Ultrastructural observations in the frog spinal cord in relation to the generation of primary afferent depolarization

The spatial distribution of different types of synaptic boutons over the motoneuron cell body and dendrites was studied both in the normal and in the chronically deafferented frog spinal cord. In both preparations, endings with S-type (spherical) vesicles were more numerous than those with F (flat) vesicles, the $\text{S}\text{F}$ synapse ratio being approximately constant when measured at the soma, the proximal dendrites or the fine dendrites. However, the proportion of S boutons, as well as the proportion of degenerated boutons of afferent fibers in close apposition with other boutons, was higher on the fine dendrites in the lateral neuropile than on the motoneuron cell body and proximal dendrites, and in all cases significantly higher than the proportion of F boutons showing close appositions with other boutons. Degenerated boutons from descending fibers were found to synapse over the motoneuron cell body and proximal dendrites, but these boutons were rarely seen to be closely apposed to other boutons. The possible relation of these findings to the mechanisms generating primary afferent depolarization is discussed.     

Synaptic organization of the Purkinje cells in the cerebellum of the cat

Summary An attempt at distinction between excitatory and inhibitory synapses is made in the cat cerebellum. The former are assumed to contain spheroid vesicles (S-type) of average diameter of 500 Å, while the latter flattened vesicles (F-type) of smaller size than the former. The elongation index (the ratio of the length of major versus minor axis of the vesicles) of S-type synaptic vesicles was about 1.2, while that of the F-type was more than 1.7. Parallel fibers of granule cells make S-type synaptic contacts (en-passant type or crossing-over synapse) mostly on the spines of the smaller branchlets of Purkinje cells. Climbing fibers make also S-type synapses on the smaller spines with short necks of the larger dendrites of Purkinje cells, but not frequently on the direct surface of them. It must be emphasized that almost no F-type synapse has been recognized which makes synaptic contacts directly on the spine of any type. It makes synaptic contacts usually on the direct surface of dendrites of Purkinje cells. Basket cell axons embrace directly the somas of the Purkinje cells. Their synaptic contacts were always of F-type and of en-passant character.The hypothesis is proposed that excitatory (E-type) synapses can be identified with synapses of S-type, whereas inhibitory (I-type) synapses would correspond to the F-type terminals.     

Ultrastructural observations on beaded alpha-motoneuron dendrites

Beaded dendrites of alpha-motoneurons intracellularly labelled with horseradish peroxidase (HRP) were studied ultrastructurally in eight adult cats. For comparison, adjacent unlabelled beaded dendrites of unknown origin were also included in the study. Electron microscopy revealed no signs of degeneration or poor fixation according to common criteria. With the exception of the HRP-reaction product no difference in structure was observed between labelled and unlabelled beaded dendrites. Both the beads and their interconnecting segments were postsynaptic to boutons of normal appearance containing spherical (S-type boutons) or flattened vesicles (F-type boutons). The values for synaptic covering and synaptic packing density of the beaded dendritic regions, which usually were located in the periphery of the dendritic trees, were clearly lower than values obtained previously for cell bodies and proximal dendrites of alpha-motoneurons.     

Recovery of synapses in axotomized adult cat spinal motoneurons after reinnervation into muscle

 Peripheral axotomy of adult cat spinal motoneurons induces a marked loss of synaptic boutons from the cell bodies and dendritic trees. The aim of the present study was to analyze the recovery of synaptic contacts in axotomized motoneurons following reinnervation into muscle. Adult cat spinal motoneurons were first deprived of their muscular contacts for 12 weeks and, then, allowed to reinnervate their target muscle. Two years later, regenerated motoneurons were labeled with horseradish peroxidase to allow quantitative ultrastructural analyses of the synaptic covering of the cell bodies and dendrites. Presynaptic boutons were classified according to their size and the shape of their synaptic vesicles. Results show that a recovery of synaptic covering occurs in the axotomized neurons after muscle reinnervation, but it affects various bouton types to different degrees. The number of S-type boutons synapsing with the soma was 70% higher after reinnervation than at 12 weeks after axotomy, while the number of F-type boutons had increased by only 13%. Compared with the normal situation, the number of S-type boutons synapsing with the proximal dendrites increased from 82% at 12 weeks after axotomy to 180% in the reinnervated state. In conclusion, in adult cat spinal motoneurons, the reestablishment of muscular contact is followed by a normalization of some of the synaptological changes induced by a prolonged state of axotomy. In certain respects restitution is incomplete, but in others it results in overcompensation. Received: 10 December 1997 / Accepted: 30 July 1998     

Ultrastructural observations on beaded α-motoneuron dendrites

Beaded dendrites of 1α-motoneurons intracellularly labelled with horseradish peroxidase (HRP) were studied ultrastructurally in eight adult cats. For comparison, adjacent unlabelled beaded dendrites of unknown origin were also included in the study. Electron microscopy revealed no signs of degeneration or poor fixation according to common criteria. With the exception of the HRP-reaction product no difference in structure was observed between labelled and unlabelled beaded dendrites. Both the beads and their interconnecting segments were postsynaptic to boutons of normal appearance containing spherical (S-type boutons) or flattened vesicles (F-type boutons). The values for synaptic covering and synaptic packing density of the beaded dendritic regions, which usually were located in the periphery of the dendritic trees, were clearly lower than values obtained previously for cell bodies and proximal dendrites of a-motoneurons.     

Exogenous brain-derived neurotrophic factor regulates the synaptic composition of axonally lesioned and normal adult rat motoneurons

Brain-derived neurotrophic factor has previously been shown to promote survival and axonal regeneration in injured spinal motoneurons and, also, to modulate synaptic transmission and regulate the density of synaptic innervation in a variety of neurons. The present light and electron microscopic study demonstrates synaptotrophic effects of exogenously applied brain-derived neurotrophic factor on the synaptic composition of both normal and axonally lesioned adult rat spinal motoneurons. After L5-L6 ventral root avulsion, a massive loss of all types of boutons occurred on the somata of the lesioned motoneurons which persisted for at least 12 weeks postoperatively. We found that (i) intrathecal infusion of brain-derived neurotrophic factor during the first postoperative week did not prevent the synaptic detachment and activation of glial cells; (ii) prolonged treatment for four weeks restored synaptic covering and significantly reduced microglial reaction; (iii) the synaptotrophic effect remained significant for at least eight weeks after cessation of the treatment; (iv) brain-derived neurotrophic factor mainly supported F-type boutons with presumably inhibitory function, while it had little effect on S-type boutons associated with excitatory action; and (v) in normal unlesioned motoneurons, four weeks of treatment with brain-derived neurotrophic factor induced sprouting of F-type boutons, a loss of S-type boutons and motoneuron atrophy.The present data show that exogenous neurotrophins not only help to restore synaptic circuitry in axonally injured motoneurons, but also strongly influence the synaptic composition in normal motoneurons.     

Monoamine innervation of the oculomotor nucleus in the rat. A radioautographic study

The serotonin and noradrenaline innervations of the rat oculomotor nucleus were examined by high resolution radioautography after in vivo labeling with tritiated 5-hydroxytryptamine and dopamine, respectively. Noradrenaline as well as serotonin endings (axonal varicosities) pervaded the entire nucleus, but the latter were at least six times more numerous (1.3 X 10(6) per mm3 of tissue) and were often found in the immediate vicinity of neuronal somata and proximal dendrites. The axon terminals of both types were of similar size and exhibited some large dense-cored vesicles in association with aggregated small and clear vesicles. The dense-cored vesicles were, however, more frequent and the content in clear vesicles more pleomorphic in serotonin than noradrenaline endings. In single thin sections, the proportion of noradrenaline and serotonin profiles exhibiting a synaptic junction was relatively small (15%). These were either symmetrical or asymmetrical when made on dendritic branches but invariably symmetrical on spines. In addition, a significant number of serotonin terminals were seen in close apposition or synaptic contact with neuronal perikarya and large dendrites, allowing for a direct, "proximal" action of serotonin. Moreover, many such terminals appeared to be coupled with unlabeled endings of another category, characterized by dispersed, uniformly round and clear synaptic vesicles, providing an alternate route for a proximal effect of serotonin in the oculomotor nucleus. In line with previous investigations on other motor nuclei, these data support the likelihood of a close involvement of both noradrenaline and serotonin in the control of motoneuronal activity.     

The dendrites of single brain-stem motoneurons intracellularly labelled with horseradish peroxidase in the cat. An ultrastructural analysis of the synaptic covering and the microenvironment

Two laryngeal motoneurons intracellularly stained with horseradish peroxidase were studied ultrastructurally. The precise position of the ultrastructural observations made along the dendrites was obtained from the computer-reconstruction of the motoneurons in three dimensions. The shape and the size of the synaptic boutons, the percentage of membrane covered by bouton appositions and active zones, the number of boutons per 100 microns2 (packing density) were analysed on the soma and on the labelled dendrites at different distances from the soma up to 1000 microns. The results revealed no important regional differences in the mean length of synaptic apposition. The packing density was in the range of 9.3-14.9 boutons per 100 microns2 and was not correlated with the distance from the soma. The percentage apposition covering was higher on the soma and the proximal part of the dendrites than on the remaining part of the dendritic arborization. Close appositions between labelled dendrite and unlabelled somata and/or dendrites together with dendro-dendritic synapses suggested the possibility that the dendrites may be involved in local cell-to-cell communication. Microdendrites emerging from the soma or the proximal dendrites were contacted by synaptic boutons which may be more efficient as revealed by computation.     

Immunoelectron microscopic analysis of the synaptic connectivity of serotoninergic neurons grafted to the 5,7-dihydroxytryptamine-lesioned rat spinal cord.

The connections between the host and 5-hydroxytryptamine-containing neurons grafted to the spinal cord have been analysed using electron microscopic immunohistochemistry. Adult rats with 5,7-dihydroxytryptamine lesions of the brain and spinal cord received implants of embryonic medullary raphé neurons at three sites in the spinal cord. Eight to 10 months after grafting, the transplanted 5-hydroxytryptamine-positive neurons had formed extensive and complex contacts with spines, dendrites, perikarya and vesicle-containing structures in both the dorsal and ventral horns. Reinnervation of laminae IV-VI was less rich. In the graft itself, connections were also made between non-immunoreactive varicosities and 5-hydroxytryptamine-containing dendrites, and somata, but the exact origin of the afferents was not determined. Outside the implant site, no obvious synaptic junctions onto grafted 5-hydroxytryptamine-immunoreactive boutons were obvious, although labelled and unlabelled varicosities were often in close apposition. Synaptic junctions in the dorsal horn were predominantly symmetric, with the presynaptic varicosity containing mostly small agranular vesicles. By contrast, in the ventral horn most junctions were asymmetric, while the presynaptic element contained both small agranular and large dense-core vesicles. The results demonstrate that the types of synaptic contacts formed between the grafted 5-hydroxytryptamine neurons and the host spinal cord are remarkably similar to those found in intact spinal cord. In addition, the division of morphological differences that exists between 5-hydroxytryptamine-containing boutons in the normal dorsal vs ventral horns is also apparent in the transplanted animals. Finally, there appear to be present several anatomical substrates for the regulation by the host of 5-hydroxytryptamine output from the grafted neurons.     

Neuronal and synaptic organization of the centromedian nucleus of the monkey thalamus: a quantitative ultrastructural study, with tract tracing and immunohistochemical observations

Summary The ultrastructure of the centromedian nucleus of the monkey thalamus was analysed qualitatively and quantitatively and projection neurons, local circuit neurons, and synaptic bouton populations identified. Projection neurons were mostly medium-sized, with oval-fusiform or polygonal perikarya, few primary dendrites, and frequent somatic spines; local circuit neurons were smaller. Four basic types of synaptic boutons were distinguished: (1) Small- to medium-sized boutons containing round vesicles (SR) and forming asymmetric contacts, identified as corticothalamic terminals. (2) Heterogeneous medium-sized boutons with asymmetric contacts and round vesicles, similar to the so-called large round (LR) boutons, which were in part of cortical origin. (3) Heterogeneous GAD-positive small- to medium-sized boutons, containing pleomorphic vesicles and forming symmetric contacts (F1 type), which included pallidothalamic terminals. (4) Presynaptic profiles represented by GAD-positive vesicle-containing dendrites of local circuit neurons. Complex synaptic arrangements, serial synapses and triads with LR and SR boutons engaging all parts of projection neuron dendrites and somata, were seen consistently, whereas classical glomeruli were infrequent. LR and SR boutons also established synapses on dendrites of local circuit neurons. F1 boutons established synapses on projection neuron somata, dendrites and initial axon segments. Compared to other previously studied motor-related thalamic nuclei, differences in synaptic coverage between proximal and distal projection neuron dendrites were less pronounced, and the density of synapses formed by local circuit dendrites on projection neuron dendrites was lower. Thus, compared to other thalamic nuclei, the overlap of different inputs was higher on monkey centromedian cells, and centromedian inhibitory circuits displayed a different organization.     

Substance P-immunoreactive elements in laminae I and II of the chicken spinal cord: a light- and electron-microscopic study

Substance P (SP)-immunoreactive (IR) elements were studied in laminae I and II of the chicken spinal cord in conjunction with an anti-SP monoclonal antibody at light- and electron-microscopic levels by means of the indirect antibody peroxidase-antiperoxidase technique. At the light-microscopic level, SP-IR elements were most intensely observed in the dorsolateral portion of the dorsal horn, laminae I and II. Electron-microscopically, SP-IR boutons contained large spherical dense-cored vesicles (diameter range: 60-125 nm) and spherical clear vesicles. They were subdivided into two groups: large SP-IR boutons, which were the central terminals in synaptic glomeruli, and small SP-IR boutons. In the synaptic glomerulus, two kinds of non-IR presynaptic profiles made axo-axonic synapses with the SP-IR central terminal: one was the presynaptic profile containing pleomorphic clear vesicles and the other was the presynaptic profile containing large dense-cored vesicles. A 'septate junction'-like structure was observed between large SP-IR boutons in synaptic glomeruli. The present results suggest that SP-containing primary afferents are modulated presynaptically by two different neurotransmitter or modulator systems.     

Vestibulospinal projections in the toad. An experimental electron microscopic study

An experimental electron microscopical study has been made on the mode of termination of the vestibulospinal projections in the toad. Degenerating fibers and boutons were observed in the ventromedial region of the spinal cord in which five types of terminals were identified in normal preparations: S-type, F-type, C-type, NfS-type and NfF-type. The latter two types represent the largest terminals in the spinal gray. In the medial region of the ventral horn they seem to belong to supraspinal fibers.Following electrolytic lesions of the dorsal and ventral nuclei nervi octavi, some of the NfF-type of boutons undergo filamentous degeneration which progresses into the dark type at longer survival periods (7–20 days). In two animals, a few NfS-type of boutons also display filamentous hypertrophy.Problems connected with the filamentous type of degeneration in amphibians and with the identification of bouton types belonging to the vestibulospinal pathway are discussed.Degenerating fibers and boutons are seen at all levels of the spinal cord, ipsilateral and also, in smaller number, contralateral to the lesions. The vestibulospinal fibers terminate predominantly on large and small dendrites, and only occasionally on the somata of large neurons on the ipsilateral side.     

The ultrastructure of the nucleus of the oculomotor nerve (somatic efferent portion) of the cat

Summary The ultrastructure, of the somatic efferent portion of the nucleus of the oculomotor nerve was studied in four adult cats. The neuronal population is composed of neurons of variable size. A continuous pattern of morphological aspects is evident between the large neurons, which show abundant cytoplasm with well developed organelles and the small neurons which have a reduced amount of cytoplasm. The dendrites are generally smooth, with few short spines. Axo-somatic and axo-dendritic synapses are numerous. Synaptic boutons are also present on the axon hillock.The neuropil is characterized by the occurrence of small groups of dendrites which may be in direct touch with their membranes. Direct membrane appositions may also occur between neighbouring neurons and between the cell somata and tangentially running dendrites. Generally beneath the site of apposition there is accumulation of mitochondria, multivesicular bodies, coated vesicles and moderately dense amorphous material. The morphological features suggest the possibility of cellular interchanges at the sites of direct membrane apposition.Five types of synaptic boutons were recognized on the basis of their vesicular content, the presence of abundant filaments in the pre-synaptic bag, the occurrence of post-synaptic specializations. The, different synaptic types and their distribution are similar to those reported in the spinal motor nuclei. Many of the synapses make synaptic contacts with two or more post-synaptic elements. Axo-axonic synapses were also observed.     

Immunocytochemical and electron microscopic study of serotonin neuronal organization in the dorsal raphe nucleus of the monkey

Serotonin neurons in the dorsal raphe nucleus were identified using an antibody to a serotonin-bovine serum albumin conjugate and the peroxidase anti-peroxidase method. Nerve cell bodies showing serotonin-like immunoreactivity ranged in size from 15 to 22 micron in diameter; their dendrites were also immunoreactive. Immunostaining was present in the cytoplasmic matrix, outer membranes of mitochondria, rough endoplasmic reticulum, multivesicular bodies and dense-cored vesicles. Heavily immunoreactive axonal varicosities contained small round vesicles (18-35 nm) and larger dense-cored vesicles (50-90 nm). Both unmyelinated (0.2-0.5 micron) and myelinated (0.8-1.1 micron) serotonin-like immunoreactive axons were found, often interspersed within bundles of similar caliber unlabeled axons. Serotonin-like immunoreactive somata and dendrites were postsynaptic to numerous unlabeled terminals that contained either (a) clear round vesicles (18-25 nm) with many small dense-cored vesicles (30-50 nm), (b) clear round vesicles (18-25 nm) with large dense-cored vesicles (90-110 nm) or (c) clear round vesicles (18-25 nm) with or without flat vesicles. In addition pairs of unlabeled terminals formed crest synapses onto serotonin-like immunoreactive dendritic spines. This variety of unlabeled terminals making contact with serotonin-like immunoreactive elements suggests that several neuronal systems with possibly different transmitters may regulate serotonin raphe neurons. We occasionally observed serotonin-like immunoreactive dendrites and terminals in apposition to other serotonin-like immunoreactive dendrites with membrane specializations at the site of contact. This might represent a possible site for the self inhibition of serotoninergic neurons reported in physiological studies of the serotonin system in the dorsal raphe nucleus.     

Morphological and biochemical properties of synaptic vesicles isolated from guinea pig brain.

Two major types of different synaptic vesicles were isolated by centrifugation from the brain homogenate of guinea-pigs. The size and shape of vesicles in each type are different. The one type is spheroid and the other is flattened. The former is larger than the latter. These two types are equivalent to the S-type and F-type vesicles in tissue sample. Sedimentation coefficients of S-type and F-type are 64s and 59s respectively. Amino acid composition of vesicles in each type is almost similar, but the content of amino acids in the S-type is larger than in the F-type except for valine.     

Quantitative electromicroscope study of the oculomotor parasympathetic neurons projecting to the ciliary ganglion in cats: comparison of the synaptic (axo-somatic and axo-proximal dendritic) organization of anterior-dorsal and ventral cell groups

The synaptic organization of the oculomotor parasympathetic preganglionic neurons (OPNs), labeled retrogradely after a horseradish peroxidase (HRP) injection into the ciliary ganglion, was studied in cats by electron microscopy. We divided the OPNs into two groups, anterior-dorsal (ADG) and ventral (VG) cell groups, based upon physiological studies in cats suggesting that accomodation-related OPNs are predominantly located anterior and dorsal to the somatic nuclei of the oculomotor nuclear complex (i.e., the anteromedian and Edinger Westphal nuclei, and the ventral central gray area), while pupillo-constriction-related OPNs are predominantly located ventral to the somatic nuclei (i.e., the ventral tegmental area). The synaptic organization of these two groups was quantitatively compared, using a nested analysis of variance to determine statistical significance (P<0.05). Partial reconstructions of the labeled somata and proximal dendrites were made from tracings of electron micrographs of every 2nd section in serial ultrathin sections that included the nucleolus or were adjacent to sections that included the nucleolus. The mean number of boutons of apposition on a reconstructed labeled soma of VG was significantly greater than that of ADG (mean ±SD; ADG, 5.3±3.3; VG, 8.6±3.2). The mean synaptic density on a VG soma was significantly greater than on an ADG soma (mean±SD; ADG, 3.74±2.11 counts/100 (m²; VG, 6.30±1.99 counts/100 m²). The mean synaptic covering ratio on a VG soma was significantly greater than on an ADG soma (mean±SD; ADG, 5.21±2.91%; VG, 10.14±3.76%). The mean estimated number of boutons of apposition on a VG soma was significantly greater than on an ADG soma (mean±SD: ADG, 53±36; VG, 100±48). Boutons were classified on the basis of the shape of their synaptic vesicles as S-type (containing spherical clear synaptic vesicles) or P-type (containing both flattened and spherical clear synaptic vesicles). The mean S-type/S+P-type bouton ratio on a VG soma was significantly greater than on an ADG soma (mean±SD; ADG, 0.31±0.20; VG, 0.67±0.18). The differences demonstrated in this study reinforce, morphologically, the assumption of functional localization of OPNs, and further allow us to estimate the relative characteristics of the synaptic organization of accommodation-related OPNs and pupillo-constriction-related OPNs.     

Ultrastructural identification of degenerating boutons of monosynaptic pathways to the lumbosacral segments in the cat after spinal hemisection

Summary Electrophysiological results suggest that the supra-spinal monosynaptic pathways to the lumbosacral motoneurones in cats are located in the vestibulospinal and reticulospinal tracts. In the present study these pathways were interrupted in the upper thoracic region by hemisection of the spinal cord. The boutons undergoing degeneration were then studied both qualitatively and quantitatively by means of serial and single sections of selected areas of the ventral horn in spinal cord segments L 6, L 7 and S 1. The degenerating boutons were distributed over the cell soma and larger dendrites (but not the most proximal portions) of the motoneurones. Two days postoperatively boutons showing signs of degeneration were readily identified in the latero-ventral motoneurone nucleus and Rexed's lamina VIII. These boutons which were swollen had a greatly increased number of neurofilaments and a decreased number of synaptic vesicles. Five days postoperatively these boutons which were now shrunken, contained swollen mitochondria and a dense, granular material. Between 5 and 8 days post-operatively the boutons were invaded by glial cell processes and digested. A small percentage of the F- and S-type boutons (approximately 0.5 and 3.0% respectively) were identified as forming the terminals of the experimentally interrupted pathways. This study did not exclude the possibility that a proportion of these boutons were the terminals of the propriospinal, monosynaptic fibers derived from the cervical region of the cord.On sabbatical leave from the Department of Zoology, University of Melbourne     

Changes in synaptology of adult cat spinal α-motoneurons after axotomy

The aim of this electron-microscopic study was to analyze the distribution of synaptic contacts on the cell bodies and dendrites of permanently axotomized adult cat spinal α-motoneurons. Following transection and ligation of the medial gastrocnemius nerve, the synaptic covering of the cell bodies and three different dendritic compartments of homonymous α-motoneurons was analyzed quantitatively at 3, 6, and 12 weeks postoperatively. The synaptic boutons were classified according to their size and the shape of their synaptic vesicles. On the soma, a transient increase in the number of boutons was noted at 3 weeks and 6 weeks postoperatively, while after 12 weeks the bouton number had decreased to half of its normal value. The transient increase was mainly due to an increase in the number of F-type boutons. At 12 weeks postoperatively, the synaptic covering was reduced by 83% on the soma and by 57% on the proximal dendrites. In the distal dendritic regions, the values for synaptic covering remained largely unchanged. In summary, axotomized motoneurons exhibit a reduction in synaptic covering which is maximal on the cell body and becomes less pronounced centrifugally along the dendrites. However, if also taking into account the loss of distal dendritic branches that occurs in axotomized motoneurons, the total loss of boutons is several times larger in the dendrites than on the soma. Received: 18 October 1996 / Accepted: 13 June 1997     

Inhibitory zinc-enriched terminals in mouse spinal cord

The ultrastructural localization of zinc transporter-3, glutamate decarboxylase and zinc ions in zinc-enriched terminals in the mouse spinal cord was studied by zinc transporter-3 and glutamate decarboxylase immunohistochemistry and zinc selenium autometallography, respectively.The distribution of zinc selenium autometallographic silver grains, and zinc transporter-3 and glutamate decarboxylase immunohistochemical puncta in both ventral and dorsal horns as seen in the light microscope corresponded to their presence in the synaptic vesicles of zinc-enriched terminals at ultrastructural levels. The densest populations of zinc-enriched terminals were seen in dorsal horn laminae I, III and IV, whereas the deeper laminae V and VI contained fewer terminals. At ultrastructural levels, zinc-enriched terminals primarily formed symmetrical synapses on perikarya and dendrites. Only relatively few asymmetrical synapses were observed on zinc-enriched terminals. In general, the biggest zinc-enriched terminals contacted neuronal somata and large dendritic elements, while medium-sized and small terminals made contacts on small dendrites. The ventral horn was primarily populated by big and medium-sized zinc-enriched terminals, whereas the dorsal horn was dominated by medium-sized and small zinc-enriched terminals.The presence of boutons with flat synaptic vesicles with zinc ions and symmetric synaptic contacts suggests the presence of inhibitory zinc-enriched terminals in the mammalian spinal cord, and this was confirmed by the finding that zinc ions and glutamate decarboxylase are co-localized in these terminals. The pattern of zinc-enriched boutons in both dorsal and ventral horns is compatible with evidence suggesting that zinc may be involved in both sensory transmission and motor control.