Ultrastructural studies on cerebellar histogenesis in the frog: the external granular layer and the molecular layer

Ontogeny of proenkephalin mRNA and enkephalin peptide expression in the cerebellar cortex of the rat: spatial and temporal patterns of expression follow maturational gradients in the external granular layer and in Purkinje cells

Proenkephalin mRNA and peptide products were examined in developing cells of the postnatal rat cerebellar cortex using in situ hybridization and immunocytochemistry. On day 7, proenkephalin mRNA was first detected as discrete cellular labeling in Golgi cells and as a diffuse hybridization signal over the Purkinje cell layer. On day 14, proenkephalin mRNA and peptide products primarily appeared in distinct subpopulations of Purkinje cells present in the posterior and lateral cerebellum. Similarly, in the external granular layer (EGL), enkephalin immunoreactivity was present only in the posterior and lateral portions of the cerebellum on day 14. However, proenkephalin mRNA was not detected in enkephalin-immunoreactive EGL cells. On day 21, the subset of Purkinje cells that expressed proenkephalin mRNA and peptides were distributed more uniformly throughout the cerebellum. On day 28, a few enkephalin-immunoreactive Purkinje cells were uniformly present throughout the cerebellum, but proenkephalin mRNA was not detected in most of these cells. The spatial gradients in proenkephalin mRNA expression evident in the Purkinje cells of younger rats were no longer present in 28-day-old rats. These findings are important, because endogenous opioids such as enkephalin have been previously shown to inhibit the growth of Purkinje cell dendrites and dendritic spines, and inhibit the rate of mitosis in EGL neuroblasts. Cells do not develop at uniform rates within the cerebellum. There are regional differences in the timing of the formation of the EGL, and in the morphogenesis of Purkinje cells. In conjunction with previous work, the present findings suggest that during development, the pattern of enkephalin immunoreactivity in Purkinje and EGL cells closely follows the spatial and temporal gradients of maturation in both these cell types. The emergence and disappearance of enkephalin immunoreactivity in Purkinje and EGL cells is spatially and temporally related, and coincides with proenkephalin mRNA expression in Purkinje cells. Thus, the transient and coordinated appearance of enkephalin in cerebellar Purkinje and EGL cells may contribute to regional differences in the rate of cerebellar maturation, and may help synchronize the developmental interactions between these two cell types.

Autoradiographic studies of cerebellar histogenesis in the premetamorphic bullfrog tadpole: II. Formation of the interauricular granular band

This study examines the origin of cells in the interauricular granular band (iagb) in the cerebellum of the frog tadpole during early stages of development by means of histological and autoradiographic methods. Premetamorphic bullfrog tadpoles were exposed to multiple doses of 3H-thymidine (10 microCi/g body weight per exposure) at developmental stages ranging from 1 week to 1 year and were killed at either 6 or 12 months of age. The autoradiographic data were examined to determine the time when cells of the iagb were generated. Our findings show that initial generation of iagb cells begins at week 3 and that a peak in the formation of postmitotic neurons is reached at the age of 10 weeks. This is followed by other peaks of cell generation at the ages of 16 weeks, 10 months, and 11.5 months. The generation cycles of iagb cells are interrupted by periods of quiescence when label cannot be detected in any of the cells. These quiescent periods occur at the ages of 20-26 weeks, 7 months, and 12 months. These findings indicate that cells of the iagb are generated in a cyclical manner over the entire 1-year period which was studied. Comparison of our present data on iagb cell formation with the generation of cells in the EGL shows that the production of these two groups of cells is overlapping, but cells of the iagb begin and cease production before those of the EGL. On the basis of our findings we propose that the cells of the iagb and the EGL belong in separate cell groups which are generated by distinct subpopulations of germinal cells in the neuroepithelial cap.

Granule cell development in the frog cerebellum during spontaneous and thyroxine-induced metamorphosis

Granule cell maturation in the cerebellum of bullfrog tadpoles was studied during both spontaneous and thyroxine-induced metamorphosis by using electron microscopy and Golgi-impregnated preparations. The production of cerebellar microneurons, a majority of which are granule cell precursors, was quantitatively compared during spontaneous and thyroxine-induced metamorphosis by using stereological methods and biochemical measurements of DNA. Granule cell migration and differentiation appeared morphologically similar during spontaneous and thyroxine-induced metamorphosis. In both instances, granule cells migrated tangentially along the pial surface, migrated into the internal granular layer, developed dendritic arbors, and formed synaptic contacts with the processes of Golgi cells and with mossy fibers. These events are similar to developmental processes that have been described in detail in other animals. Quantitative stereological measurements demonstrated similar overall patterns of change during spontaneous and thyroxine-induced metamorphosis. Most notably, increases in the volume of the external granule layer correlated with increases in the relative and total amounts of DNA. However, measurements of total DNA were consistently reduced during the period of accelerated change that occurs in thyroxine-induced metamorphosis, although external granular layer volume was greater in these tadpoles after 2 and 3 weeks of thyroxine treatment than in spontaneously metamorphosing tadpoles. While granule cell development in the frog is largely dependent on thyroid hormone, differences between thyroid-hormone-induced and spontaneously metamorphosing tadpoles suggest that normal patterns of cerebellar development are also dependent on events that occur in premetamorphic tadpoles in the absence of thyroid hormone.

Stellate cell development in the frog cerebellum during spontaneous and thyroxine-induced metamorphosis

Stellate cell development was studied in the bullfrog cerebellum during spontaneous and thyroxine-induced metamorphosis using the Golgi-Kopsch method and electron microscopy. Cells that possessed axosomatic synapses and resembled stellate cells were present even in the incipient molecular layer of the cerebellum in the premetamorphic tadpole. These cells may have originated from the early, transient wave of external granule cells that have been reported in the cerebellum of premetamorphic tadpoles in the first 6 months of development, and may constitute the variant population of stellate cells that are present later during development or the degenerating cells that have been observed during metamorphosis as scattered dying cells in the molecular layer. Typical stellate cells, whose development resembled the genesis and differentiation of stellate cells in birds and mammals, were initially observed at the outer border of the molecular layer that is adjacent to the external granular layer during the onset of metamorphosis. These stellate cells were bipolar with processes extending in a plane perpendicular to elongating parallel fibers, and with progressive development, became multipolar with dendrites oriented in various directions with respect to the pia. Stellate cell axons innervate the dendrites and somata of Purkinje cells and other stellate cells, and can be categorized into two types: (1) axons with extensive branching near the soma of origin, and (2) long axons with few branches that occasionally terminate in the Purkinje cell layer. Atypical neurons that did not resemble typical stellate cells were also present in the molecular layer; these might be classified as a stellate cell variant. The generation and differentiation of stellate cells can be induced 1 to 2 years prematurely by administering thyroid hormone to premetamorphic tadpoles. Like most events of cerebellar histogenesis in the frog, stellate cell development also appears to be largely a thyroid-dependent phenomenon.

Abnormal visual cortex development in the kitten associated with maternal dietary taurine deprivation

We have examined the visual cortex of newborn and 8-week-old kittens born to mothers consuming a taurine-supplemented or taurine-free diet using the rapid Golgi technique. Kittens from taurine-supplemented mothers exhibited normal development of the visual cortex. Kittens from taurine-depleted mothers showed striking differences. In newborn kittens, neuroblasts are aggregated both at the ventricular and pial zones, having failed to migrate and differentiate normally. Eight weeks after birth, only few pyramidal and nonpyramidal neurons are found. Those present have heavily spined dendritic processes indicative of poor arborization. Protoplasmic astrocytes are represented by undifferentiated cellular masses. The taurine concentration in the cortex of such kittens is four- to fivefold smaller than in kittens from taurine-supplemented mothers. These results suggest that normal concentrations of taurine in visual cortex, as well as cerebellum, are required for normal ontogeny of neurons. Once such deficits are established in the prenatal and immediate postnatal period, they result in permanent abnormalities. These findings have clear implications for vegetarian women who intend to have children, since virtually no taurine is present in plants and vegetables.

Cerebellar external granule cells are attached to the basal lamina from the onset of migration up to the end of their proliferative activity

The development of cerebellar external granule cells in rats was studied from the time of demarcation of the cerebellar anlage on embryonal day 12 up to the time of their disappearance on postnatal day 20. Two types of cells were found. The first was orientated tangentially to the cerebellar surface and was characterized by a persistent contact to the basal lamina via an external process, with a lamellopodial tip and a cytoskeleton characteristic for migratory cells, and a retracting internal process featuring a single cilium. This cell type was the first to appear on embryonal day 14 in the caudolateral angle of the cerebellar anlage and, later, spread over the whole cerebellar surface. It disappeared after the external granular layer was completely expanded over the cerebellum. The second cell type appeared for the first time on embryonal day 16 in the caudal part of the cerebellar anlage and disappeared on postnatal day 20. It was orientated radially and also had contact with the basal lamina either with its cell body or with one or two short, radial processes, whose morphology differed from that of the external process of tangential cells by the absence of a lamellopodium and a prominent cytoskeleton. After postnatal day 17 contacts of external granule cells with the basal lamina decreased rapidly in length and number and were absent on postnatal day 20. We interpret these findings to indicate that tangential external granule cells are migrating before taking on a radial orientation characteristic for the mitotic cycle of proliferating external granule cells. In the light of increasing evidence implicating extracellular matrix in various developmental events of the nervous system we propose that the basal lamina of the cerebellum may be used as substrate and guidance structure by migrating external granule cells, and, furthermore, that the persistent contact with the basal lamina may mediate stimuli maintaining external granule cells in a proliferative state.

Derivation of cerebellar Golgi neurons from the external granular layer: evidence from explantation of external granule cells in vivo

The present report provides evidence to challenge the traditional view that cerebellar Golgi cells are derived from the ventricular neuroepithelium, postulating instead that they originate from external granule cells. Supporting evidence for this assertion comes from three sources: 1) Typical Golgi cells are found in ectopic granule cell colonies, both outside the cerebellum (in the subarachnoid space) and also within the cerebellar cortex between fused folia. Because ectopic granule cell colonies are derived from external granule cells, which become displaced after treatment with 6-hydroxydopamine (6-OHDA), it was assumed that the ectopic Golgi cells also stem from such displaced external granule cells. 2) In order to demonstrate that Golgi cell precursors migrate from the external granular layer into the Purkinje cell plate, the development of the cerebellar cortex was studied over the period of Golgi cell genesis. On E19 the external granular layer in the rat is subdivided into an outer proliferative and an inner subproliferative zone. At the inner margin of the external granular layer, and in the marginal zone, radially oriented, darkly staining cells are present that exhibit all the characteristics of migrating neurons possessing a leading process oriented toward the Purkinje cell plate, a somatic cilium, and a close association with radial glia fibers. In later stages, these cells are also found deep to the Purkinje cell plate. Because Golgi cells arise during the period between E19 and postnatal day 2 in the rat (Altman and Bayer, '77, '78) and as the basket cells, the first neurons of proven origin from the external granular layer, are not produced before the second postnatal day (Altman, '72), the earlier migrating neurons are presumed to be Golgi cells. 3) Available data from cell kinetic 3H-thymidine studies show that there is no unequivocal evidence for Golgi cell genesis from the ventricular neuroepithelium, because, at the time of Golgi cell birth, ventricular and external granular stem cell populations are proliferating, and with the present methods it is not possible to decide which of these are the precursors of Golgi cells. Thus, taken together, the findings of this study show that Golgi cells are more likely to arise from the external granular layer than from the ventricular neuroepithelium. This concept would unify cerebellar histogenesis by proposing that projection neurons arise from the ventricular neuroepithelium, whereas all interneurons of the cerebellar cortex are descendants of the external granular layer.

Purkinje cell maturation in the frog cerebellum during thyroxine-induced metamorphosis

Purkinje cell maturation during thyroxine-induced metamorphosis in premetamorphic bullfrog tadpoles was studied using electron microscopy and Golgi (silver-impregnated) preparations. Cerebella from tadpoles were examined following 1, 2, or 3 weeks of thyroxine treatment. Particular attention was paid to possible differences between the two populations of Purkinje cells previously described, i.e. (i) the smaller population located in the dorsal part of the cerebellum, where the Purkinje cells show dendritic arborization long before the appearance of the external granular layer, and (ii) the larger population located in the middle and ventral regions of the cerebellum, where the Purkinje cells begin to undergo maturation during metamorphosis when the external granular layer is established. Following thyroxine treatment, both populations of Purkinje cells showed rapid maturational change. In the mature (dorsal) group, dendritic growth resumed in the presence of an external granular layer increasing the complexity of their dendritic arbors. Moreover, climbing fiber synapses translocated from contacts on the soma to the thorns of growing dendrites, and somatic processes often disappeared. The immature (ventral) group showed dramatic differentiation of the perikaryon including polarization of cytoplasm with subsequent dendritic outgrowth and formation of somatic processes in the presence of climbing fibers. Stellate cell contacts appeared on the smooth portion of the soma of many Purkinje cells. Dendritic growth during thyroxine-induced metamorphosis was characterized by growth (elongation) with minimal branching, which is initially observed during spontaneous metamorphosis. Typically, these growing dendrites ended in growth cones, some with one or several filopodia. Developing Purkinje cell dendritic spines formed synapses with parallel fibers. The present study has provided an example of the dramatic nature of thyroxine's action in inducing the complex series of detailed maturational changes in the cerebellum 1-2 yr ahead of schedule. In addition, the results show that thyroxine-induced Purkinje cell maturation is more rapid and synchronous than that seen during spontaneous metamorphosis. It is concluded that Purkinje cell maturation during metamorphosis is largely dependent on thyroid hormone.

Arguments in favour of endocytosis of glycoprotein components of the membranes of parallel fibers by Purkinje cells during the development of the rat cerebellum

Chloroquine (a drug known to induce a dysfunction of lysosomes) was used to study the behavior of Concanavalin A binding glycoproteins located on the axolemma of parallel fibers in young rat cerebella, and abundant on these membranes at a period preceding synaptogenesis with the dendrites of Purkinje cells. Chloroquine induces in Purkinje cells a large accumulation of grains consisting of membrane whorls in lysosomes. These grains stain for Concanavalin A, and do not stain either for a mitochondrial marker (aspartate aminotransferase mitochondrial isoenzyme) or for a marker of the Purkinje cell internal membrane (PSG). It is suggested that the material accumulating in the Purkinje cells under the effect of chloroquine comes from the parallel fibers. Together with the observation that alpha-D-mannosidase (involved in the degradation of these glycoproteins) is exclusively located inside Purkinje cells, these results provide a firm indication that this material enters the Purkinje cells through pinocytosis. The absence of ATPase activity (ATPase is a glycoprotein plasma membrane marker highly concentrated on parallel fibers) within these grains suggested that not all the components of these membranes are pinocytosed, but that the process is specific for certain molecules. These results are compatible with the ultrastructural observations of others, and support the arguments in favour of the pinocytosis phenomenon being one of the first steps of synapse formation. The observed specificity of pinocytosis for certain molecules suggests that a receptor-mediated recognition of some glycans of glycoproteins is the preliminary event in the establishment of synapses.

Developmental relationships between trigeminal ganglia and trigeminal motoneurons in chick embryos. I. Ganglion development is necessary for motoneuron migration

The migration and early development of trigeminal (V) motoneurons were studied in chick embryos in which two different populations of primary trigeminal sensory neurons had been removed prior to the birthdate of the V motoneurons. Ablation of mesencephalic neural crest cells, which eliminates monosynaptic sensory input, did not affect the migration, early development, or later differentiation of the V motoneurons. However, when the anlagen of the V ganglion were removed, the V motor root did not exit from the brainstem and the V motor nucleus did not develop. Although the neurons of the V ganglion do not innervate adult V motoneurons, these populations are related developmentally. In those embryos in which the V ganglion did not develop, medial column cells, which are midline, postmitotic, premigratory V motoneurons, and a few medial, elongated cells (possibly migratory) were present until days 5-6, but these cells did not complete their lateral migration to form the lateral nucleus of V. In cases where the ganglion anlagen were not completely removed, the number of postmigratory V motoneurons was positively correlated to the size of the ganglion remnant. There also was a correlation between the axial position of the postmigratory V motoneurons and the ganglion remnants. If a caudal remnant developed, only caudal V motoneurons, whose axons reached the ganglion, migrated; if a rostral remnant developed, only rostral V motoneurons, with axons reaching this remnant, migrated. Additionally, if the central axons of the ganglion remnant entered the metencephalon in either dorsal or ventral ectopic positions, the V motor nucleus was located in a corresponding aberrant position. Thus, some characteristic of the V ganglion cells appears to guide the motor axons and somas to their final brainstem position.

Ultrastructural studies on the ventricular surface of the frog cerebellum

Ultrastructural studies of the ventricular surface of the frog cerebellum showed regional differences. In the midline region of the adult cerebellum was found a band of profusely ciliated squamous ependymal cells. In the rest of the cerebellum the ependymal cells were columnar and each had a single cilium. In the cerebellum of the premetamorphic tadpole, the squamous ependymal cells of the midline region also were monociliated. During metamorphosis they gradually became multiciliated. Additionally, supraependymal cells and synaptic elements were present on the ventricular surface of the cerebellum of adult frogs as well as in late metamorphic tadpoles. In contrast, supraependymal cells were rarely observed in premetamorphic tadpoles, and it was concluded that the supraependymal system develops during metamorphosis. It is postulated that the band of cilia may be associated with the circulation of cerebrospinal fluid, and supraependymal synaptic elements function in neuroendocrine regulation.

The mode of migration of neurons to the hippocampus: a Golgi and electron microscopic analysis in foetal rhesus monkey

The mode of neuron migration from the site of their origin in the ventricular zone to area CA1 of the hippocampus was analysed with Golgi and electron microscopic methods during the first half of gestation in the foetal rhesus monkey. In the inner portion of the intermediate zone, the migrating cells have a bipolar form with one, or oaccasionally two, leading processes which do not reach the ammonic plate and with a single trailing process which usually ends within the intermediate zone. Both the nucleus and the cytoplasm of the migrating cells are relatively electron-dense and the latter contains organelles typical of young neurons as described in other brain regions. Analysis of electron micrographs from serial sections reveals that the length of the somata and of the leading and trailing processes of the migrating neurons is apposed to fascicles of radially oriented, electron-lucent, microtubule-filled fibres which are ultrastructurally similar to the radial glial fibres of the neocortex and to the Bergmann glial fibres of the cerebellum. The close (20 nm) apposition between the membranes of the migrating cell and the radial fibre is maintained even in areas where the fibres bend or curve tortuously. Migrating neurons situated at progressively more superficial levels of the intermediate zone become progressively more differentiated and complex. Thus, in the outer portion of the intermediate zone, the migrating cells acquire several additional cytoplasmic processes and occasionally a long thin axon-like process which courses into the incipient alveus. These cells have somewhat larger somata and less electron-dense nuclei and cytoplasm than the migrating neurons still situated in the inner part of the intermediate zone. Cells close to the ammonic plate usually have one to three cytoplasmic processes that enter the ammonic plate and terminate near their presumed final position. Migrating neurons situated at the lower border of the ammonic plate have a single large apical process which intermingles with neurons already in their final position and which sometimes traverses the ammonic plate. The apposition of the migrating neurons to the radial glial processes becomes less explicit as the cell soma enters the ammonic plate, reflecting the more complex three-dimensional intercellular relationships. However, the present analysis indicates that during the middle and late stages of neuronal migration to the hippocampus radial glial fibres may guide postmitotic young neurons across the intermediate zone to the ammonic plate in the same way that they guide neurons migrating to the superficial and middle layers of the neocortical plate.