Mono- and dual-frequency fast cerebellar oscillation in mice lacking parvalbumin and/or calbindin D-28k

Calbindin is a fast Ca2+-binding protein expressed by Purkinje cells and involved in their firing regulation. Its deletion produced approximately 160-Hz oscillation sustained by synchronous, rhythmic Purkinje cells in the cerebellar cortex of mice. Parvalbumin is a slow-onset Ca2+-binding protein expressed in Purkinje cells and interneurons. In order to assess its function in Purkinje cell firing regulation, we studied the firing behavior of Purkinje cells in alert mice lacking parvalbumin (PV-/-), calbindin (CB-/-) or both (PV-/- CB-/-) and in wild-type controls. The absence of either protein resulted in Purkinje cell firing alterations (decreased complex spike duration and pause, increased simple spike firing rate) that were more pronounced in CB-/- than in PV-/- mice. Cumulative effects were found in complex spike alterations in PV-/- CB-/- mice. PV-/- and CB-/- mice manifested approximately 160-Hz oscillation that was sustained by Purkinje cells firing rhythmically and synchronously along the parallel fiber axis. This oscillation was dependent on GABA(A), N-methyl-D-aspartate and gap junction transmission. PV-/- CB-/- mice exhibited a dual-frequency (110 and 240 Hz) oscillation. The instantaneous spectral densities of both components were inversely correlated. Simple and complex spikes of Purkinje cells were phase-locked to one of the two oscillation frequencies. Mono- and dual-frequency oscillations presented similar pharmacological properties. These results demonstrate that the absence of the Ca2+ buffers parvalbumin and calbindin disrupts the regulation of Purkinje cell firing rate and rhythmicity in vivo and suggest that precise Ca2+ transient control is required to maintain the normal spontaneous arrhythmic and asynchronous firing pattern of the Purkinje cells.     

'New' functions for 'old' proteins: the role of the calcium-binding proteins calbindin D-28k,calretinin and parvalbumin,in cerebellar physiology. Studies with knockout mice

Calretinin (CR), calbindin D-28k (CB) and parvalbumin (PV) belong to the large family of EF-hand calcium-binding proteins, which comprises more than 200 members in man. Structurally these proteins are characterized by the presence of a variable number of evolutionary well-conserved helix-loop-helix motives, which bind Ca2+ ions with high affinity. Functionally, they fall into two groups: by interaction with target proteins, calcium sensors translate calcium concentrations into signaling cascades, whereas calcium buffers are thought to modify the spatiotemporal aspects of calcium transients. Although CR, CB and PV are currently being considered calcium buffers, this may change as we learn more about their biology. Remarkable differences in their biophysical properties have led to the distinction of fast and slow buffers and suggested functional specificity of individual calcium buffers. Evaluation of the physiological roles of CR, CB and PV has been facilitated by the recent generation of mouse strains deficient in these proteins. Here, we review the biology of these calcium-binding proteins with distinct reference to the cerebellum, since they are particularly enriched in specific cerebellar neurons. CR is principally expressed in granule cells and their parallel fibres, while PV and CB are present throughout the axon, soma, dendrites and spines of Purkinje cells. PV is additionally found in a subpopulation of inhibitory interneurons, the stellate and basket cells. Studies on deficient mice together with in vitro work and their unique cell type-specific distribution in the cerebellum suggest that these calcium-binding proteins have evolved as functionally distinct, physiologically relevant modulators of intracellular calcium transients. Analysis of different brain regions suggests that these proteins are involved in regulating calcium pools critical for synaptic plasticity. Surprisingly, a major role of any of these three calcium-binding proteins as an endogenous neuroprotectant is not generally supported.     

Differences in locomotor behavior revealed in mice deficient for the calcium-binding proteins parvalbumin, calbindin D-28k or both

We investigated the role of the two calcium-binding proteins parvalbumin (PV) and calbindin D-28k (CB) in the locomotor activity and motor coordination using null-mutant mice for PV (PV-/-), CB (CB-/-) or both proteins (PV-/-CB-/-). These proteins are expressed in distinct, mainly non-overlapping populations of neurons of the central and peripheral nervous system and PV additionally in fast-twitch muscles. In a test measuring repeated locomotor activity during 18-20 days, the analysis revealed a slightly increased activity in mice lacking either protein, while the lack of both decreased the number of beams crossed during active periods. An increase in the characteristic speed during the first 8 days could be attributed to PV-deficiency, while the elimination of CB in CB-/- and double-KO mice decreased the percentage of fast movements at all time points. In the latter, additionally a reduction of the fastest speed was observed. The alterations in locomotor activity (fast movements, fastest speed) strongly correlate with the impairment in locomotor coordination in mice deficient for CB evidenced in the runway assay and the rotarod assay. The graded locomotor phenotype (CB>PV) is qualitatively correlated with alterations in Purkinje cell firing reported previously in these mice. The presence or absence of either protein did not affect the spontaneous locomotor activity when animals were placed in a novel environment and tested only once for 30 min. In summary, the lack of these calcium-binding proteins yields characteristic, yet distinct phenotypes with respect to locomotor activity and coordination.     

Initial junctions between developing parallel fibers and Purkinje cells are different from mature synaptic junctions

During postnatal development of cerebellar cortex, junctions are formed between parallel fiber axons and the shafts of Purkinje cell dendrites. These shaft junctions resemble synaptic junctions on spines in thin sections, in that the axon contains a cluster of synaptic vesicles, and the pre- and postjunctional membranes are lined by electron-dense material. The shaft junctions do not have the aggregate of particles arrayed on the extracellular half of the postjunctional membrane that is characteristic of mature spine synaptic junctions in freeze-fractured preparations, however, and so presumably have a different protein composition. Shaft junctions are transient specializations, present only in developing tissue, but do not appear simply to be intermediates in the formation of mature spine synaptic junctions. In normal development, spines are formed by the Purkinje cell dendrite at sites not occupied by shaft junctions. Moreover, in certain neurological mutant mice, shaft junctions form in the absence of spine synapses, and in other systems spines develop in the absence of shaft junctions. We suggest that shaft junctions are a class of synapse formed during development which is distinguished by its capacity to dissociate, and that only some fraction of the parallel fibers forming shaft synapses with a developing Purkinje cell will have established spine synapses in the adult.     

The role of parvalbumin and calbindin D28k in experimental scrapie

Aims: Prion diseases are generally characterized by pronounced neuronal loss. In particular, a subpopulation of inhibitory neurones, characterized by the expression of the calcium-binding protein parvalbumin (PV), is selectively destroyed early in the course of human and experimental prion diseases. By contrast, nerve cells expressing calbindin D28k (CB), another calcium-binding protein, as well as PV/CB coexpressing Purkinje cells, are well preserved. Methods: To evaluate, if PV and CB may directly contribute to neuronal vulnerability or resistance against nerve cell death, respectively, we inoculated PV- and CB-deficient mice, and corresponding controls, with 139A scrapie and compared them with regard to incubation times and histological lesion profiles. Results: While survival times were slightly but significantly diminished in CB–/–, but not PV–/– mice, scrapie lesion profiles did not differ between knockout mice and controls. There was a highly significant and selective loss of isolectin B₄-decorated perineuronal nets (which specifically demarcate the extracellular matrix surrounding the 'PV-expressing' subpopulation of cortical interneurones) in scrapie inoculated PV+/+, as well as PV–/– mice. Purkinje cell numbers were not different in CB+/+ and CB–/– mice. Conclusions: Our results suggest that PV expression is a surrogate marker for neurones highly vulnerable in prion diseases, but that the death of these neurones is unrelated to PV expression and thus based on a still unknown pathomechanism. Further studies including the inoculation of mice ectopically (over)expressing CB are necessary to determine whether the shortened survival of CB–/– mice is indeed due to a neuroprotective effect of this molecule.     

Structure of the Purkinje cell membrane in staggerer and weaver mutant mice

The structure of the plasma membrane of Purkinje cell dendrites was examined in weaver and staggerer mutant mice. Purkinje spines in weaver mice have clusters of intramembrane particles which resemble those at normal synapses with parallel fibers, even though no parallel fibers are formed in this mutant. There are very few spines in the staggerer, and these manifest normal intramembrane structure at contacts with climbing fibers. The spines which would normally be involved in synapses with parallel fibers are never formed in the staggerer, and the intramembrane structures which would have been associated with these spine synapses are also lacking. Thus, during postnatal cerebellar development in the mutants, acquisition of intramembrane specializations requires Purkinje spine formation but can occur independently of the development of parallel fibers.     

Electron microscopic analysis of postnatal histogenesis in the cerebellar cortex of staggerer mutant mice

Postnatal development of the cerebellar cortex has been compared in staggerer mutant and unaffected littermate mice. From postnatal day 3 to about day 21 the external granular layer in staggerer mice is decreased in thickness and area, and the number of postmitotic granule cell neurons is reduced. Those granule cells that are generated seem to differentiate normally, with the remarkable exception that they form only primitive junctions with Purkinje cell dendritic shafts. These specialized junctions are not superseded by the normal parallel fiber:Purkinje spine synapses and disappear by the third week. Purkinje cell somata and dendrites are smaller than normal at all stages examined. The dendrites are not confined to the sagittal plane as in the normal and, unique among mutant or other animals described to date, they exhibit virtually no branchlet spines. All other cortical synaptic relations of granule and Purkinje cells, including climbing fiber:Purkinje spine synapses, appear qualitatively normal. However, by 28 days virtually all staggerer granule cells have degenerated. While the primary genetic defect remains unknown, we postulate that the morphological abnormalities may be attributable to a block in the normal developmental relationship between granule cells and Purkinje cells. The small cell size and failure to form branchlet spines suggest that the Purkinje cell abnormality may be closer to the primary effect of the mutant gene than the more flagrant hypoplasia and degeneration of granule cell neurons.     

Evidence for loss of Purkinje cell dendrites during late development: a morphometric Golgi analysis in the mouse

The late postnatal development of the Purkinje cell dendritic tree in mouse cerebellar vermis was investigated in Golgi preparations by morphometric techniques in order to determine at what age adult characteristics of the Purkinje cell are achieved in the rodent brain which grows continuously throughout adult life. B6D2F1 hybrid mice were sacrificed at 9, 15, 20, 35 and 250 days of age. "Hind-brain" weights (by direct weighing) and vermis volume (determined histometrically from Golgi sections), both increased rapidly from 9 to 20 days of age and continued to increase steadily with advancing age. The growth of Purkinje dendritic field areas, determined by planimetric measurements from Golgi sections paralleled the growth curves for vermis cross-sectional area, vermis volume and "hindbrain" weight. However, stereological determinations revealed an unexpected disparity between the growth of the Purkinje dendritic field areas and changes in the total length of dendrites of Purkinje cells. The total dendritic branch length per Purkinje cell increased sharply up to 20 days of age but thereafter declined with advancing age. Dendritic spine counts on Purkinje cells revealed no change in the number of dendritic spines per unit length of dendrites between 20 and 250 days of age, however, since the Purkinje cell total branch length declined-calculations suggest that the total number of spines per cell declined after 20 days of age. Thus, the size of the cerebellum and the Purkinje cell dendritic tree continued to enlarge during late postnatal development; however, the total dendritic surface area and the total number of dendritic spines on each Purkinje cell, after reaching a peak at 20 days of age, declined with advancing age. The data suggest that the late postnatal development of the Purkinje cell dendritic tree is characterized by resorption as well as dendritic growth. The functional significance of such developmental remodelling is unknown.     

Spontaneous electrical activity and dendritic spine size in mature cerebellar Purkinje cells

Previous experiments have shown that in the mature cerebellum both blocking of spontaneous electrical activity and destruction of the climbing fibres by a lesion of the inferior olive have a similar profound effect on the spine distribution on the proximal dendrites of the Purkinje cells. Many new spines develop that are largely innervated by parallel fibers. Here we show that blocking electrical activity leads to a significant decrease in size of the spines on the branchlets. We have also compared the size of the spines of the proximal dendritic domain that appear during activity block and after an inferior olive lesion. In this region also, the spines in the absence of activity are significantly smaller. In the proximal dendritic domain, the new spines that develop in the absence of activity are innervated by parallel fibers and are not significantly different in size from those of the branchlets, although they are shorter. Thus, the spontaneous activity of the cerebellar cortex is necessary not only to maintain the physiological spine distribution profile in the Purkinje cell dendritic tree, but also acts as a signal that prevents spines from shrinking.     

Robust axonal sprouting and synaptogenesis in organotypic slice cultures of rat cerebellum exposed to increased potassium chloride

Organotypic slices of the rat cerebellum, cultured in physiological levels [K+]o (5 mM) for 14 days, loose the majority of granule cells in the anterior lobe resulting in few axons and atypical Purkinje cell dendrites with vacant spines. When the culture medium was switched from 5 mM to 20, 30 or 40 mM [K+]o during the last 7 days of cultures, slices developed axons with numerous vesicle-filled boutons that made synaptic contact with Purkinje cell spines. Most boutons had one or two spine profile contacts, while some were unusually large. Enlarged boutons abutted Purkinje cell somata or their dendrites, causing intervening spines to invaginate terminals to form rosette synaptic complexes. Calbindin immuno-labeling excluded Purkinje cell axonal collaterals as the source of rosette boutons and suggested a granule cell origin. Quantification of vacant spines as compared to those on boutons revealed a threshold for potassium, between 10 and 20 mM, where the number of synaptic spines increased and vacant spines decreased drastically. These findings suggest that elevated [K+]o triggers an activity-dependent plasticity in rat cerebellar slice cultures by promoting axonal sprouting with formation of vesicle-filled boutons and synaptogenesis on open receptor sites of Purkinje cell spines.     

The diffusional properties of dendrites depend on the density of dendritic spines

We combined computational modeling and experimental measurements to determine the influence of dendritic structure on the diffusion of intracellular chemical signals in mouse cerebellar Purkinje cells and hippocamal CA1 pyramidal cells. Modeling predicts that molecular trapping by dendritic spines causes diffusion along spiny dendrites to be anomalous and that the value of the anomalous exponent (d(w) ) is proportional to spine density in both cell types. To test these predictions we combined the local photorelease of an inert dye, rhodamine dextran, with two-photon fluorescence imaging to track diffusion along dendrites. Our results show that anomalous diffusion is present in spiny dendrites of both cell types. Further, the anomalous exponent is linearly related to the density of spines in pyramidal cells and d(w) in Purkinje cells is consistent with such a relationship. We conclude that anomalous diffusion occurs in the dendrites of multiple types of neurons. Because spine density is dynamic and depends on neuronal activity, the degree of anomalous diffusion induced by spines can dynamically regulate the movement of molecules along dendrites.     

Parvalbumin deficiency affects network properties resulting in increased susceptibility to epileptic seizures

Networks of GABAergic interneurons are of utmost importance in generating and promoting synchronous activity and are involved in producing coherent oscillations. These neurons are characterized by their fast-spiking rate and by the expression of the Ca(2+)-binding protein parvalbumin (PV). Alteration of their inhibitory activity has been proposed as a major mechanism leading to epileptic seizures and thus the role of PV in maintaining the stability of neuronal networks was assessed in knockout (PV-/-) mice. Pentylenetetrazole induced generalized tonic-clonic seizures in all genotypes, but the severity of seizures was significantly greater in PV-/- than in PV+/+ animals. Extracellular single-unit activity recorded from over 1000 neurons in vivo in the temporal cortex revealed an increase of units firing regularly and a decrease of cells firing in bursts. In the hippocampus, PV deficiency facilitated the GABA(A)ergic current reversal induced by high-frequency stimulation, a mechanism implied in the generation of epileptic activity. We postulate that PV plays a key role in the regulation of local inhibitory effects exerted by GABAergic interneurons on pyramidal neurons. Through an increase in inhibition, the absence of PV facilitates synchronous activity in the cortex and facilitates hypersynchrony through the depolarizing action of GABA in the hippocampus.     

The glutamate receptor 2 subunit controls post-synaptic density complexity and spine shape in the dentate gyrus

In adult brain the majority of AMPA glutamate receptor (GluR) subunits contain GluR2. In knock-out (KO) mice the absence of GluR2 results in consequences for synaptic plasticity including cognitive impairments. Here the morphology of dendritic spines and their synaptic contacts was analysed via three-dimensional reconstruction of serial electron micrographs from dentate gyrus (DG) of adult wild type (WT) and GluR2 KO mice. Pre-embedding immunocytochemical staining was used to examine the distribution and subcellular localization of AMPA receptor GluR1 and N -methyl- d -aspartate receptor NR1 subunits. There were no significant changes in synapse density in the DG of GluR2 KO compared with WT mice. However, in GluR2 KO mice there was a significant decrease in the percentage of synapses on mushroom spines but an increase in synapses on thin spines. There was also a large decrease in the proportion of synapses with complex perforated/segmented post-synaptic densities (PSDs) (25 vs. 78% in WT) but an increase in synapses with macular PSDs (75 vs. 22%). These data were coupled in GluR2 KO mice with significant decreases in volume and surface area of mushroom spines and their PSDs. In both GluR2 KO and WT mice, NR1 and GluR1 receptors were present in dendrites and spines but there was a significant reduction in NR1 labelling of spine membranes and cytoplasm in GluR2 KO mice, and a small decrease in GluR1 immunolabelling in membranes and cytoplasm of spines in GluR2 KO compared with WT mice. Our data demonstrate that the absence of GluR2 has a significant effect on both DG synapse and spine cytoarchitecture and the expression of NR1 receptors.     

Primary degeneration of the granular layer of the cerebellum (Norman type)

Summary Purkinje cells, impregnated with the rapid Golgi method, in a patient with primary degeneration of the granular layer showed abnormal orientation of the perikaryon and dendrites, reduction in size of the dendritic arbor, absence of spiny branchlets, and large numbers of stubby spines and hypertrophic spines on secondary dendritic branches; stubby spines and thorn-like formations were seldom observed on the primary dendrites and perikaryon of some Purkinje cells. These findings are similar to those described in the cerebellum of the homozygous weaver mutant mouse and in the cerebella of experimentally induced agranular phenocopies, thus suggesting that similar plastic changes occur in human and animal Purkinje cells as a result of the absence of parallel fibres input in early developmental stages. In addition, Purkinje cells in this patient showed club-shaped deformities in the distal region of primary dendrites, which were filled with radially oriented, short dendrites covered with stubby spines and hypertrophic spines. These latter structures appear to be fully impregnated asteroid bodies observed in paraffin sections.     

Guide for authors

Lurcher is an autosomal semidominant murine mutation. Lurcher heterozygotes, (+/Lc) lose all their cerebellar Purkinje cells by adulthood. Explants from 2 days postnatal (P2) wild-type (+/+) and +/Lc cerebellar cortex were grown in vitro to investigate the role of local neuronal environment and afferent input on the degenerating +/Lc Purkinje cell. In Lurcher explants, Purkinje cells were maintained for up to 25 days in vitro. No significant difference was observed between +/+ and +/Lc Purkinje cell numbers from 10 to 20 days in vitro, as revealed by calbindin-D immunoreactivity. Growing +/Lc explants in association with + / + explants resulted in no significant difference in Purkinje cell survival (10–20 days in vitro). Image analysis of the gross morphology of calbindin-D-immunostained Purkinje cells from +/+ and +/Lc explants grown in vitro revealed a significant decrease in the total area and dendritic lengths of +/Lc Purkinje cells (15 and 20 days in vitro). The fine structure of +/Lc and +/+ Purkinje cells was examined under the electron microscope (10–25 days in vitro). No difference in ultrastructure was observed between +/Lc and +/+ Purkinje cells grown in vitro, and many features similar to normal Purkinje cell development in vivo were present. These included monosynaptic parallel fibre synapses with Purkinje cell dendritic spines, other interneuron synapses with Purkinje cell dendrites and soma, astroglial investment, and minimal extracellular space in the neuropil. Unusual features observed included a persistence of the perisomatic spines in some Purkinje cells, an absence of Nissl bodies in the Purkinje cell perikaryon, naked Purkinje cell dendritic spines, and occasional heterol, ogous synapses. The results are discussed in the light of previous chimeric analysis of the Lurcher mutation, and a hypothesis is put forward to explain the survival of + /Lc Purkinje cells in vitro. © 1995 Wiley-Liss, Inc.     

Quantitative study of the Purkinje cell dendritic spines in the rat cerebellum

The number of spines on an individual Purkinje cell in the cerebellar cortex of the rat was determined by stereological methods. Investigations were based on thin section electron micrographs, freeze fracture replicas, and horseradish peroxidase labeled cells. Purkinje cell dendritic spines in our embedded material had a mean length of 1.4 +/- 0.05 micron and mean neck and head diameters of 0.22 +/- 0.01 micron and 0.45 +/- 0.02 micron, respectively. From these dimensions, an estimate of spine volume in embedded material of 0.132 micron 3 was obtained. The density of dendritic spines in our fixed material was 8.15 x 10(8) or 7.24 x 10(8) per microliters of molecular layer from volume fraction and density per mm2, respectively. The number of spines per linear micron of Purkinje cell spiny branchlet was 17.2 from freeze fracture and 17.6 from horseradish peroxidase labeled dendrites. These all indicate that there are between 154,000 and 175,000 spines on the dendritic tree of each Purkinje cell, considerably more than previously reported for the rat.     

Postnatal maturation of rat Purkinje cells cultivated in the absence of two afferent systems: an ultrastructural study.

Organized cultures of newborn rat cerebellum were established in Maximow chambers in order to study the maturation of Purkinje cells in absence of afferent systems. In the first model, standard cultures were devoid of extracerebellar afferents mossy and climbing fibers. Despite this absence, somatic spines appeared upon Purkinje cells during the first week in vitro and maturation proceeded normally except for the almost absence of spiny branchlets. Large dendritic trunks were studded with numerous spines, some of which were naked, a few bearing isolated post-synaptic densities and others occupied by boutons of parallel fibers. Stellate and basket axons made synapses upon the smooth portions of dendrites and soma. In a second model, the cultures were fed the antimitotic drug methylazoxymethanol (MAM) to prevent multiplication of granule cell precursors. Despite the absence of climbing and parallel fibers, the elongation of Purkinje dendrites was not prevented, but again the dendritic arbor consisted of large trunks studded with spines; somatic as well as dendritic spines were contacted by large boutons identified as Purkinje recurrent collaterals (PRC). It is concluded that the Purkinje cell possesses a large autonomy from afferent systems as to the growth of soma and dendrites. Conversely, the geometry of the dendrite and especially the spiny branchlets depend on the presence of both climbing and parallel fibers. One may conclude from the above experiments that specificity of synaptic contacts is maintained as long as postsynaptic sites are not devoid of their normal afferents. Heterologous synapses are formed when postsynaptic sites are present, their normal afferents absent and aberrant ones increasing by collateral sprouting. Such is probably the case in the second model of this study.     

Heterologous synapses upon purkinje cells in the cerebellum of the reeler mutant mouse: An experimental light and electron microscopic study

The projections of the spinal cord upon the cerebellum of normal and Reeler mutant mice were compared by light and electron microscopic methods after hemicordotomy. In both genotypes this afferent system projects to the cerebellar cortex and to the roof nuclei. In the Reeler, there is an additional projection among the Purkinje cells and interneurons of the central cerebellar mass. In both normal and Reeler cerebellar cortex this mossy fiber system terminates as large glomeruli. In Reeler the spinal projection also gives rise to a smaller terminal which is distributed both to the cortex and the central cerebellar mass. In both genotypes the dendrites of granule cells and the somata and dendrites of Golgi cells are synaptic targets of the glomeruli of the cortical projection. In Reeler both the glomeruli and smaller terminals also form heterologous synaptic contacts with dendritic spines of heterotopic intracortical and subcortical Purkinje cells. In both genotypes the synapses are exclusively type I. A second class of heterologous synapse, a type I junction between axons of Golgi cells and Purkinje cell spines, is also recognized in electron micrographs. The present study is the first unequivocal demonstration by experimental hodologic method of heterologous synaptic junctions in the mammalian central nervous system. The existence of such junctions in the cytoarchitectonically anomolous cerebellum of this mutant emphasizes the critical role played by the cellular environment in shaping neural circuits in the developing nervous system.     

Reorganization in granuloprival cerebellar cultures after transplantation of granule cells and glia. II. Ultrastructural studies

Cytosine arabinoside-induced granuloprival cerebellar cultures lack both granule cells and differentiated glia and demonstrate marked synaptic reorganization. After kainic acid-exposed cerebellar explants, which contain granule cells and mature glia, were transplanted to the granuloprival cultures, the following ultrastructural features were noted: (1) parallel fibers formed normal synapses with Purkinje cell dendritic spines as well as with basket/stellate cell somata; (2) sprouted Purkinje cell recurrent axon collateral terminals were markedly reduced in number; (3) Purkinje cells matured and lost perisomatic spines; (4) astroglia formed sheaths around Purkinje cell somata and dendrites; and (5) axonal myelination occurred. The transplanted cultures demonstrated ultrastructural restitution toward normal after addition of missing elements.     

Mature Purkinje cells in cerebellar tissue cultures: An ultrastructural study

Mature Purkinje cells in mouse cerebellar tissue cultures were morphologically analyzed by electron microscopy. Explants maintained for 19 to 31 days in vitro contained Purkinje cells that were similar in most respects to those described in vivo except for incomplete arborization of the dendritic trees. Typical features included (1) absence of Purkinje cell perisomatic spines; (2) a paucity of naked Purkinje cell dendritic spines; (3) a 1:1 relationship of Purkinje cell dendritic spines to parallel fiber terminals; and (4) almost complete astroglial investment of Purkinje cell somata and dendrites. Minimal extracellular space was present in the neuropil of the explants and unusual synapses involving Purkinje cells were absent. Atypical features described by some investigators may be a function of retarded development in suboptimal culture conditions and do not represent the limit of tissue culture methodology.