fire rhythmically during fictive mastication, while neurons of the ventral half tend to fire tonically (Westberg et al., 2001). This paper describes the changes in the pattern of rhythmical mastication elicited by stimulation of the sensorimotor cortex during inhibition or excitation of neurons in this nucleus and adjacent parts of nucleus reticularis gigantocellularis (Rgc) in the anaesthetized rabbit. Masticatory movements and electromyographic (EMG) activity of the masseter and digastric muscles produced by cortical stimulation were recorded before, during and after injections of a local anaesthetic (lidocaine) or excitatory amino acid N-methyl-d-aspartate (NMDA) into nPontc and Rgc through a microsyringe with attached microelectrode to record neuronal activity. Lidocaine inhibited local neurons and modified the motor program, and the effects varied with the site of injection. Most injections into the ventral half of nPontc increased cycle duration, digastric burst duration and burst area. The action of lidocaine in dorsal nPontc was more variable, although burst duration and area were often decreased. The effects on the muscle activity were always bilateral. Lidocaine block of the rostromedial part of Rgc had no effect on movements or on EMGs. Injections of NMDA excited local neurons and when injected into ventral nPontc, it completely blocked mastication. Dorsal injections either had no effect or increased cycle frequency, while decreasing burst duration and area. No increases in EMG burst duration or area were observed with NMDA. Our findings suggest that neurons of ventral nPontc tonically inhibit other parts of the central pattern generator during mastication, while dorsal neurons have mixed effects. We incorporated these findings into a new model of the masticatory central pattern generator.

Ultrastructural features of synapse from dorsal parvocellular reticular formation neurons to hypoglossal motoneurons of the rat

The dorsal parvocellular reticular formation (PCRt) receives projection of the trigeminal mesencephalic nucleus neurons. It contains the dorsal group of interneurons that integrate and coordinate activity of the oral motor nuclei. Ultrastructural features of synaptic connection from the dorsal PCRt neurons to the motoneurons of the hypoglossal nucleus (XII) were examined at both the light and electron microscopic levels in rats. Biotinylated dextran amine (BDA) was initially iontophoresed into the dorsal part of PCRt unilaterally. Seven days later horseradish peroxidase (HRP) was injected into the body of the tongue. After histochemical reaction for visualization of HRP and BDA, the BDA-labeled fibers and terminals were seen distributing bilaterally in XII with ipsilateral predominance. BDA-labeled terminals were closely apposed upon HRP retrogradely labeled somata and dendrites of the XII motoneurons. A total of 1408 BDA-labeled boutons were examined ultrastructurally, which had mean size of 1.22+/-0.37 microm in diameter. Five hundred-ninety three of these boutons in both the ipsilateral (n=401) and contralateral (n=192) XII were seen to synapse on both the dendrites and somata of HRP-labeled motoneurons. The vast majorities of synapses were axodendritic (98%, 580/593), while 2% of them were axosomatic. Of the 1408 BDA-labeled boutons, 69.6% of them were S-type boutons containing small clear and spherical synaptic vesicles and 30.4% of them were PF-type boutons containing pleomorphic and flattened synaptic vesicles. Approximately 64% of synapses between BDA-labeled boutons and HRP-labeled motoneurons were asymmetric, and 33% of synapses were symmetric. No axoaxodendritic or axoaxosomatic synaptic triad was observed. The present study illustrated the anatomical pathway and synaptological characteristics of neuronal connection between the dorsal PCRt premotor neurons and the XII motoneurons. Its functional significance in coordinating activity of XII motoneurons during oral motor behaviors has been discussed.

Role of GABA(B) receptor in the regulation of excitatory synaptic transmission in trigeminal motoneurons

The aim of the present study was to determine if excitatory synaptic transmission onto trigeminal motoneurons is subject to a presynaptic modulation by gamma-aminobutyric acid (GABA) via GABA(B) receptor in this system. Whole cell recordings were made from trigeminal motoneurons in longitudinal brain stem slices taken from 8-day-old rats. Monosynaptic excitatory postsynaptic potential (EPSP) activity was evoked by placing bipolar stainless steel electrodes dorsal-caudal to the trigeminal motor nucleus. Bath application of the GABA(B) receptor agonist, baclofen, produced a marked reduction in the mean amplitude and variance of evoked EPSPs and also increased the portion of transmission failures. It also produced a decrease in the frequency, but not in the mean amplitude, of spontaneous miniature EPSPs. Bath application of GABA(B) receptor antagonists 6-hydroxy-saclofen and CGP35348 increased both the amplitude and frequency of miniature EPSP activity. Taken together the above results suggest that the excitatory synaptic inputs onto trigeminal motoneurons are controlled by tonic presynaptic modulation by GABA(B) receptor.

IPSC kinetics at identified GABAergic and mixed GABAergic and glycinergic synapses onto cerebellar Golgi cells

In the rat cerebellum, Golgi cells receive serotonin-evoked inputs from Lugaro cells (L-IPSCs), in addition to spontaneous inhibitory inputs (S-IPSCs). In the present study, we analyze the pharmacology of these IPSCs and show that S-IPSCs are purely GABAergic events occurring at basket and stellate cell synapses, whereas L-IPSCs are mediated by GABA and glycine. Corelease of the two transmitters at Lugaro cell synapses is suggested by the fact that both GABA(A) and glycine receptors open during individual L-IPSCs. Double immunocytochemical stainings demonstrate that GABAergic and glycinergic markers are coexpressed in Lugaro cell axonal varicosities, together with the mixed vesicular inhibitory amino acid transporter. Lugaro cell varicosities are found apposed to glycine receptor (GlyR) clusters that are localized on Golgi cell dendrites and participate in postsynaptic complexes containing GABA(A) receptors (GABA(A)Rs) and the anchoring protein gephyrin. GABA(A)R and GlyR/gephyrin appear to form segregated clusters within individual postsynaptic loci. Basket and stellate cell varicosities do not face GlyR clusters. For the first time the characteristics of GABA and glycine cotransmission are compared with those of GABAergic transmission at identified inhibitory synapses converging onto the same postsynaptic neuron. The ratio of the decay times of L-IPSCs and of S-IPSCs is a constant value among Golgi cells. This indicates that, despite a high cell-to-cell variability of the overall IPSC decay kinetics, postsynaptic Golgi cells coregulate the kinetics of their two main inhibitory inputs. The glycinergic component of L-IPSCs is responsible for their slower decay, suggesting that glycinergic transmission plays a role in tuning the IPSC kinetics in neuronal networks.

Involvement of GABA(A) receptor in modulation of jaw muscle activity evoked by mustard oil application to the rat temporomandibular joint

The effect of intrathecal administration of the GABA(A) receptor antagonist bicuculline methylbromide on jaw muscle electromyographic (EMG) activity evoked by mustard oil injection into the rat temporomandibular joint was studied. Bicuculline given prior to mustard oil augmented the EMG activity evoked by mustard oil, and "rekindling" of EMG activity was induced by bicuculline given 30 min after mustard oil. These results suggest that central GABA(A) receptors modulate reflex responses to noxious craniofacial stimuli.

Development of inhibitory synaptic transmission to motoneurons

The inhibitory effects of the neurotransmitters glycine and gamma-aminobutyric acid (GABA) on motoneurons and their role in mediating the timing of motor output have been understood for some years. Recent work, however, has revealed that these neurotransmitters function very differently in developing motor circuits. Most strikingly, both GABA and glycine depolarize neonatal motoneurons, and, in many instances, provide excitatory drive to developing motor networks. Additionally, the relative contributions of GABA and glycine to inhibitory synaptic transmission in a circuit or, indeed, within the same synapse, change with postnatal development. Here, we review three fundamental properties of inhibitory neurotransmission that are altered postnatally and may be important in shaping the unique behaviors of these synapses early in development.

Formation of mixed glycine and GABAergic synapses in cultured spinal cord neurons

In the spinal cord, GABA and glycine mediate inhibition at separate or mixed synapses containing glycine and/or GABA(A) receptors (GlyR and GABA(A)R, respectively). We have analysed here the sequence of events leading to inhibitory synapse formation during synaptogenesis of embryonic spinal cord neurons between 1 and 11 days in vitro (DIV). We used immunocytochemical methods to detect simultaneously an antigen specific to inhibitory terminals, the vesicular inhibitory amino acid transporter (VIAAT), and one of the following postsynaptic elements: GlyR, GABA(A)R or gephyrin, the anchoring protein of GlyR, which is also associated with GABA(A)R. Quantitative analysis revealed that until 5 DIV most gephyrin clusters were not adjacent to VIAAT-positive profiles, but became associated with them at later stages. In contrast, GlyR and GABAAR clustered predominantly in front of VIAAT-containing terminals at all stages. However, about 10% of receptor aggregates were detected at nonsynaptic loci. The two receptors colocalized in 66.2+/-2.5% of the inhibitory postsynaptic domains after 11 DIV, while 30.3+/-2.6% and 3.4+/-0.8% of them contained only GlyR and GABA(A)R, respectively. Interestingly, at 3 DIV GABA(A)R clustered at a postsynaptic location prior to gephyrin and GlyR; GABA(A)R could thus be the initiating element in the construction of mixed glycine and GABAergic synapses. The late colocalization of gephyrin with GABA(A)R, and the demonstration by other groups that, in the absence of gephyrin, postsynaptic GABA(A)R is not detected, suggest that gephyrin is involved in the stabilization of GABA(A)R rather than in its initial accumulation at synaptic sites.

Identification of brainstem interneurons projecting to the trigeminal motor nucleus and adjacent structures in the rabbit

Neurons of several nuclei within the medial pontomedullar reticular formation are active during mastication, but their relationship with other elements of the pattern generating circuits have never been clearly defined. In this paper, we have studied the connection of this area with the trigeminal motor nucleus and with pools of last-order interneurons of the lateral brainstem. Retrograde tracing techniques were used in combination with immunohistochemistry to define populations of glutamatergic and GABAergic neurons. Injections of tracer into the Vth motor nucleus marked neurons in several trigeminal nuclei including the ipsilateral mesencephalic nucleus, the contralateral Vth motor nucleus, the dorsal cap of the main sensory nucleus and the rostral divisions of the spinal nucleus bilaterally. Many last-order interneurons formed a bilateral lateral band running caudally from Regio h (the zone surrounding the Vth motor nucleus), through the parvocellular reticular formation and Vth spinal caudal nucleus. Injections of tracer into Regio h, an area rich in last-order interneurons, marked, in addition to the areas listed above, a large number of neurons in the medial reticular formation bilaterally. The major difference between injection sites was that most neurons projecting to the Vth motor nucleus were located laterally, whereas most of those projecting to Regio h were found medially. Both populations contained glutamatergic and GABAergic neurons intermingled. Our results indicate that neurons of the medial reticular formation that are active during mastication influence Vth motoneurons output via relays in Regio h and other adjacent nuclei.

GABA- and glycine-immunoreactive terminals contacting motoneurons in lamprey spinal cord

Double postembedding GABA- and glycine-immunostaining was performed on the lamprey (Lampetra fluviatilis) spinal cord after previous HRP labeling of motoneurons. Immunopositive boutons contacting motoneurons were counted and distinguished as GABA (39%), glycine (30%) and both GABA+glycine-immunopositive (31%). Densely-packed, flattened synaptic vesicles were only observed in glycine-immunopositive boutons while GABA-immunoreactive and GABA+glycine-immunoreactive boutons contained rounded or oval synaptic vesicles. Dense-core vesicles of different diameters were associated with conventional synaptic vesicles in 74% of GABA-only-immunopositive boutons, 50% of double GABA+glycine-immunopositive boutons, but were only observed in 9% of glycine-only-immunopositive boutons. The presence of terminals immunoreactive to either GABA or glycine contacting the motoneurons suggests that there is a morphological substrate for both GABAergic and glycinergic postsynaptic inhibition of motoneurons in the lamprey spinal cord.

Distribution pattern of inhibitory and excitatory synapses in the dendritic tree of single masseter alpha-motoneurons in the cat

Little is known about the differences in the distributions of inhibitory and excitatory synapses in the dendritic tree of single motoneurons in the brainstem and spinal cord. In this study, the distribution of gamma-aminobutyric acid (GABA)-, glycine-, and glutamate-like immunoreactivity in axon terminals on dendrites of cat masseter alpha-motoneurons, stained intracellularly with horseradish peroxidase, was examined by using postembedding immunogold histochemistry in serial ultrathin sections. The dendritic tree was divided into three segments: primary (Pd) and distal (Dd) dendrites and intermediate (Id) dendrites between the two segments. Quantitative analysis of 175, 279, and 105 boutons synapsing on 13 Pd, 54 Id, and 81 Dd, respectively, was performed. Fifty percent of the total number of studied boutons were immunopositive for GABA and/or glycine and 48% for glutamate. Among the former, 27% showed glycine immunoreactivity only and 14% were immunoreactive to both glycine and GABA. The remainder (9%) showed immunoreactivity for GABA only. As few as 3% of the boutons were immunonegative for the three amino acids. Most boutons immunoreactive to inhibitory amino acid(s) contained a mixture of spherical, oval, and flattened synaptic vesicles. Most boutons immunoreactive to excitatory amino acid contained clear, spherical, synaptic vesicles with a few dense-cored vesicles. When comparisons of the inhibitory and excitatory boutons were made between the three dendritic segments, the proportion of the inhibitory to the excitatory boutons was high in the Pd (60% vs. 37%) but somewhat low in the Id (46% vs. 52%) and Dd (44% vs. 53%). The percentage of synaptic covering and packing density of the inhibitory synaptic boutons decreased in the order Pd, Id, and Dd, but this trend was not applicable to the excitatory boutons. The present study provides possible evidence that the spatial distribution patterns of inhibitory and excitatory synapses are different in the dendritic tree of jaw-closing alpha-motoneurons.

c-fos expression in brainstem premotor interneurons during cholinergically induced active sleep in the cat

The present study was undertaken to identify trigeminal premotor interneurons that become activated during carbachol-induced active sleep (c-AS). Their identification is a critical step in determining the neural circuits responsible for the atonia of active sleep. Accordingly, the retrograde tracer cholera toxin subunit B (CTb) was injected into the trigeminal motor nuclei complex to label trigeminal interneurons. To identify retrograde-labeled activated neurons, immunocytochemical techniques, designed to label the Fos protein, were used. Double-labeled (i.e., CTb(+), Fos(+)) neurons were found exclusively in the ventral portion of the medullary reticular formation, medial to the facial motor nucleus and lateral to the inferior olive. This region, which encompasses the ventral portion of the nucleus reticularis gigantocellularis and the nucleus magnocellularis, corresponds to the rostral portion of the classic inhibitory region of. This region contained a mean of 606 +/- 41.5 ipsilateral and 90 +/- 32.0 contralateral, CTb-labeled neurons. These cells were of medium-size with an average soma diameter of 20-35 micrometer. Approximately 55% of the retrogradely labeled cells expressed c-fos during a prolonged episode of c-AS. We propose that these neurons are the interneurons responsible for the nonreciprocal postsynaptic inhibition of trigeminal motoneurons that occurs during active sleep.

Glycine-immunoreactive synaptic terminals in the nucleus tractus solitarii of the cat: ultrastructure and relationship to GABA-immunoreactive terminals

Postembedding immunogold labeling methods applied to ultrathin and semithin sections of cat dorsomedial medulla showed that neuronal perikarya, dendrites, myelinated and nonmyelinated axons, and axon terminals in the nucleus tractus solitarii contain glycine immunoreactivity. Light microscopic observations on semithin sections revealed that these immunoreactive structures were unevenly distributed throughout the entire nucleus. At the electron microscopic level, synaptic terminals with high levels of glycine-immunoreactivity, assumed to represent those releasing glycine as a neurotransmitter, were discriminated from terminals containing low, probably metabolic levels of glycine-immunoreactivity, by a quantitative analysis method. This compared the immunolabeling of randomly sampled terminals with a reference level of labeling derived from sampling the perikarya of dorsal vagal neurones. The vast majority of these "glycinergic" terminals contained pleomorphic vesicles, formed symmetrical synaptic active zones, and targeted dendrites. They appeared to be more numerous in areas of the nucleus tractus solitarii adjoining the tractus solitarius, but rather scarce caudally, medially, ventrally, and in the dorsal motor vagal nucleus. In a random analysis of the entire nucleus tractus solitarii, 26.2% of sampled terminals were found to qualify as glycine-immunoreactive. In contrast, boutons immunoreactive for gamma-aminobutyric acid (GABA) were more evenly distributed throughout the dorsal vagal complex and accounted for 33.7% of the synaptic terminals sampled. A comparison of serial ultrathin sections suggested three subpopulations of synaptic terminals: one containing high levels of both GABA- and glycine-immunoreactivities (21% of all terminals sampled), one containing only GABA-immunoreactivity (12.7%), and relatively few terminals (5.2%) that were immunoreactive for glycine alone. These results were confirmed by dual labeling of sections using gold particles of different sizes. This study reports the first analysis of the ultrastructure of glycinergic nerve terminals in the cat dorsal vagal complex, and the pattern of coexistence of glycine and GABA observed provides an anatomical explanation for our previously reported inhibitory effects of glycine and GABA on neurones with cardiovascular and respiratory functions in the nucleus tractus solitarii.

Ultrastructural anatomy of physiologically identified jaw-muscle spindle afferent terminations onto retrogradely labeled jaw-elevator motoneurons in the rat

Neuronal microcircuits involving jaw-muscle spindle afferents and jaw-elevator motoneurons were studied via retrograde and intracellular labeling in rats. Initially, trigeminal motoneurons were retrogradely labeled from horseradish peroxidase (HRP) injections into the temporalis and masseter muscles. The intracellular response of jaw-muscle spindle afferent neurons was then characterized during palpation, ramp and hold, and sinusoidal stretching of the jaw-closing muscles. Biotinamide was injected into these neurons, and the tissue was processed for the visualization of HRP and biotinamide. The ultrastructure of 243 intracellularly stained jaw-muscle spindle afferent boutons located within the trigeminal motor nucleus (Vmo) was examined. Eighty-five of these boutons synapsed with motoneurons retrogradely labeled with HRP, and 158 boutons synapsed with unlabeled structures within the Vmo. All spindle afferent boutons contained clear, spherical synaptic vesicles. Although the majority of boutons were S type, a few labeled jaw-muscle spindle afferent boutons possessed a long, narrow cleft, with a subsynaptic cistern comparable to previous descriptions of C-type boutons. Sixty-eight percent of spindle afferent boutons synapsed with large or medium-sized, retrogradely labeled motoneuron dendrites, and 32% synapsed with retrogradely labeled somata. In numerous instances, spindle afferent boutons synapsed with trigeminal motoneuron dendritic or somatic spines. Most of the synapses between spindle afferent boutons and trigeminal motoneuron dendrites were asymmetric, and the greatest percentage of axosomatic synapses between spindle afferents and trigeminal motoneurons were symmetric. Approximately 24% of spindle afferent boutons constituted the intermediate element of a axoaxodendritic or axoaxosomatic assemblage, implying that some jaw-muscle spindle afferent synapses with trigeminal motoneurons are presynaptically modulated.

The vesicular GABA transporter, VGAT, localizes to synaptic vesicles in sets of glycinergic as well as GABAergic neurons

A transporter thought to mediate accumulation of GABA into synaptic vesicles has recently been cloned (McIntire et al., 1997). This vesicular GABA transporter (VGAT), the first vesicular amino acid transporter to be molecularly identified, differs in structure from previously cloned vesicular neurotransmitter transporters and defines a novel gene family. Here we use antibodies specific for N- and C-terminal epitopes of VGAT to localize the protein in the rat CNS. VGAT is highly concentrated in the nerve endings of GABAergic neurons in the brain and spinal cord but also in glycinergic nerve endings. In contrast, hippocampal mossy fiber boutons, which although glutamatergic are known to contain GABA, lack VGAT immunoreactivity. Post-embedding immunogold quantification shows that the protein specifically associates with synaptic vesicles. Triple labeling for VGAT, GABA, and glycine in the lateral oliva superior revealed a higher expression of VGAT in nerve endings rich in GABA, with or without glycine, than in others rich in glycine only. Although the great majority of nerve terminals containing GABA or glycine are immunopositive for VGAT, subpopulations of nerve endings rich in GABA or glycine appear to lack the protein. Additional vesicular transporters or alternative modes of release may therefore contribute to the inhibitory neurotransmission mediated by these two amino acids.

Brainstem mechanisms underlying feeding behaviors

The essential elements controlling trigeminal motoneurons during feeding lie between the trigeminal and facial motor nuclei. These include populations of neurons in the medial reticular formation and pre-motoneurons in the lateral brainstem that reorganize to generate various patterns. Orofacial sensory feedback, antidromic firing in spindle afferents and intrinsic properties of motoneurons also contribute to the final masticatory motor output.

Electron microscopic analysis of gamma-aminobutyric acid and glycine colocalization in rat trigeminal subnucleus caudalis

Postembedding immunogold methods were used to examine the distribution of gamma-aminobutyric acid (GABA) and glycine and especially their colocalization in glomerular neuronal profiles adjacent to trigeminal primary afferent profiles in lamina II of rat subnucleus caudalis. We found that 60% of the profiles adjacent to the trigeminal primary afferent terminals exhibited colocalization of GABA and glycine. GABA alone was found to localize in 17% of the adjacent profiles. Glycine alone was found to localize in 18% of the adjacent profiles. Of interest, 10% of the trigeminal primary afferent fibers showed glycine localization. All the profiles with colocalization of GABA and glycine were identified as presynaptic axonal terminals, suggesting a possible cumulative effect by these two inhibitory neurotransmitters in presynaptic inhibition. These findings show that GABA and glycine colocalize in a subpopulation of presynaptic axonal terminals within lamina II of the subnucleus caudalis. The possible origins of these axons are discussed, as well as their potential involvement in presynaptic inhibition of orofacial nociception.