Spine Necks Research Articles

Large GABA cells of chick ectostriatum: anatomical evidence suggesting a double GABAergic disinhibitory mechanism. An electron microscopic immunocytochemical study

In an extension of our previous light microscopic observations, a type of neuron which shows GABA-like immunoreactivity was identified and described in the ectostriatal core of young domestic chicks, using pre- and postembedding electron microscopic immunocytochemistry. Large GABA immunopositive (GABA+) cells are characterized by an ovoidal or polygonal soma of 12-16 micron diameter, uniformly distributed nuclear chromatin, a prominent Golgi apparatus and an abundance of rough endoplasmic reticulum. In addition to axodendritic terminals, large GABA neurons receive numerous axosomatic synapses of both symmetrical and asymmetrical types covering a substantial part of their perikaryal surface. Axosomatic terminals with symmetrical junctions are usually immunoreactive to GABA whereas the boutons with asymmetrical synaptic specialization are immunonegative. GABA+ boutons also synapse with dendritic spine necks presumably belonging to projection neurons. These terminals usually contain loosely packed synaptic vesicles without any marked accumulation near the synaptic cleft. Large GABA+ terminals with densely packed vesicles were found to synapse with axon hillocks. Based on known descriptions of ectostriatal cytoarchitecture and synaptology, it is suggested that the GABA+ cells of chick ectostriatum represent inhibitory interneurons which may be equivalent to GABAergic non-pyramidal neuronal types of mammalian visual cortex. GABA+ axosomatic synapses afferent to large GABA cells are likely to form the structural basis for a disinhibitory mechanism in the avian ectostriatum.

In the rat medial nucleus accumbens, hippocampal and catecholaminergic terminals converge on spiny neurons and are in apposition to each other

The nucleus accumbens septi (Acb) represents an interface between limbic and motor systems and a site for modulation of these integrative functions by ascending catecholaminergic, principally dopaminergic, axons. This modulatory regulation is most likely attributed to pre- or postsynaptic associations between limbic telencephalic and brainstem afferents. In the present investigation, we examined the ultrastructure and synaptic associations of hippocampal afferents, as well as their relation to catecholaminergic terminals, in the medial Acb of adult rats. Hippocampal afferents were identified by anterograde transport of wheat germ agglutinin-conjugated horseradish peroxidase (WGA-HRP) injected in the ventral subiculum, and by anterograde degeneration seen 2–3 days following lesion of the fimbria. Specific comparisons between these methods were made (1) to determine whether similar populations of terminals were labeled and (2) to assess the feasibility of combining degeneration with immunoperoxidase labeling for the catecholamine synthesizing enzyme, tyrosine hydroxylase (TH). Hippocampal afferents labeled with HRP were finely myelinated or unmyelinated and gave rise to small terminals (mean diameter 0.58 μm) containing mostly clear, round vesicles. Of the HRP-labeled terminals which made recognizable junctions, 85% (104/122) formed asymmetric synapses with the heads of dendritic spines. The remainder either formed asymmetric axodendritic synapses or symmetric junctions. Degenerating terminals were significantly smaller (mean diameter 0.35 μm) than terminals labeled with HRP. However, these also formed principally asymmetric axospinous synapses (89/102, 87%). Whether identified by HRP transport or anterograde degeneration, the hippocampal afferents comprised approximately 10% of all terminals and 30% of all asymmetric axospinous synapses in the medial Acb. In contrast to hippocampal afferents, TH-labeled terminals formed primarily symmetric contacts with dendritic shafts and the heads and necks of spines. Quantitative analysis of sections containing both anterograde degeneration and TH-immunoreactivity showed that 25% (26/104) of associations formed by degenerating hippocampal terminals involved convergent inputs with TH-labeled terminals on the same postsynaptic structure. These included dual input either to the same spine head or to different parts of the same dendrite. In addition, the plasma membranes of hippocampal and TH-labeled terminals were often directly apposed to each other (10/58, 17% of axo-axonal associations formed by degenerating terminals), without recognizable synaptic specializations. These results provide ultrastructural evidence consistent with (1) hippocampal excitation (asymmetric junctions) and dopaminergic inhibition (symmetric junctions) of certain common spiny neurons, (2) modulatory interactions between these afferents at the postsynaptic level, and (3) presynaptic modulation between transmitters in two physiologically diverse pathways to the medial Acb.

Dendritic spines of CA 1 pyramidal cells in the rat hippocampus: serial electron microscopy with reference to their biophysical characteristics

Serial electron microscopy and 3-D reconstructions of dendritic spines from hippocampal area CA 1 dendrites were obtained to evaluate 2 questions about relationships between spine geometry and synaptic efficacy. First, under what biophysical conditions are the spine necks likely to reduce the magnitude of charge transferred from the synapses on the spine heads to the recipient dendrite? Simulation software provided by Charles Wilson (1984) was used to determine that if synaptic conductance is 1 nS or less, only 1% of the hippocampal spine necks are sufficiently thin and long to reduce charge transfer by more than 10%. If synaptic conductance approaches 5 nS, however, 33% of the hippocampal spine necks are sufficiently thin and long to reduce charge transfer by more than 10%. Second, is spine geometry associated with other anatomical indicators of synaptic efficacy, including the area of the postsynaptic density and the number of vesicles in the presynaptic axon? Reconstructed spines were graphically edited into head and neck compartments, and their dimensions were measured, the areas of the postsynaptic densities (PSD) were measured, and all of the vesicles in the presynaptic axonal varicosities were counted. The dimensions of the spine head were well correlated with the area of PSD and the number of vesicles in the presynaptic axonal varicosity. Spine neck diameter and length were not correlated with PSD area, head volume, or the number of vesicles. These results suggest that the dimensions of the spine head, but not the spine neck, reflect differences in synaptic efficacy. We suggest that the constricted necks of hippocampal dendritic spines might reduce diffusion of activated molecules to neighboring synapses, thereby attributing specificity to activated or potentiated synapses.

An hypothesis regarding the antipsychotic effect of neuroleptic drugs

The antipsychotic effect of neuroleptic drugs appears gradually over the course of several weeks of chronic drug administration. Neuroleptic drugs are thought to act by blocking dopamine receptors; however, the dopamine-blocking effect of neuroleptics appears rapidly. One effect of dopamine antagonists which develops slowly is dopaminergic supersensitivity. It is suggested that this dopaminergic supersensitivity is related to the development of tolerance to some of the acute sedative properties of neuroleptics, but not to the antipsychotic effect. A population of glutamate receptors which are postsynaptic to the cortico-striatal afferents is located on the same neurons as striatal dopamine receptors. These glutamate synapses are located on the heads of dendritic spines of striato-nigral projection neurons, while the dopaminergic synapses are predominantly located on the necks of these same dendritic spines. Similar relationships could exist for mesolimbic and mesocortical dopamine systems. In peripheral systems, postjunctional denervation supersensitivity is known to be nonspecific; in other words, denervation of a single innervation of an excitable cell can alter the response to a range of stimuli. The antipsychotic effect of neuroleptics is therefore suggested to be due to nonspecific postjunctional subsensitivity at glutamate synapses, which develops concomitant with supersensitivity at dopaminergic synapses.

Dendritic spines of rat cerebellar Purkinje cells: serial electron microscopy with reference to their biophysical characteristics

We have used serial electron microscopy and 3-dimensional reconstructions of dendritic spines from Purkinje spiny branchlets of normal adult rats to evaluate 2 questions about the relationship of spine geometry to synaptic efficacy. First, do relationships between spine geometry and other anatomical indicators of synaptic activity suggest that spine size and shape might be associated with synaptic efficacy? Reconstructed spines were graphically edited into head and neck compartments; the area of the postsynaptic density (PSD) was measured; the volume of spine smooth endoplasmic reticulum (SER) was computed; and all of the vesicles in the axonal varicosities were counted. Spine head volume and the volume of SER contained in the head are well correlated with the area of the PSD and the number of vesicles in the presynaptic axonal varicosity. Spine neck diameter does not fluctuate with PSD area, head volume, or the vesicle number. These results suggest that the dimensions of the spine head, but not of the spine neck, are likely to reflect differences in synaptic efficacy. Second, does the geometry of cerebellar spine necks reduce the transfer of synaptic charge to the recipient dendrite from the theoretical maximum that could be transferred if the synapse were on a dendritic shaft? Comparison of volume to surface area showed that the spine heads are approximately spherical and the necks are approximately cylindrical. Application of results from a biophysical model that assumed these geometrical shapes for spines (Wilson, 1984) showed that the cerebellar spine necks are unlikely to reduce transfer of synaptic charge by more than 5-20% even if their SER were to completely block passage of current through the portion of the neck that it occupies. We suggest that the constricted spine neck diameter might serve to isolate metabolic events in the vicinity of activated synapses by reducing diffusion to neighboring synapses, without significantly influencing the transfer of synaptic charge to the postsynaptic dendrite.

Synaptic organization of the striatum.

The striatum, the main component of the basal ganglia, is composed of mainly one type of neuron, the so-called medium spiny neuron. This neuron cell type, which constitutes over 90% of striatal neurons, is the major output neuron of the striatum. Combined ultrastructural neuroanatomical methods have elucidated the organization of afferent connectivity to these neurons. The major physiologic function of striatal efferent activity appears to be inhibition of tonically active GABAergic neurons in the globus pallidus and substantia nigra pars reticulata. Thus, the excitatory input from the cerebral cortex, whose afferents make asymmetric synapses with the spines of medium spiny neurons, appears to drive the efferent activity of the striatum. Other extrinsic and intrinsic afferent synapses are situated in a position to regulate the effect of the corticostriatal excitatory input to the medium spiny neurons. For example, dopaminergic afferents from the midbrain make mainly symmetric synapses with the spine necks and dendritic shafts of the medium spiny neurons. Medium spiny neurons themselves have local axon collaterals, in addition to their efferent axon that exits the striatum, which serve to link together local clusters of medium spiny neurons. These local axon collaterals, which contain either GABA, substance P, or enkephalin, also make mainly symmetric synapses with the necks of spines or dendritic shafts of medium spiny neurons. Other afferents with similar synaptic connections to these neurons arise from cholinergic or somatostatinergic striatal intrinsic neurons. Additionally, the patterns of extrinsic and intrinsic afferents to medium spiny neurons and their extrinsic projections are related to the organization of medium spiny neurons into two mosaically organized macroscopic compartments, the striatal patches and matrix.

The ultrastructural localization of calcium-activated protease "calpain" in rat brain.

Calpain I, a calcium-activated neutral protease which degrades a number of cytoskeletal proteins, has been implicated in the rapid turnover of structural proteins that may participate in synaptic plasticity. In the present study, an antibody raised against purified erythrocyte calpain I was biochemically characterized and demonstrated to specifically bind the Mr = 80,000 subunit of both rat erythrocyte and brain calpain I. This antibody was used to examine the cellular distribution of calpain I at the electron microscopic level in rat brain and spinal cord using the avidin-biotin immunocytochemical technique. Reaction product was observed throughout neuronal perikarya, within both axonal and dendritic processes, and within spine heads and necks. Postsynaptic densities in both shaft and spine synapses were also immunoreactive. Glial cell bodies and processes were densely stained. In both neurons and glia, the reaction product was deposited along cytoskeletal elements. The localization of calpain I immunoreactivity to glial processes suggests this degradative enzyme may play a role in the glial hypertrophy and process retraction seen in brain. The presence of the enzyme in spines and postsynaptic densities is consistent with the hypothesis that it is involved in the turnover of synaptic cytoskeleton, thus providing a means through which transient physiological events effect lasting changes in the chemistry and morphology of spines.

Ultrastructural pathology of dendritic spines in epitumorous human cerebral cortex.

Ultrastructural changes in the human epitumorous cerebral cortex were examined. A swelling of nerve cell perikarya, dendrites and axon terminals, and an occasional hyperplasia and disarray of microtubules and neurofilaments were observed. Accumulations of lysosomes, tubuloreticular structures and intracytoplasmic or intramitochondrial crystalloid inclusions were also found. Some myelinated axons were degenerated. Astrocytes and their processes were focally swollen. A mild swelling was found also in microgliocytes. Oligodendrocytes occasionally contained accumulations of dense bodies. Special attention was paid to dendritic spines. The spine surface morphology changed distinctly on swollen dendritic segments. The necks of most spines were short and wide, and numerous sessile forms were present in this location. The spine apparatus was often hypertrophied and disorganized. Smoothing out of spines on swollen dendrites is described and a possible functional significance of the observed changes in the symptomatology of brain tumors is hypothesized.

Fracture prevalence and bone mineral mass in osteoporosis measured with computed tomography and dual energy photon absorptiometry.

In 174 adults presenting with backache, bone densitometry was performed on the lumbar spine, both femoral necks, and one femoral shaft employing dual energy photon absorptiometry (DPA); in 112 of these, densitometry was undertaken on L4 using single energy quantitative computer assisted tomography (CT). Radiographs of the spine were obtained in all patients and those with known or suspected malignant disease were excluded. The subjects were divided into two groups according to the presence (n = 128) or absence (n = 44) of signs of vertebral collapse or compression. The predictive value of the densitometric results for the presence of vertebral fractures was calculated and used as an estimate of fracture risk. While CT showed somewhat higher predictive values than DPA of the spine or combinations of DPA results from spine and femoral necks, the difference was not significant. It is concluded that with both CT and DPA the probability of the presence or absence of fracture can be raised to 75-80% when the probability prior to the investigation is 50% and when the threshold values of the measurements are chosen so that their sensitivity and specificity are about equal.

Efferent synaptic connections of grafted dopaminergic neurons reinnervating the host neostriatum: a tyrosine hydroxylase immunocytochemical study

In adult rats with a unilateral 6-hydroxydopamine-induced destruction of the nigrostriatal dopamine (DA) pathway, grafts of embryonic substantia nigra can establish a new dopaminergic terminal fiber plexus in the previously denervated neostriatum and compensate for some of the behavioral deficits induced by the nigrostriatal lesion. In the present study the synaptic connections of the ingrowing DA fibers from the graft were analyzed ultrastructurally, using immunocytochemical localization of tyrosine hydroxylase (TH), in animals whose lesion-induced motor asymmetry had been completely compensated for by the nigral grafts. In two of the animals, horseradish peroxidase-wheatgerm agglutinin conjugate was injected into the graft in order to trace possible reciprocal afferent connections to the graft from the host striatum. TH-immunoreactive axons from the graft were seen to make abundant symmetric synapses with neuronal elements in the host neostriatum. Between 85 and 90% of these synapses were on dendritic shafts and spines, and the rest were on neuronal perikarya. Two principal targets were identified: dendrites of spiny neurons, the majority of which are likely to be striatal projection neurons; and the cell bodies of giant neurons, most (or perhaps all) of which are known to be cholinergic interneurons. The synapses made on dendritic spines, which constituted about 40% of all TH-positive synapses formed by the TH-positive neurons in the graft, resembled those seen in normal animals, both in that they made contacts with spine necks and in that they invariably were associated with an asymmetric TH-negative synapse contacting the spine head. The innervation of the giant cell perikarya, which constituted about 6% of all TH-positive synapses found, was strikingly abnormal in that the graft-derived TH-positive fibers formed dense pericellular "baskets" selectively around the giant cell bodies. Such arrangements were never seen in the normal striatum, nor did they occur in the intact contralateral striatum in the grafted animals. It is proposed that this apparent dopaminergic hyperinnervation from the graft could provide a powerful inhibition of the cholinergic interneurons in the reinnervated host striatum, and that such an inhibitory mechanism could assist in the graft-induced functional recovery by potentiating the functional effects of DA synapses terminating on the spiny efferent neurons. This dual innervation may thus help to explain why restoration of only a small proportion of the striatal DA innervation by the graft is sufficient to induce complete compensation of, e.g., motor asymmetry in the lesioned rats.(ABSTRACT TRUNCATED AT 400 WORDS)

Tyrosine hydroxylase-immunoreactive boutons in synaptic contact with identified striatonigral neurons, with particular reference to dendritic spines

Tyrosine hydroxylase-immunoreactive fibres in the rat neostriatum were studied in the electron microscope in order to determine the nature of the contacts they make with other neural elements. The larger varicose parts of such fibres contained relatively few vesicles and rarely displayed synaptic membrane specializations; however, thinner parts of axons (0.1–0.4 μm) contained many vesicles and had symmetrical membrane specializations, indicative of en passant type synapses. By far the most common postsynaptic targets of tyrosine hydroxylase-immunoreactive boutons were dendritic spines and shafts, although neuronal cell bodies and axon initial segments also received such input. Six striatonigral neurons in the ventral striatum were identified by retrograde labelling with horseradish peroxidase and their dendritic processes were revealed by Golgi impregnation using the section-Golgi procedure. The same sections were also developed to reveal tyrosine hydroxylase immunoreactivity and so we were able to study immunoreactive boutons in contact with the Golgi-impregnated striatonigral neurons. Each of the 280 immunoreactive boutons examined in the electron microscope displayed symmetrical synaptic membrane specializations: 59% of the boutons were in synaptic contact with the dendritic spines, 35% with the dendritic shafts and 6% with the cell bodies of striatonigral neurons. The dendritic spines of striatonigral neurons that received input from immunoreactive boutons invariably also received input, usually more distally, from unstained boutons that formed asymmetrical synaptic specializations. A study of 87 spines along the dendrites of an identified striatonigral neuron showed that the most common type of synaptic input was from an individual unstained bouton making asymmetrical synaptic contact (53%), while 39% of the spines received one asymmetrical synapse and one symmetrical immunoreactive synapse. It is proposed that the spatial distribution of presumed dopaminergic terminals in synaptic contact with different parts of striatonigral neurons has important functional implications. Those synapses on the cell body and proximal dendritic shafts might mediate a relatively non-selective inhibition. In contrast, the major dopaminergic input that occurs on the necks of dendritic spines is likely to be highly selective since it could prevent the excitatory input to the same spines from reaching the dendritic shaft. One of the main functions of dopamine released from nigrostriatal fibres might thus be to alter the pattern of firing of striatal output neurons by regulating their input.

Ultrastructural morphology of dopaminergic nerve terminals and synapses in the striatum of the rat using tyrosine hydroxylase immunocytochemistry: A topographical study

Structures immunoreactive for TH were examined in the rat striatum (including caudateputamen, nucleus accumbens and globus pallidus) by electron microscopy using the indirect peroxidase-labeled antibody method. Axon profiles and nerve terminals were the only structures stained by DAB precipitates in the axoplasm. The reactive boutons frequently contained a population of large pleomorphic vesicles (40–60 nm in diameter) but their interiors remained free of reactions. The synaptic contacts formed belonged principally to the symmetric type 2 of Gray while asymmetric Gray's type 1 synapses were rarely observed. The former were mostly apposed to dendritic trunks (rarely to perikarya) and the latter to dendritic spines. In addition, numerous immunoreactive nerve terminals were often found in close contact with small structures identified as the neck of dendritic spines. The active zone of these presumed synapses was characterized by a prominent thickening of the presynaptic membrane but the post-synaptic thickening was lacking. For similar reasons, it was difficult to assert the existence of one axo-axonic synapse when a positive nerve terminal was closely apposed to another one (generally unreactive). The exact morphology of dopaminergic synapses, or even their existence, have not been firmly established owing to large discrepancies between previous reports. No synapses, or synaptic contacts of either asymmetric or symmetric type were described according to the technique used. Our work was undertaken to elucidate further this problem, and in particular, we thought that regional differences in the synaptic organization might explain the divergent data. However, regional quantitative analysis performed in this study did not show significant differences in the percentage of either kind of synapses in the various striatal regions.

Preferential localization of polyribosomes under the base of dendritic spines in granule cells of the dentate gyrus

Electron microscopic studies of the dentate gyrus of the rat have revealed an apparent association between polyribosomes and dendritic spines. The present study was designed to elucidate the nature of this association. Our qualitative observations revealed that polyribosomes appeared primarily in two locations within the dendrite: (1) beneath the base of identified spines just subjacent to the intersection of the spine neck with the main dendritic shaft and (2) beneath mounds in the dendritic membrane which had the appearance of the base of a spine which extended out of the plane of section. To quantitatively define the nature of the apparent association, we attempted to determine (1) the proportion of spines with associated polyribosomes and (2) the proportion of the polyribosomes within dendrites which are associated with spine bases. Evaluation of profiles which were identifiable as spine neck-dendritic shaft intersections in a single section revealed that an average of 12.2% had associated polyribosomes. A serial section analysis revealed a somewhat higher incidence, however. Of a collection of 34 through-sectioned spines, 29% had polyribosomes which were revealed in one or more of the sections comprising the series. To evaluate what proportion of polyribosomes within the dendrite was associated with spines, we evaluated a series of photographs covering approximately 1250 micrometer2 of the dentate molecular layer from five animals, identifying all polyribosomes within dendrites and scoring their location as being (1) under spines, (2) under mounds, or (3) other. An average of 9.6% of the polyribosomes were found under processes identifiable as spine neck-dendritic shaft intersections, while 71.4% of the polyribosomes were found under mounds. Only 19% were not obviously associated with spines or mounds. Spine bases and mounds comprise only 3 of 35%, respectively, of the outline of dendritic profiles, however, indicating that the high incidence of polyribosomes under these elements cannot be accounted for by chance. To attempt to determine whether the mounds represent the base of dendritic spines, 68 mounds in 21 dentritic profiles were selected from the middle of the series of 20 serial sections. Ninetine of these mounds (28%) were continuous with an identified spine, and an additional 31% were continuous with thin processes of the size and appearance of spine necks. Thus, most of the mounds probably do represent the base of spines which extend out of the plane of a single section.

Morphological and physiological properties of neostriatal neurons: An intracellular horseradish peroxidase study in the rat

The physiological and morphological (light and electron microscopic) properties of four categories of neostriatal neurons (two types of medium spiny cells and two types of aspiny cells) were analyzed using the technique of intracellular recording and intracellular labeling with horseradish peroxidase. All of the neurons in this study had excitatory responses following stimulation of the cortex and substantia nigra except for the large aspiny neuron for which only substantia nigra inputs were tested. Morphologically, these neurons differed with respect to the size and shape of their somata, density and distribution of dendritic spines and distribution of their axons and axon collaterals. Ultrastructurally, observed somatic differences included the quantity and distribution of organelles and conformation of the nuclear envelope. The axons of one type of medium spiny neuron and the large aspiny neuron were myelinated. Unmyelinated axon collaterals arose from the axons of both types of medium spiny neurons and formed synapses on the dendritic shafts and possibly with the necks of spines of other neostriatal neurons. The parent axons of the most common type of medium spiny neurons were followed to the globus pallidus and, in some cases, to the internal capsule.

Distribution of thalamic input to different dendrites of a spiny stellate cell in mouse sensorimotor cortex

A Golgi impregnated, gold-toned [2] spiny stellate cell from layer IV of mouse SmI cortex was reconstructed in three dimensions from serial thin sections to assess the spatial relationships of the synapses onto its dendrites. The distribution of thalamocortical (TC) synapses with the reconstructed dendrites is presented in this report. Thalamocortical axon terminals were labeled by lesion induced degeneration which, in mouse SmI cortex, may reliably indicate the numbers of thalamocortical axon terminals. Results indicate that thalamocortical synapses, which are distributed over most regions of the dendritic tree, are arranged in a regular, periodic fashion on parts of two of the reconstructed dendrites. In these regions, the necks of spines receiving thalamocortical input attach to the dendrite shaft at intervals of about 5 μm. In many other regions of the dendritic tree, two spines receiving thalamocortical synapses are separated by a similar interval. Further studies are expected to determine the extent to which dendrites of spiny stelate cells and of other kinds of cortical neurons are contacted in a periodic fashion by thalamocortical axon terminals.

Early stages of uptake and transport of horseradish-peroxidase by cortical structures, and its use for the study of local neurons and their processes.

AbstractIt is well known that, when horseradish‐peroxidase (HRP) is injected into the central nervous system, it is taken up by nerve terminals and retrogradely transported to their cells of origin. We have now injected HRP into the visual cortex of cats in order to study light‐microscopically the pattern of its spread, the neural structures that take it up locally, and the initial steps of its retrograde descent towards the dorsal lateral geniculate nucleus (LGNd). Special attention was given to post‐injection times of ten minutes to eight hours, but animals with up tofive days' survival were also analyzed.It was found that the brown spot produced by the initial spread (primary diffusion) of HRP remains constant in size for about two hours and then spreads out (secondary diffusion) to involve neighboring visual cortical areas. Secondary diffusion, however, does not result in further retrograde labelling of LGNd neurons. Local nervous structures labelled during primary diffusion of HRP include a large number of infracortical axons, many of which show bifurcations directed towards the cortex. As HRP descends within these axons, fewer and deeper axonal branchings appear labelled, as though, after initial uptake, HRP were no longer taken up. The possible reasons for the lack of further HRP uptake, including observed local inflammatory reactions, are discussed. Intracortical labelled structures include neuron somata and processes. These resemble Golgi‐impregnated structures and, particularly at short survival times, show fine details such as dendritic spine necks, terminal bouton stalks, synaptic contacts, and a remarkable degree of axonal preterminal and terminal branching. Some of these structures were also examined electron‐microscopically.