Induction of neuron-specific tropomyosin mRNAs by nerve growth factor is dependent on morphological differentiation.

We have examined the expression of brain-specific tropomyosins during neuronal differentiation. Both TmBr-1 and TmBr-3 were shown to be neuron specific. TmBr-1 and TmBr-3 mRNA levels increased during the most active phase of neurite outgrowth in the developing rat cerebellum. In PC12 cells stimulated by nerve growth factor (NGF) to differentiate to the neuronal phenotype, TmBr-1 and TmBr-3 levels increased with an increasing degree of morphological differentiation. Induction of TmBr-1 and TmBr-3 expression only occurred under conditions where PC12 cells were permitted to extend neurites. NGF was unable to maintain levels of TmBr-1 and TmBr-3 with the loss of neuronal phenotype by resuspension of differentiated PC12 cells. The unique cellular expression and regulation in vivo and in vitro of TmBr-1 and TmBr-3 strongly suggests a critical role of these tropomyosins in neuronal microfilament function. The findings reveal that the induction and maintenance of the neuronal tropomyosins is dependent on morphological differentiation and the maintenance of the neuronal phenotype.

Differential induction and intracellular localization of SCG10 messenger RNA is associated with neuronal differentiation.

The differentiation of neurons involves the establishment of distinct molecular compartments which regulate neuronal shape and function. This requires targeting of specific gene products to growth-associated regions of the neuron. We have investigated the temporal and spatial regulation of SCG10 gene expression during neuronal differentiation. There are two SCG10 messenger RNAs, 1 and 2 kg in length, which encode the same growth-associated protein. These messenger RNAs were found to be differentially regulated during the onset of neurite outgrowth in early rat cerebellum development. In PC12 cells, the two SCG10 messenger RNAs were shown to be differentially induced by nerve growth factor. Regulation of the 2 kb messenger RNA, but not the 1 kb messenger RNA, is dependent on the differentiation of PC12 cells, indicating that post-transcriptional regulation of SCG10 expression during neurite outgrowth. Spatial regulation of the 2 kb SCG10 messenger RNA distribution during brain development was examined by in situ hybridization. The 2 kb messenger RNA was found to be localized to the neuronal pole where outgrowth was occurring, within differentiating neurons in vivo. Intracellular localization of SCG10 messenger RNA was also observed in differentiating primary cultured neurons, with the 2 kb messenger RNA transported into growing neurites during the development of neuronal polarity. In neurons which had developed polarity, the 2 kb SCG10 messenger RNA was consistently found in the cell body and axon. This study demonstrates both temporal and spatial post-transcriptional regulation of SCG10 expression which is associated with neurite outgrowth. The directed transport and positional translation of SCG10 messenger RNA provide a potential mechanism for protein targeting and the creation of molecular compartments during neuronal differentiation.

AvGp50, a predominantly axonally expressed glycoprotein, is a member of the IgLON's subfamily of cell adhesion molecules (CAMs).

We have previously reported a 50 kDa glycoprotein (AvGp50) expressed specifically in the chick nervous system [Hancox, K.A., Sheppard, A.M. and Jeffrey, P.L., Characterisation of a novel glycoprotein (AVGP50) in the avian nervous system, with a monoclonal antibody, Dev. Brain Res., 70 (1992) 25-37], and we present its molecular characterization. A PCR fragment was generated following sequencing of peptide and N-terminal fragments derived from purified AvGp50. A 1.58 kb clone (pUEX762) containing the 5'-UTR, the entire coding sequence and a short 3'-UTR was then isolated from a chick embryonic day 18 forebrain library. The deduced amino acid sequence encodes a 338 amino acid peptide containing a 31 amino acid signal peptide at the N-terminal and a 19 amino acid phosphatidylinositol glycan linkage sequence at the C-terminal. The mature protein contains three C2-immunoglobulin-like domains and a glycosyl phosphatidylinositol anchor and shares significant homology to other members of the immunoglobulin superfamily, including neural cell adhesion molecule (N-CAM), myelin-associated glycoprotein (MAG) and the Drosophila protein Amalgam. AvGp50 exhibits highest sequence identity to a recently classified subgroup of the immunoglobulin superfamily (IgLONs - immunoglobulin LAMP, OBCAM and neurotrimin - classified by Pimenta et al. [Pimenta, A.F., Zhukareva, V., Barbe, M.F., Reinoso, B.S., Grimley, C., Henzel, W., Fischer, I. and Levitt, P., The limbic system-associated membrane protein is an Ig superfamily member that mediates selective neuronal growth and axon targeting, Neuron, 15 (1995) 287-297], comprising the opioid binding cell adhesion molecule (OBCAM), neurotrimin and the limbic system-associated membrane protein (LAMP) suggesting that AvGp50 is a member of this subgroup. AvGp50 is expressed predominantly on the cell surface of axons, in particular Purkinje cell and granule cell axons in the cerebellum. In cerebellar and forebrain neuronal cultures, protein expression is exclusively located at the cell surface. Despite its cell surface localization, AvGp50 does not directly influence the outgrowth of neurons from explant cultures from ED8 to ED10 chick forebrain, prompting the suggestion that AvGp50 may act in later maturational events.

Molecular cloning and primary structure of the avian Thy-1 glycoprotein.

We have previously characterised, both biochemically and immunohistochemically, a 23 kDa putative avian Thy-1 protein homologue. In this report we have examined the carbohydrates present on the protein and determined the partial protein sequence of enzymatically and CNBr-produced peptides. The protein sequences enabled us to clone an essentially full-length (1854 bp) cDNA using PCR and colony screening of an embryonic day (ED) 18 forebrain pUEX-1 cDNA library. Analysis of deduced amino acid sequence shows the 23 kDa protein to be 110 amino acids in length compared to mouse (112) and human and rat (111) while still retaining the conserved cysteine residues. The N-glycosylation site at position 61 is the same as that in the human protein, but is different from that in the rodent (position 75). Northern blot analysis of Thy-1 mRNA expression in the forebrain, cerebellum and tectum show that all three tissues have low levels at ED4 (forebrain and tectum) and ED8 (cerebellum), rising most rapidly between ED16 and the first few days post-hatch. Analysis of various tissues at hatch and adult show expression to be predominantly neuronal with very low levels in some other organs, mainly at hatch, indicating the possibility of subsets of cells, which we have also seen in histological sections, in these tissues expressing Thy-1 mRNA. Bone marrow and blood cells were also negative for Thy-1 protein.

Purification of chick retinal ganglion cells for molecular analysis: combining retrograde labeling and immunopanning yields 100% purity.

Retinal ganglion cells (RGCs) from embryonic and posthatch chickens were 100% purified by a novel combination of three steps: (1) Retrograde labeling by injection of the fluorescent carbocyanine tracer DiI into the optic nerve, (2) immunopanning of dissociated retinal cells with Thy1 antibodies, and (3) micro-aspiration of labeled RGCs into glass capillaries. The retina was dissected and dissociated with trypsin 12-15 h after the injection of DiI. DiI-labeled cells were identified on immunopanned dishes by fluorescence and collected for molecular analysis within 3 h after dissociation. This technique allowed the collection of up to 500 RGCs per capillary tube and 1500 labeled RGCs per retina. Extraction of RNA and molecular analysis by RT-PCR from 600 RGCs shows that expression of rare genes, such as those of neurotrophic factors, can be detected. This is the first description of a rapid and reliable technique for a 100% purification of RGCs with sufficient yield for molecular analysis of rare gene expression. The protocol can be modified for the purification of other cell types. The advantages and limitations of the three-step purification method are compared with previous RGC purification protocols.

Epigenetic factors controlling the development of avian Purkinje neurons.

Using a unique protocol, we have developed an avian neuron culture system in which a high yield of Purkinje neurons is obtained more readily than with pre-existing methods. Purkinje neurons were identified in vitro using the specific antibodies calbindin and cyclic GMP-kinase. Survival of Purkinje neurons was dependent on astrocyte contact and enhanced by astrocytic factors supplied to the medium by a monolayer of astrocytes grown on coated membranes suspended in the culture wells but not in contact with the neurons. The age of the cerebellum from which astrocytes were obtained was shown to affect the morphological development of the Purkinje neurons suggesting the developmentally-regulated expression of growth factors. However, in the presence of the astrocytes, Purkinje neurons could only progress to a limited stage of development based on morphological criteria. The addition to the culture of cerebellar granule neurons at a time of Purkinje neuron development that they would expect to encounter them in vivo resulted in a shift of Purkinje neurons to a mature phenotype. This maturation effect was increased in response to increasing levels of granule neurons, but was independent of the granule neuron ages used. This system offers significant advantages over other Purkinje neuron culture systems and will be useful for studying the extrinsic factors involved in Purkinje neuron development and histogenesis.

Thy-1 expression in the retinotectal system of the chick.

Previous investigations into the occurrence of Thy-1 in the chick retina have not clearly defined when the antigen first appears and have not adequately described its expression during the relatively early phases of retinal ontogeny. We have investigated these issues, using improved immunohistochemical procedures and show that Thy-1 is associated with the retinal ganglion cells from the time they begin to differentiate by extending their axonal projections. In addition, we have found that its expression reflects the growth of the optic fibre layer and the elaboration of the ganglion cell dendritic processes into the inner plexiform layer. For the first time we describe the appearance and the developmental expression of Thy-1 in the chick tectum. We have found that Thy-1 is associated with retinal axons from the time of their arrival at the tectum and that its expression reflects the elaboration of the stratum opticum. Within the tectum proper Thy-1 appears first in 3 distinct layers all of which are plexiform in nature. By the time that tectal histogenesis is essentially complete the antigen is expressed by all the layers of the tectum. The implications of these findings are discussed in terms of the development of the individual tissues and with respect to the elaboration of the retinotectal pathway.

The developmental appearance of Thy-1 in the avian cerebellum.

The cellular localization of the Thy-1 antigen during development of the chick cerebellum has been investigated using a monoclonal antibody SB1-20.11. Improved cellular morphology and retention of both membrane and intracellular antigenicity was achieved by the immunohistochemical labelling of polyester wax sections using an indirect peroxidase visualization protocol. A parallel histological investigation was carried out using a modified silver staining procedure based on that of Bodian. Immunoreactivity was found throughout development in the soma and dendritic tree of the Purkinje cell, in the internal granular layer, white matter and elements of the deep cerebellar nuclei. The antigen's expression closely correlates to the morphological maturation of Purkinje cell population. Furthermore, it appears to reflect the formation of glomeruli and the basket cell interaction with the Purkinje cell. An association of Thy-1 with climbing fibres, as reported previously in rodent species, cannot be unambiguously shown in the chick because of the high levels of Thy-1 expressed throughout development on the Purkinje cell dendrites in the molecular layer. The spatial and temporal pattern of expression in the chick cerebellum suggests that Thy-1 contributes to the definition of synaptic fields.

The metabolic turnover of the major proteins of the postsynaptic density.

We have used the method of Austin, Lowry, Brown and Carter, to measure the steady-state metabolic half-life of tubulin (alpha and beta individually) and actin (beta and gamma together) in the total cytosolic (S3), microsomal (P3), synaptic plasma membrane (SPM) and synaptic junction (SJ) subcellular fractions from 6-day-old and adult chicken forebrain. In the SPM and SJ fractions we also measured the steady-state metabolic half-life of the major postsynaptic density protein (mPSDp). In SPM and SJ fractions from 6-day-old chickens tubulin and actin turned over approximately twice as slowly (t1/2 approximately equal to 24 days) as tubulin and actin in the S3 fraction (t1/2 approximately equal to 13 days). This difference was unlikely merely to be due to association with membranes since the t1/2 values for the proteins were the same in P3 and S3. The estimated t1/2 values for mPSDp were similar to that for tubulin and actin in SPM and SJ fractions. Similar results were obtained in adult chickens except that all t1/2 values in all fractions were approximately 30% larger. The calculated t1/2 values did not change between labelling periods of 4 and 6.5 h suggesting that the lag phase of incorporation of newly synthesized PSD proteins is sufficiently rapid to not produce this result artefactually. When the brain from a non-labelled chicken was homogenized in the presence of the S3 fraction from a labelled chicken and sub-fractionated the relative specific activities of the SPM and SJ fractions produced were 1-2% of those from the labelled brain. These results support the notion that tubulin and actin are intrinsic components of the PSD.

Neural membrane glycoproteins associated with chicken Thy-1: an anti-idiotypic antibody study.

In order to investigate the possible binding of Thy-1 to other neuronal cell surface proteins, anti-idiotypic antibodies were raised using a panel of anti-Thy-1 monoclonal antibodies. Anti-idiotypic antibodies were selected for their ability to bind to day-old chick brain membrane components in enzyme-linked immunosorbent assays (ELISA), and to bind to membrane glycoproteins as determined by Western transfer immunoblotting assays. The 5 monoclonal anti-idiotypic antibodies bind to a membrane glycoprotein component of 70 kDa, and one of the antibodies also binds to 3 higher molecular weight components of 160 kDa, 120 kDa and 90 kDa. These antibodies bind to areas of the chicken cerebellum known to be rich in Thy-1. It is postulated that these molecules are associated with Thy-1, and that the role of Thy-1 on the neuronal cell surface, may be to form complexes with, and/or to stabilise these higher molecular weight glycoproteins during synaptic development.

The developmental appearance of Thy-1 antigen in the avian nervous system.

A monoclonal antibody against chicken Thy-1 glycoprotein was utilised in an indirect binding assay and an immunohistochemical technique to investigate the developmental appearance of Thy-1 in the chicken nervous system. The largest increase of Thy-1 levels in chicken forebrain during embryonic development coincided with the major period of neuronal development and synapse formation but preceded the major period of myelination. Immunohistochemical localisation studies on sections of chicken cerebellum, retina and spinal cord during this period showed that Thy-1 first appeared in regions of tissue which contained differentiated synapses of axons; however, not all such areas were Thy-1 positive. Areas of tissue rich in actively dividing neuroblasts or postmitotic undifferentiated neurons, such as the external granular layer of the cerebellum, showed very little staining. The disappearance of Thy-1-specific staining with age observed in the white matter of sections of cerebellum and spinal cord was due to the masking effect of myelination and not to a loss of Thy-1 from neuronal membranes.

Distribution of Thy-1 in the avian nervous system: immunohistochemical and absorption analyses with a monoclonal antibody.

A monoclonal antibody against chicken Thy-1 has been used to study the histochemical localisation of Thy-1 in chicken nervous and lymphoid tissues and to quantitate the relative amounts of Thy-1 in different brain subregions, subcellular fractions and non-neural tissues. An indirect ELISA using chicken brain membranes as a target established that the highest levels of Thy-1 were present in chicken forebrain, followed by midbrain, brainstem, spinal cord, cerebellum, retina and sciatic nerve. Analyses of subcellular fractions of chicken forebrain revealed a generalised localisation of Thy-1 on membranes comprising both the junctional and extrajunctional components of the synaptosome. Consistent with this finding are the immunohistochemical studies on cryostat sections where Thy-1 was localised to certain axonal and synaptic regions of chicken nervous tissue. Strong monoclonal antibody binding was found in the molecular layer and white matter of chicken cerebellar sections with fibrous staining on the axons running through the granule cell layer. No continuous staining could be seen on the perikaryal membranes of Purkinje cells or granule cells and no staining was present within the cell bodies. The Bergmann glia of the cerebellum were Thy-1-negative. The monoclonal antibody showed preferential binding to the inner plexiform and optic fibre layers of the chicken retina, suggesting a retinal ganglion cell localisation for chicken Thy-1, as has been suggested for the rat and mouse homologues. Surprisingly the lymphocytes of both the bursa and thymus gland were Thy-1-negative, however some extracellular staining was observed of interlobular connective tissue of the bursa.

Characterisation of a novel glycoprotein (AvGp50) in the avian nervous system, with a monoclonal antibody.

A size fractionated lentil lectin-positive fraction derived from a deoxycholate extract of 1-day-old chick forebrain membranes was used to generate a series of monoclonal antibodies (Mabs) against neural antigens. One of these, MabSA1.7 recognises a glycoprotein which is enriched in synaptic plasma membranes, designated AvGp50. Polyacrylamide gel electrophoresis and Western blots show that AvGp50 is comprised of at least two glycoforms, with M(r)s of 53 kDa and 49 kDa respectively. AvGp50 is nervous system specific and most abundantly expressed in the forebrain, tecta and cerebellum where its pattern of expression is developmentally regulated. Immunohistochemical data localises AvGp50 to regions characterised by highly concentrated synapses, in particular, the molecular and granule cell layers of the cerebellum and in the inner and outer plexiform layers in the retina. Solubilization of the protein with the detergent Triton X-100 shows that AvGp50 is predominantly a cytoskeletally associated glycoprotein. However, when a synaptic plasma membrane fraction was treated with Triton X-114, AvGp50 partitioned into the detergent phase. Digestion of the protein with N-glycanase cleaved five N-linked carbohydrate side chains and reduced the molecular weight to approximately 34 and 31 kDa. Removal of the carbohydrate side chains led to an almost complete loss of recognition of the 34 kDa glycoform by the MabSA1.7, suggesting that the monoclonal antibody recognises a carbohydrate rather than peptide epitope.

Maturation of synapses in chicken forebrain.

The detergent treatment used to prepare synaptic junction and post-synaptic density (PSD) fractions cause a much greater disruption and solubilization of the PSD in young chicken forebrain than in adult forebrain. The disruption is manifested as a large decrease in the amount of the major PSD protein present in the isolated fractions. This 'fragility' of the PSD persists throughout the first month after hatching and gradually disappears during the second month. Artificial elevation of systemic testosterone levels speeded up this maturation of the PSD but only if the hormone was injected during the second month.

A population of Thy-1 molecules associated with the cytoskeleton.

Using a monoclonal antibody, SB1 38.456, raised against chicken neuronal Thy-1, we have shown that a significant amount of Thy-1 is associated with the non-ionic detergent-insoluble fraction of chicken brain membranes. Furthermore, this population of Thy-1 is not released by any biochemical method which disrupts microfilaments, microtubules or intermediate filaments, but is released by raising the temperature to 40 degrees C in high ionic strength, cation-containing buffer. We propose, therefore, that a population of Thy-1 molecules is associated with the cytoskeleton either via a linker protein or via other cell surface glycoproteins which are directly associated with the cytoskeleton.

The RNA of chick forebrain synaptosomes.

The RNA species present in myelin, synaptosomal, mitochondrial, microsomal and post-microsomal fractions obtained from chick forebrain have been investigated using polyacrylamide gels. The primary species associated with the synaptosomal fraction are the 28S, 18S and 5S rRNAs, in addition to species L and 4S tRNA. No detectable mitochondrial rRNA has been found. The significance of these results is discussed in relation to the study of synaptosomal protein synthesis.

Denervation exposes cryptic concanavalin A binding sites in skeletal muscle membranes.

Denervation produces significant changes in glycoproteins on the surfaces of skeletal muscles which can be detected as an increase in membrane-bound carbohydrate and an increase in specific binding of labelled lectins. The increased binding of concanavalin A to denervated membranes is detectable only in intact membranes and is abolished by treatments which disrupt or perturb membrane structure whereas the increased binding of Ricinus communis agglutinin is detectable under both conditions. We suggest that, in addition to chemical modification of existing carbohydrate chains in glycoproteins and the synthesis of new glycoproteins, denervation results in an altered geometric arrangement of existing membrane glycoproteins.

Glycosyltransferase activities in chicken brain synaptic junctions.

Isolated synaptic junction fractions from adult chicken brain were found to contain a number of glycoprotein glycosyltransferase with different donor and acceptor specificities. The activities of sialyl- and galactosyltransferases towards endogenous acceptors in the synaptic junction fraction were low but increased greatly after the addition of deglycosylated fetuin or mucin as exogenous acceptors. The activity of the fucosyltransferases towards endogenous acceptors in the fraction was much higher than that of the other transferases. The addition of deglycosylated fetuin caused a smaller increase in the activity than with the other transferases and deglycosylated mucin had no effect. We find no evidence for the enrichment of any glycosyltransferase in the synaptic junction. If anything, our results suggest that the synaptic junction may be specifically depleted in one type of fucosyltransferase.

Thy-1 antigen is specific to ganglion cells in chicks.

The cellular localization of Thy-1 in the chick retina was investigated by selectively destroying certain populations of neurons with toxins. In control retinae four weeks after intravitreal injection of vehicle, there was strong immunoreactivity for Thy-1 in the nerve fibre layer, ganglion cell layer and inner plexiform layer. By contrast, 4 weeks after intraocular injection with 1.25 nmol of colchicine, virtually all ganglion cells had been destroyed, but most amacrine cells remained. Very little Thy-1 immunoreactivity was evident in these retinae. Four weeks after intraocular injection of 2 mumol of N-methyl-D-aspartic acid (NMDA), a large proportion of amacrine cells had been destroyed, but most ganglion cells remained. In these retinae Thy-1 immunoreactivity was present in the nerve fibre, ganglion cell and inner plexiform layers, in the latter with greater intensity than in controls. We conclude that in chicks the Thy-1 antigen is principally, if not exclusively restricted to ganglion cells.

Thermodynamic behaviour of membrane enzymes in Duchenne muscular dystrophy.

Erythrocyte ghost preparations have been prepared from blood of Duchenne patients (DMD), female carriers of the disease and controls. Arrhenius plots of Na+, K+-ATPase activity of these membrane preparations show a biphasic response for controls. For 75% of DMD and carriers the response is monophasic. This is not an inherent property of the membrane since it can vary over time in the one individual and it can be induced in normal membranes by preincubation with DMD plasma. Arrhenius plots of AChE activity showed no such difference between the three sources of blood.