Search

The developmental regulation of tubulin and several of its posttranslational modifications was examined during the differentiation of rat cerebellar granule cells in vitro. In particular, we have noted that the glutamylation of alpha- and beta-tubulin subunits varies during development and becomes more prominent in differentiating neuronal processes. These results indicate that glutamylation of tubulin may be important in the stabilization of microtubules during the maturation of the neuronal cytoskeleton.

T alpha 1 and T26 alpha-tubulin mRNA expression was examined during the differentiation of rat cerebellar granule cells in vitro and in situ. High levels of T alpha 1 transcript correlated with neurons extending processes and hence may implicate T alpha 1 with neuritogenesis. In comparison, T26 labeling was much less prominent, appeared more constitutive and was possibly associated with cell proliferation. Such profiles indicate that the different isotypes play different roles in the assembly and function of microtubules during neuronal differentiation.

Microtubule-associated protein 2 (MAP2) is a major component of the neuronal cytoskeleton and is known to promote the assembly and stabilization of microtubules, functions which have important implications in neuronal differentiation. MAP2 consists of high molecular weight (HMW) proteins MAP2a, MAP2b and a low molecular weight (LMW) isoform MAP2c which are produced from a single gene by alternative splicing. In this study, we describe the expression of the various MAP2 mRNA isotypes and protein isoforms during the development of rat cerebellar granule cell neurons over a 21-day period in vitro. In situ hybridization was used to detect MAP2 mRNA isotypes which corresponded to HMW- and LMW-MAP2 proteins. The distribution of MAP2 mRNAs in the developing P7 cerebellar cortex was related to the different stages of granule neuron development in situ. During early stages of neuronal differentiation in vitro, high levels of MAP2c mRNA were observed which gradually decreased as development progressed. Throughout the period studied, MAP2ab mRNA concentrations remained low although a small transient rise was noted during the first 14 days in vitro (div). The profile of MAP2 protein variants showed further developmental regulation. The expression of the LMW-MAP2c isoform closely mirrored that of its mRNA whilst HMW-MAP2b protein concentrations rose during the first 10 div and were maintained in older cultures. HMW-MAP2a appeared after 4 div and gradually increased throughout the remainder of the study. Clearly, the outline of HMW-MAP2 protein did not relate to its encoding mRNA and such disparity may be due to the operation of different transcriptional and/or posttranslational mechanisms. Immunocytochemical analyses of MAP2 variants provided further information concerning their localization during neurite outgrowth. These results describe the developmental regulation of MAP2 mRNA and protein variants and that the profile of their expression relates to the formation of processes during the differentiation of granule neurons in vitro.

Tau microtubule-associated proteins constitute a group of developmentally regulated neuronal proteins which promote microtubule polymerization and stabilization and hence have important implications during neuronal morphogenesis. We have examined the expression of tau mRNA and protein levels during the differentiation of cerebellar granule neurons over a period of 3 weeks in vitro. Oligonucleotide probes directed towards either immature or mature forms of tau mRNA were detected by in situ hybridization. Such experiments demonstrated that the time interval between 1 and 4 days in vitro represents a developmental epoch in the regulation of tau mRNA whereby the dominant immature tau messages were gradually replaced by mature mRNAs. Analysis of the profile of the various tau isoforms showed further developmental regulation with the transient rise in immature tau variants followed by the appearance of mature isoforms in older cultures. The increase in tau heterogeneity during granule neuron differentiation was enhanced by and could be attributed to intensive post-translational phosphorylation. Dephosphorylation of cell cultures demonstrated that the majority of tau was phosphorylated and that such a modification had profound affects on the localization of tau within developing neurons by immunocytochemistry. This study describes the profile of tau protein and mRNA levels expressed by differentiating cerebellar granule neurons in vitro and clearly demonstrates that tau is developmentally regulated and that important changes in tau expression occur at a time when processes are consolidating their first contacts.

GAP-43 is a growth-associated protein that has been implicated in the developmental outgrowth of axons. We have examined the profile of GAP-43 levels in rat cerebellar granule cells during their development in vitro. During the first 1-2 days after plating, the majority of cells expressed neurites and after 8 days a complex neuronal network had developed. In situ hybridization studies showed that GAP-43 mRNA levels rapidly increased to peak at 1-2 days and gradually returned to initial values after 7-8 days. Analysis of GAP-43 protein levels followed a similar transient profile. Initially, granule cell perikarya and structures associated with neuritogenesis all displayed GAP-43 immunoreactivity. In older cultures, perikaryal labelling was lost after 10 days whilst process staining decreased more gradually. During the first 48 hours detailed analysis of GAP-43 mRNA revealed two populations of granule cells. It was suggested that cells with significant label originated from the external germinal layer which displays much GAP-43 mRNA in cerebellar sections. Cells with little or no GAP-43, however, probably originated from the internal granular layer since this region displayed no specific labelling. Granule cells within clumps expressed more GAP-43 mRNA compared to isolated cells perhaps indicating cell-cell regulation of expression. These results describe the transient rise in GAP-43 protein and mRNA levels expressed by developing cerebellar granule cell neurons in vitro and provide further evidence for the role GAP-43 plays during neuritogenesis.

Short-term (up to 5 h post-plating) cerebellar granule cell cultures were prepared from the week-old rat and maintained in a micro-incubator during time lapse video microscopy to examine normal and N-methyl-D-aspartate (NMDA)-evoked neurite extension. In untreated cultures growth of neurites was stochastic but proceeded at an average rate of 12.0 +/- 1.4 microns/h. Growth cone morphology was variable. The classical filopodia and lamellipodia possessing tips were motile or non-motile, while those processes ending in a club shape were rarely seen to extend. Individual growth cones passed through several different morphologies during growth. New processes extended more rapidly (13.0 +/- 1.7 microns/h) than those already present (9.0 +/- 0.5 microns/h). Addition of the NMDA receptor antagonist, aminophosphonovalerate (APV), caused a marked retraction of pre-existing processes. Stimulation of the receptor with 50 microM NMDA caused a marked increase in growth rate compared to controls (15.0 microns/h and 1.7 microns/h, respectively). When the presence of actin-rich structures was examined using rhodamine-phalloidin labelling it was found that NMDA increased the proportion of neuronal processes that possessed a growth cone by 28%. Conversely, inhibition of NMDA receptor activity with APV reduced the formation of lamellipodia from neuronal cell bodies.

Week-old rat cerebellar granule cells were grown in the presence of the cell-permeable calcium chelating agent BAPTA-acetoxy methyl ester (BAPTA-AM) for the first 8 h in vitro. There was a dose-dependent inhibition of process outgrowth with an IC50 of approximately 5 microM. Neurite outgrowth could be partially recovered by the addition of N-methyl-D-aspartate (NMDA; 50 microM) to BAPTA-AM-treated cells. Phorbol ester stimulation of treated cells evoked a profound inhibition of neuritogenesis compared to a stimulatory effect on control cultures. The inhibition of growth caused by phorbol esters could not be reversed by NMDA co-addition. Neurites extended by BAPTA-AM-treated granule cells were thinner than in control cultures and did not form elaborate growth cones even when growth was stimulated by NMDA. The distribution of tyrosinated and acetylated alpha-tubulin in the processes of BAPTA-AM-treated cells appeared similar to that in controls. However, rhodamine-phalloidin labelling of microfilaments in the cell cultures emphasised the loss of an elaborate actin-rich growth cone in BAPTA-AM-treated cells even when neurite formation was partially recovered. These results indicate the importance of [Ca2+]i in the production of neurites from cerebellar granule cells in vitro.

The present study has examined excitatory amino acid transmitter binding sites as measured autoradiographically in cryostat sections prepared from developing rat cerebella during the period of granule cell neuritogenesis. The external germinal layer (EGL) and molecular layer (ML), which during development contain granule cells at early stages of axon growth, contained only low levels of NMDA-displaceable L-[3H]glutamate binding sites. Similarly, [3H]glycine binding to the NMDA receptor linked binding site was not enriched in the EGL. Radioligand binding to the NMDA receptor was always greater in the granular layer (GL) than in the ML. The developmental increases in NMDA-displaceable L-[3H]glutamate and in [3H]glycine binding to the GL were similar but NMDA displaceable L-[3H]glutamate binding density increased before [3H]glycine binding sites. Glycine increased NMDA-displaceable L-[3H]glutamate binding only in the adult cerebellum. These results suggest that NMDA stimulation of neuritogenesis in granule cell cultures may reflect stimulation of dendritogenesis in the developing glomerulus rather than a stimulation of axon growth in the EGL. Also, NMDA receptors may be present in an immature form during cerebellar development and have different properties to the adult receptor. Binding sites for [3H]kainate and [3H]AMPA were present in both the GL and ML and increased during development. At all times the amount of binding sites for [3H]kainate were highest in the GL whereas those for [3H]alpha-amino-3-hydroxy-5-methylisoxazole-4-propionate were highest in the ML.

The distributions of the non-tyrosinated M alpha 4 alpha-tubulin gene product and its tyrosinated form M alpha 4 + Y were examined in immunoblots and sections of adult rat brain. In cerebellar sections, M alpha 4 and M alpha 4 + Y immunoreactivities were enriched in neurons, anti-M alpha 4 + Y labeling thus being more restricted in distribution than that of the general alpha-tubulin antibody YL1/2. The results indicate that the M alpha 4 gene product does not constitute the large non-tyrosinatable pool of alpha-tubulin in brain.

We have used short-term (8 h) cultures of week-old rat cerebellar granule cells to examine the effects on neuritogenesis of activation and down-regulation of protein kinase C by phorbol esters. We have previously demonstrated that endogenously released glutamate promoted neurite outgrowth in the same system acting via N-methyl-D-aspartate receptors. Low levels (0.1-1 nM) of the phorbol ester 12-O-tetradecanoylphorbol-13-acetate (TPA) evoked increases in the number of granule cells which extended neurites; higher levels (10-250 nM) which caused a down-regulation of total protein kinase C, inhibited outgrowth in a dose-dependent manner. N-Methyl-D-aspartate by itself also stimulated process outgrowth but could not reverse the inhibition evoked by either TPA or the protein kinase C inhibitor sphingosine. Stimulation of protein kinase C with 0.1 nM TPA resulted in a general increase in the incorporation of 32P-labelled inorganic orthophosphate into granule cell polypeptides. The results indicate that the activation of protein kinase C is involved in neuritogenesis in granule cells and are consistent with the idea that N-methyl-D-aspartate receptor activation may exert its effect on neuritogenesis through protein kinase C.

The tyrosine kinase substrate p36 (calpactin heavy chain) is a calcium-dependent membrane- and cytoskeletal-binding protein. Using an affinity-purified antiserum raised against the p36 heavy chain from bovine adrenal medulla, we have examined the cellular distribution of p36 in developing and adult cerebellum. From immunoblotting, the level of p36 in cerebellum was found to decline during development. In dissociated cell cultures of P4 cerebellum, all cell types were labeled by anti-p36. In vibratome sections from cerebella of P10 rats, anti-p36 stained Purkinje cell bodies strongly and all other cell types less strongly, with the exception of cells in the external germinal layer, which were unstained. By 18 days postnatally p36 was present at higher levels in Bergmann glia and astrocytes of the white matter. In sections of adult cerebella, anti-p36 staining was restricted to Bergmann glial processes and to the processes of a subclass of astrocytes in the granular layer and the white matter. At no developmental stage was anti-p36 staining detectable in axons or dendrites in vibratome sections. These results suggest that p36 plays a role in some aspect of cellular differentiation common to all cerebellar cell types and may have additional functions in astrocytes of the adult cerebellum.

The effect of endogenous glutamate on neurite outgrowth from cerebellar granule cells in culture was examined. Neurite outgrowth was inhibited by enzymatic removal of endogenous glutamate from the culture medium. The broad-spectrum glutamate receptor antagonist kynurenate also inhibited neurite outgrowth from granule cells in serum-containing and serum-free cultures; the inhibition by kynurenate was reversed by exogenous glutamate. Neurite outgrowth was inhibited to the same extent by the NMDA receptor antagonist APV. These results indicate that endogenous glutamate, possibly released by granule cells themselves, stimulated neurite outgrowth through activation of the NMDA class of glutamate receptors. Activation of NMDA receptors on developing neurons may be an important mechanism for the regulation of neuronal growth and differentiation.

Cerebellar granule cells derived from one-week-old rat pups have been placed into culture to examine the expression of microfilament, microtubular and neurofilament proteins in developing neurites up to 2 days in vitro. Immunofluorescence and rhodamine-phalloidin studies have shown that the first cytoskeletal element to appear in the developing neurite was the microfilament network which was present in cell processes and expansions from at least 8 minutes in vitro. Rhodamine-phalloidin fluorescence was maintained in the developing neurite over 2 days and was enriched in some growth cones. At between 30 and 60 min in culture microtubules (reacting with anti alpha-tubulin antibodies) appeared in the neurites but did not ramify throughout the growth cone at this or later time points. The expression of tyrosylated alpha-tubulin in microtubules of the developing neurite appeared to precede that of acetylated alpha-tubulin. Only one of the neurofilament subunits, the 200 kDa subunit, could be positively identified in granule cell processes in these cultures. The developmental pattern of cytoskeletal protein expression in granule cell neurites in vitro may reflect a process of stabilisation of the growing neurite behind an active, motile growth cone.

We describe the structure and expression of a mammalian beta-tubulin isotype (M beta 6) that is weakly expressed in testis but is abundant in developing brain, with transcripts declining to lower levels in the adult brain. The expression of M beta 6 was undetectable in any other mouse tissue examined. A serum specific for this isotype was prepared using a cloned fusion protein as immunogen. M beta 6 is one of five known beta-tubulin isotypes expressed in brain, and using the anti-M beta 6 serum along with sera, anti-M beta 2, anti-M beta 3/4 and anti-M beta 5, previously characterized, we have examined the pattern of expression of beta-tubulin isotypes in rat cerebellum. The isotypes each have characteristic cell-type specific patterns of localization in cerebellum. M beta 2, M beta 3/4 and M beta 5 are present in both neuronal and non-neuronal cells, but in contrast M beta 6 was only detectable in neurons in tissue sections and in dissociated cerebellar cell culture. The majority of sequence differences among the beta-tubulin isotypes lie at the carboxy terminus, the region of beta-tubulin involved in MAP binding. In the case of M beta 2 and M beta 6, the patterns of expression are similar or identical to the patterns of expression of MAP3 and MAP1A respectively. These results suggest that beta-tubulin isotypes may contribute to the determination of the specific association of MAPs with microtubules of diverse function. However, the strict subcellular segregation of other MAPs in brain may be determined by other factors.

The microtubule-associated proteins 1A, 1B and 2 are present at high levels in Purkinje cell dendrites of normal adult rat cerebellum but show characteristic changes in localization during cerebellar development that allow examination of the effects of hypothyroidism on the development of both Purkinje cells and granule cells. Neonatal rats were made hypothyroid by treatment with propylthiouracil from the day of birth (post-natal day 0, P0). The expression of the microtubule-associated proteins 1A, 1B and 2 in the cerebellum of hypothyroid animals was examined using immunocytochemical techniques and compared to the normal developmental pattern in control animals. The normal developmental decrease in microtubule-associated protein 1A and 1B levels in parallel fibres was delayed in the cerebellum of hypothyroid animals and these proteins persisted in parallel fibres until after P20. Microtubule-associated protein 1B but not 1A was still present in parallel fibres in less mature folia at P30 in hypothyroid rats suggesting that the expression of these two microtubule-associated proteins is regulated separately. In the molecular layer staining with anti-microtubule-associated protein 2 was enriched in Purkinje cell dendrites in normal and hypothyroid cerebella and the stained Purkinje cell dendrites in hypothyroid cerebellum demonstrated a typical deformed morphology at P15. The results show that the restricted subcellular localization of these microtubule-associated proteins is maintained in the cerebellum of hypothyroid rats but the developmental changes in their expression are delayed.

We have recently shown that acetylated alpha-tubulin containing microtubules (acetyl-MTs; labeled by antibody 6-11B-1) constitute a cold-stable subset of the microtubule network of nonneuronal cells in rat primary forebrain cultures [Cambray-Deakin and Burgoyne: Cell Motil. 8(3):284-291, 1987b]. In contrast, tyrosinated alpha-tubulin containing MTs (tyr-MTs; labeled by antibody YL1/2) are cold-labile. Here we have examined the distribution of acetyl-MTs and tyr-MTs in cultures of newborn rat forebrain astrocytes and simultaneously investigated the distribution of mitochondria and glial filaments. In double-label immunofluorescence experiments a marked colocalisation of acetyl-MTs and glial filament bundles was observed. Tyr-MTs did not show a similar colocalisation with glial filament bundles. Furthermore, the distribution of mitochondria closely followed that of the acetyl-MT and glial filament bundles. When cells were exposed to short-term (30-min) treatments with MT-disrupting agents such as colchicine and nocodazole, the tyr-MT network was removed but the distributions of acetyl-MTs, glial filaments, and mitochondria were unchanged. Increased exposure to colchicine (9-16 hr) caused a progressive disruption of the acetyl-MTs and the collapse of glial filaments and mitochondria to the perinuclear region. These results suggest that acetyl-MTs and glial filaments but not tyr-MTs may be involved in the intracellular transport of organelles and/or in the control of their cytoplasmic distribution.

The distribution of acetylated alpha-tubulin in rat cerebellum was examined and compared with that of total alpha-tubulin and tyrosinated alpha-tubulin. From immunoperoxidase-stained vibratome sections of rat cerebellum it was found that acetylated alpha-tubulin, detectable with monoclonal 6-11B-1, was preferentially enriched in axons compared with dendrites. Parallel fiber axons, in particular, were labeled with 6-11B-1 yet unstained by an antibody recognizing tyrosinated alpha-tubulin, indicating that parallel fibers contain alpha-tubulin that is acetylated and detyrosinated. Axonal microtubules are known to be highly stable and the distribution of acetylated alpha-tubulin in other classes of stable microtubules suggests that acetylation and possibly detyrosination may play a role in the maintenance of stable populations of microtubules.

We have examined the distribution of acetylated alpha-tubulin using immunofluorescence microscopy in fibroblastic cells of rat brain meninges. Meningeal fibroblasts showed heterogeneous staining patterns with a monoclonal antibody against acetylated alpha-tubulin ranging from staining of primary cilia or microtubule-organising centers (MTOCs) alone to extensive microtubule networks. Staining with a broad spectrum anti-alpha-tubulin monoclonal indicated that all cells possessed cytoplasmic microtubule networks. From double-labeling experiments using an antibody against acetylated alpha-tubulin (6-11B-1) and antibodies against either tyrosinated or detyrosinated alpha-tubulin, it was found that acetylated alpha-tubulin and tyrosinated alpha-tubulin were often segregated to different microtubules. The microtubules containing acetylated but not tyrosinated alpha-tubulin were cold stable. Therefore, it appeared that in general meningeal cells possessed two subset of microtubules: One subset contained detyrosinated and acetylated alpha-tubulin and was cold stable, and the other contained tyrosinated alpha-tubulin and was cold labile. These results are consistent with the idea that acetylation and detyrosination of alpha-tubulin are involved in the specification of stable microtubules.

The expression of the microtubule-associated protein MAP1A and the synaptic vesicle protein p65 by rat cerebellar granule cells developing in vivo and in vitro was examined. MAP1A in common with a series of previously described cytoskeletal proteins was present in immature parallel fibres and began to disappear from parallel fibres around P15. From the earliest age (P6) at which the molecular layer could be identified parallel fibres throughout the molecular layer expressed p65 with no evidence of any gradient of expression from young to mature axons. Granule cells in culture did not express p65 until later than 1 day in vitro. By 7 days in vitro numerous p65 containing varicosities were present on granule cell processes though MAP1A was still expressed. The results show that transport of synaptic vesicles and formation of varicosities is an early event in granule cell differentiation and precedes a major reorganisation of the axonal cytoskeleton.

Cerebellar granule cells in vivo and in vitro have been widely used in the study of the cellular neurobiology of neuronal development. We have described the basic neuroanatomical data on the granule cell in the developing and mature cerebellum. The importance of the cytoskeleton in determining the morphology of the granule cell and in process outgrowth and cell migration has been described. Extensive information is now available on the composition of the granule cell cytoskeleton. Cell surface glycoproteins are thought to be involved in the control of cell adhesion and cellular interactions during development. A number of surface molecules belonging to either the N-CAM or the Ng-CAM groups of glycoproteins have been studied in detail in the cerebellum. The role of these proteins in cell adhesion and in granule cell-astroglial interactions during granule cell migration has been reviewed. The survival and differentiation of neurones is controlled by soluble trophic factors. Several factors have been described which act as trophic factors for granule cells in vitro and may do the same in vivo. The numerous studies that have been carried out on the cerebellar granule cell have allowed us to describe certain aspects of the cellular neurobiology of this class of neurones as an example with general significance for the understanding of neuronal differentiation and function.

The expression of the 200-kDa neurofilament subunit in developing rat cerebellar granule cells in vivo and in vitro was examined by immunocytochemistry and immunoblotting. Granule cells in vitro were found to express neurofilament-like immunoreactivity. In cryostat sections, neurofilament immunoreactivity was evident between P6 and P13 as a band of staining in the molecular layer indicative of staining of parallel fibres. No labelling of parallel fibres was seen at P18 or in adult animals.