Search

Glycobiology, 9 (8), ISSN:0959-6658

I. El Ayachi, N. Baeza, C. Fernandez, C. Colin, D. Scavarda, P. Pesheva and D. Figarella-Branger (2010) Neuropathology and Applied Neurobiology36, 399–410 KIAA0510, the 3′-untranslated region of the tenascin-R gene, and tenascin-R are overexpressed in pilocytic astrocytomas Aims: Studying the molecules and signalling pathways regulating glioma invasiveness is a major challenge because these processes determine malignancy, progression, relapse and prognosis. We took advantage of our previous study focused on genes that were critical in tumour invasion to further study here an unknown sequence, referred to as KIAA0510, the chromosomal location of which was 1q25, described as a 5596-bp long mRNA and that we found to be significantly overexpressed in pilocytic astrocytomas compared with glioblastomas. Methods and results: Using in silico analysis as well as Polymerase chain reaction techniques, we decipher the full genomic characterization of the KIAA0510 sequence and demonstrate that KIAA0510 constitutes the 3′-untranslated region of tenascin-R gene. We have clearly confirmed the overexpression of tenascin-R in pilocytic astrocytomas vs. glioblastomas at mRNA and protein levels. We also analysed a large series of various brain tumours and found that in the group of astrocytic tumours, tenascin-R expression decreased with malignancy, whereas oligodendrogliomas sometimes retained a high level of tenascin-R even in high-grade tumours. Gangliogliomas strongly expressed tenascin-R too. In contrast, ependymomas and meningiomas were negative. In normal brain, tenascin-R was exclusively expressed by normal oligodendrocytes and subsets of neurones during post-natal development and in adulthood, where it could differentially affect cellular adhesiveness and/or differentiation. Conclusion: KIAA0510, the 3′-untranslated region of the tenascin-R gene, and tenascin-R are overexpressed in pilocytic astrocytomas. Gangliogliomas shared with pilocytic astrocytomas strong tenascin-R expression. Whether tenascin-R overexpression negatively influences brain invasion remains to be determined.

It is currently unclear whether retinal ganglion cell (RGC) axon regeneration depends on down-regulation of axon growth-inhibitory proteins, and to what extent outgrowth-promoting substrates contribute to RGC axon regeneration in reptiles. We performed an immunohistochemical study of the regulation of the axon growth-inhibiting extracellular matrix molecules tenascin-R and chondroitin sulphate proteoglycan (CSPG), the axon outgrowth-promoting extracellular matrix proteins fibronectin and laminin, and the axonal tenascin-R receptor protein F3/contactin during RGC axon regeneration in the lizard, Gallotia galloti. Tenascin-R and CSPG were expressed in an extracellular matrix-, oligodendrocyte/myelin- and neuron-associated pattern and up-regulated in the regenerating optic pathway. The expression pattern of tenascin-R was not indicative of a role in channeling or restriction of re-growing RGC axons. Up-regulation of fibronectin, laminin, and F3/contactin occurred in spatiotemporal patterns corresponding to tenascin-R expression. Moreover, we analyzed the influence of substrates containing tenascin-R, fibronectin, and laminin on outgrowth of regenerating lizard RGC axons. In vitro regeneration of RGC axons was not inhibited by tenascin-R, and further improved on mixed substrates containing tenascin-R together with fibronectin or laminin. These results indicate that RGC axon regeneration in Gallotia galloti does not require down-regulation of tenascin-R or CSPG. Presence of tenascin-R is insufficient to prevent RGC axon growth, and concomitant up-regulation of axon growth-promoting molecules like fibronectin and laminin may override the effects of neurite growth inhibitors on RGC axon regeneration. Up-regulation of contactin in RGCs suggests that tenascin-R may have an instructive function during axon regeneration in the lizard optic pathway.

Whereas the animal lectin galectin-3 is only barely detectable in the adult sciatic nerve, its expression level is substantially upregulated in the damaged nerve (1). In the present study, we demonstrate that in the mouse, the expression of galectin-3 increases during normal peripheral nerve development and is further potentiated in the adult sciatic nerve under conditions of in vitro degeneration. In cell cultures derived from neonatal sciatic nerve galectin-3 is expressed by about 50% of L1-positive Schwann cells. Comparable results were obtained for cultured late immature Schwann cells derived from hindlimbs of 17.5-day-old mouse embryos. In cell cultures derived from hindlimbs of 14-day-old embryos, by contrast, only about 5% of the L1-positive early immature Schwann cells expressed galectin-3. Our findings demonstrate that the upregulation of galectin-3 expression by activated adult Schwann cells is not a part of a dedifferentiation program, but rather attributes to a phenotype specific for injury-activated Schwann cells.

An important biological consequence of the initial interactions between the cell surface and its extracellular environment is the diversity of cellular responses ranging from overt repulsion or avoidance reaction to stable adhesion or final positioning. It is now evident that positive and negative guiding mechanisms are equally relevant to normal pattern formation during development and decisive for the outcome of a regenerative process. In this context, the present review summarizes the knowledge about the extracellular matrix glycoprotein tenascin-R, a member of the tenascin gene family. In contrast to all other known family members, tenascin-R is exclusively expressed in the central nervous system of vertebrates by oligodendrocytes and neuronal subsets at later developmental stages and in adulthood. We focus on the glycoprotein's structure, tissue distribution and functional implications in the molecular control of axon targeting, neural cell adhesion, migration and differentiation during nervous system morphogenesis and pathology.

Tenascin-R (TN-R), an extracellular matrix constituent of the central nervous system (CNS), has been implicated in a variety of cell–matrix interactions underlying axon growth inhibition/guidance, myelination and neural cell migration during development and regeneration. Although most of the functional analyses have concentrated exclusively on the role of the core protein, the contribution of TN-R glycoconjugates present on many potential sites for N- and O-glycosylation is presently unknown. Here we provide first evidence that TN-R derived from whole rat brain or cultured oligodendrocytes expresses chondroitin sulfate (CS) glycosaminoglycans (GAGs), i.e., C-4S and C-6S, that are recognized by CS-56, a CS/dermatan sulfate-specific monoclonal antibody. Based on different in vitro approaches utilizing substrate-bound glycoprotein, we found that TN-R-linked CS GAGs (1) promote oligodendrocyte migration from white matter microexplants and increase the motility of oligodendrocyte lineage cells; (2) similar to soluble CS GAGs, induce the formation of glial scar-like structures by cultured cerebral astrocytes; and (3) contribute to the antiadhesive properties of TN-R for neuronal cell adhesion in an F3/F11-independent manner, but not to neurite outgrowth inhibition, by mechanism(s) sensitive to chondroitinase or CS-56 treatments. Furthermore, after transection of the postcommissural fornix in adult rat, CS-bearing TN-R was found to be stably upregulated at the lesion site. Our findings suggest the functional impact of TN-R-linked CS on neural cell adhesion and migration during brain morphogenesis and the contribution of TN-R to astroglial scar formation (CS-dependent) and axon growth inhibition (CS-independent), i.e., suppression of axon regeneration after CNS injury. J. Neurosci. Res. 60:21–36, 2000 © 2000 Wiley-Liss, Inc.

We have previously shown that the extracellular matrix molecule tenascin-C inhibits fibronectin-mediated cell adhesion and neurite outgrowth by an interaction with a cellular RGD-independent receptor which interferes with the adhesion and neurite outgrowth promoting activities of the fibronectin receptor(s). Here we demonstrate that the inhibitory effect of tenascin-C on beta1integrin-dependent cell adhesion and neurite outgrowth is mediated by the interaction of the protein with membrane-associated disialogangliosides, which interferes with protein kinase C-related signaling pathways. First, in substratum mixtures with fibronectin, an RGD sequence-containing fragment of the molecule or synthetic peptide, tenascin-C inhibited cell adhesion and spreading by a disialoganglioside-dependent, sialidase-sensitive mechanism leading to an inhibition of protein kinase C. Second, the interaction of intact or trypsinized, i.e., cell surface glycoprotein-free, cells with immobilized tenascin-C was strongly inhibited by gangliosides or antibodies to gangliosides and tenascin-C. Third, preincubation of immobilized tenascin-C with soluble disialogangliosides resulted in a delayed cell detachment as a function of time. Similar to tenascin-C, immobilized antibody to GD2 (3F8) or sphingosine, a protein kinase C inhibitor, strongly inhibited RGD-dependent cell spreading. Finally, the degree of tenascin-C-induced inhibition of cell adhesion was proportional to the degree of disialoganglioside levels of expression by different cells suggesting the relevance of such mechanism in modulating integrin-mediated cell-matrix interactions during pattern formation or tumor progression.

In the present study we have analyzed the expression of galectin-3, a beta-galactoside-specific soluble animal lectin, by microglial cells in vitro. In enriched microglial cell cultures derived from neonatal mouse brain after 2 to 3 weeks in vitro, almost all microglial cells expressed galectin-3 intracellularly and about 90% expressed the molecule on the cell surface. Western blot analyses of lysates from microglial cells using galectin-3-specific antibodies revealed a single band with an apparent molecular weight of 29 kD. The carbohydrate recognition domain of microglia-derived galectin-3 was functional as the molecule could be affinity purified on lactose-agarose. Upon an incubation with lactose-, but not with sucrose-containing buffers the amount of cell surface expressed galectin-3 was strongly reduced, suggesting that the molecule appears to be associated with the plasma membrane via its carbohydrate recognition domain. The total amount as well as the portion of cell surface expressed galectin-3 increased upon treatment with granulocyte-macrophage colony-stimulating factor. Our findings suggest that galectin-3 expression is subject to regulation by growth factors supposed to be involved in the cascade of microglial activation under pathological conditions.

O4+oligodendrocyte (OL) progenitors in the mammalian CNS are committed fully to terminal differentiation into myelin-forming cells. In the absence of other cell typesin vitro, OL differentiation reproduces thein vivodevelopment with a correct timing, suggesting the existence of an intrinsic regulatory mechanism that presently is unknown. We have examined the effect of two isoforms of the extracellular matrix (ECM) molecule tenascin-R (TN-R), which is expressed by OLs during the process of myelination, on the adhesion and maturation of OLsin vitro. Here we show that the substrate-bound molecules supported the adhesion of O4+OLs independently of the CNS region or age from which they were derived. At the molecular level this process was mediated by protein binding to membrane surface sulfatides (Sulf), as indicated by the interference of O4 antibody and Sulf with the attachment of OLs or other Sulf+cells, erythrocytes, to TN-R substrates and by direct protein–glycolipid binding studies. In the absence of platelet-derived growth factor (PDGF), exogenous TN-R induced myelin gene expression and the upregulation of its own synthesis by cultured cells, resulting in a rapid terminal differentiation of O4+progenitors. Our findings strongly suggest that TN-R represents an intrinsic regulatory molecule that controls the timed OL differentiation by an autocrine mechanism and imply the relevance of TN-R for CNS myelination and remyelination.

Tenascin-R is an extracellular matrix glycoprotein that is restricted to the central nervous system, where it acts as a multifunctional and versatile molecule. We report spatial and temporal distribution of tenascin-R in the developing human cerebral cortex for the first time. At 7.5 gestational weeks (GW), tenascin-R was expressed in a restricted area of the basal telencephalon. At 9.5 and 11 GW, it showed a unique double band expression pattern that delineated the boundaries of the future cortical plate. From 14 to 30 GW, tenascin-R labeling extended to the whole cortex from the deep layers toward the marginal zone with an inside-to-outside progression pattern reminiscent of neuronal migration. Moreover, tenascin-R labeling initially appeared in the form of thin, straight, or slightly tortuous intercellular processes directed toward the surface in parallel with the axis of neuronal migration. At the end of pregnancy and at adulthood, diffuse and homogeneous immunolabeling of the whole cortex thickness was observed. The striatum and thalamus were faintly positive for TNR as early as 14 GW, and this positivity intensified with brain maturation. At all developmental stages, the germinative zone, the corpus callosum, the anterior commissure, and the internal capsule appeared clearly negative for tenascin-R immunostaining whereas the adjacent parenchyma was immunopositive. Our results show that tenascin-R expression is tightly regulated in a spatiotemporal manner during brain development, especially cortical plate formation. Its pattern of expression suggests a role for tenascin-R in corticogenesis. J. Comp. Neurol. 519:2379–2389, 2011. © 2011 Wiley-Liss, Inc.

Axon growth in vitro may be inhibited by contact with oligodendrocytes, but most axons grow readily on the surface of astrocyte monolayers. Since both cell types are in close contact with one another in the damaged nervous system, we have examined the growth of axons on cultures which contain both astrocytes and oligodendrocytes. Cultures derived from neonatal rat forebrain develop with a monolayer of large flat astrocytes attached to the culture dish, and with many smaller cells of the oligodendrocyte lineage on their surface. Dorsal root ganglia placed on these cultures grow axons readily, the overall extent of growth being unaffected by the presence or absence of oligodendrocytes, many of which express galactocerebroside and the inhibitory molecule janusin. A previous set of experiments had shown that growth of these axons is inhibited by oligodendrocytes by themselves. Scanning electron microscopy coupled with silver-intensified immunostaining reveals that the axons grow on the surface of the astrocytic layer, underneath the oligodendrocytes, and are therefore in contact with both cell types as they grow. The presence of astrocytes therefore alters the results of axonal contact with oligodendrocytes.

Runx2 is a member of the Runx family of transcription factors (Runx1–3) with a restricted expression pattern. It has so far been detected predominantly in skeletal tissues where, inter alia, it regulates the expression of the β-galactoside-specific lectin galectin-3. Here we show that, in contrast to Runx3, Runx1 and Runx2 are expressed in a variety of human glioma cells. Runx2 expression pattern in these cells correlated completely with that of galectin-3, but not with that of other galectins. A similar correlation in the expression pattern of galectin-3 and Runx2 transcripts was detected in distinct types of 70 primary neural tumors, such as glioblastoma multiforme, but not in others, such as gangliocytomas. In glioma cells, Runx2 is directly involved in the regulation of galectin-3 expression, as shown by RNAi and transcription factor binding assays demonstrating that Runx2 interacts with a Runx2-binding motif present in the human galectin-3 promoter. Knockdown of Runx2 was thus accompanied by a reduction of both galectin-3 mRNA and protein levels by at least 50%, dependent on the glial tumor cell line tested. Reverse transcriptase–polymerase chain reaction analyses, aimed at finding other potential target genes of Runx2 in glial tumor cells, revealed the presence of bone sialoprotein, osteocalcin, osteopontin, and osteoprotegerin. However, their expression patterns only partially overlap with that of Runx2. These data suggest a functional contribution of Runx-2-regulated galectin-3 expression to glial tumor malignancy. © 2008 Wiley-Liss, Inc.

Tenascin-R (TN-R) is a member of the tenascin family of multidomain matrix glycoproteins that is expressed exclusively in the central nervous system by oligodendrocytes and small neurons during postnatal development and in the adult. TN-R contributes to the regulation of axon extension and regeneration, neurite formation and synaptogenesis, and neuronal growth and migration. TN-R can be modified with three distinct sulfated oligosaccharide structures: HNK-1 (SO(4)-3-GlcUAbeta1,3Galbeta1,4GlcNAc), GalNAc-4-SO(4), and chondroitin sulfate. We have determined that TN-R expressed in dendrite-rich regions of the rat cerebellum, hippocampus, and cerebral cortex is one of the major matrix glycoproteins that bears N-linked carbohydrates terminating with beta1,4-linked GalNAc-4-SO(4). The syntheses of these unique sulfated structures on TN-R are differentially regulated. Levels of HNK-1 on TN-R rise and fall in parallel to the levels of TN-R during postnatal development of the cerebellum. In contrast, levels of GalNAc-4-SO(4) are regulated independently from those of TN-R, rising late in cerebellar development and continuing into adulthood. As a result, the pattern of TN-R modification with distinct sulfated carbohydrate structures changes dramatically over the course of postnatal cerebellar development in the rat. Because TN-R interacts with a number of different matrix components and, depending on the circumstances, can either activate or inhibit neurite outgrowth, the highly regulated addition of these unique sulfated structures may modulate the adhesive properties of TN-R over the course of development and during synapse maintenance. In addition, the 160-kDa form of TN-R is particularly enriched for terminal GalNAc-4-SO(4) later in development and in the adult, suggesting additional levels of regulation.

Galectin-3 (gal-3) is a member of the galectin family of lectins whose expression strongly depends on the cellular state. Here we show that in PC12 cells the expression of gal-3 protein is regulated via Ras- and mitogen-activated protein kinase (MAPK)-dependent and independent signalling pathways and correlates with nerve growth factor (NGF)-mediated neuronal differentiation. Gal-3 expression, activation of the MAPK ERK1/2 and neurite outgrowth are induced by NGF and basic fibroblast growth factor (bFGF), but not by ciliary neurotrophic factor (CNTF), epidermal growth factor, insulin or interleukin-6 (IL-6). In addition, in NGF-treated PC12 cells, gal-3 expression, ERK1/2 activation and neurite outgrowth could be specifically inhibited at the level of TrkA, Ras and MAPK-kinase, whereas expression of an oncogenic form of Ras leads to gal-3 expression and neurite outgrowth in the absence of growth factors. In NGF-primed PC12 cells, subsequent treatment with CNTF or IL-6 induces ERK1/2 activation and neurite outgrowth, but not gal-3 expression. Treatment of PC12 cells with staurosporine induces gal-3 expression and neurite outgrowth without ERK1/2 activation. NGF- and staurosporine-induced gal-3-expression is also regulated at the transcriptional level. Our data suggest the presence of complex induction mechanisms of gal-3 expression in neuronally differentiating PC12 cells involving NGF-, but not CNTF- and IL-6-driven (in NGF-primed cells) Ras/MAPK-related signalling pathways. Staurosporine, in contrast, induces gal-3 expression by a Ras/MAPK-independent mechanism.

Lipid rafts are microdomains of the plasma membrane which are enriched in glycosphingolipids and specific proteins. The reported interactions of several raft-associated proteins (such as, e.g., F3) with tenascin C and tenascin R prompted us to consider that these oligomeric multidomain glycoproteins of the extracellular matrix (ECM) could associate with rafts. Here, we show punctate immunocytochemical distributions of tenascin C (TN-C) and tenascin R (TN-R) at the membrane surface of neural cells resembling the pattern reported for raft-associated proteins. Moreover, cholesterol depletion with methyl-beta-cyclodextrin reduced the punctate surface staining of TN-C. Consistently, TN-C was associated with lipid rafts of neonatal mouse brain according to sucrose density gradient centrifugation experiments. Furthermore, TN-R was also found in rafts prepared from myelin of adult mice. Thus, brain-derived tenascins are able to associate with lipid rafts. (C) 2002 Elsevier Science (USA). All rights reserved.

Chondroitin sulfate proteoglycan (CS-PG) expression is increased in response to CNS injury and limits the capacity for axonal regeneration. Previously we have shown that neurocan is one of the CS-PGs that is upregulated (Asher et al., 2000). Here we show that another member of the aggrecan family, versican, is also upregulated in response to CNS injury. Labeling of frozen sections 7 d after a unilateral knife lesion to the cerebral cortex revealed a clear increase in versican immunoreactivity around the lesion. Western blot analysis of extracts prepared from injured and uninjured tissue also revealed considerably more versican in the injured tissue extract.In vitrostudies revealed versican to be a product of oligodendrocyte lineage cells (OLCs). Labeling was seen between the late A2B5-positive stage and the O1-positive pre-oligodendrocyte stage. Neither immature, bipolar A2B5-positive cells, nor differentiated, myelin-forming oligodendrocytes were labeled. The amount of versican in conditioned medium increased as these cells differentiated. Versican and tenascin-R colocalized in OLCs, and coimmunoprecipitation indicated that the two exist as a complex in oligodendrocyte-conditioned medium. Treatment of pre-oligodendrocytes with hyaluronidase led to the release of versican, indicating that its retention at the cell surface is dependent on hyaluronate (HA). In rat brain, approximately half of the versican is bound to hyaluronate. We also provide evidence of a role for CS-PGs in the axon growth-inhibitory properties of oligodendrocytes. Because large numbers of OLCs are recruited to CNS lesions, these results suggest that OLC-derived versican contributes to the inhospitable environment of the injured CNS.

The extracellular matrix glycoprotein tenascin-R (TN-R) has been implicated in a variety of cell-matrix interactions involved in the molecular control of axon guidance and neural cell migration during development and regeneration of the central nervous system (CNS). Whereas TN-R is amply expressed in the early postnatal and adult mammalian CNS, the protein has so far not been detected in different compartments of the peripheral nervous system (PNS). Here we provide first evidence that TN-R (predominantly TN-R 160 isoform) is transiently expressed in the sciatic nerve of late embryonic (E14-18) and neonatal mice, while at later developmental stages, both protein and mRNA are downregulated. In vitro, TN-R protein was found to be expressed by both undifferentiated and neuronally differentiated PC12 cells and by L1-positive Schwann cells (SC), but not by other neural and non-neural cell types in cell cultures derived from embryonic (E17/18) hindlimbs and neonatal sciatic nerves. In the developing PNS, TN-R expression correlated with axon growth and SC migration during the period of skeletal muscle innervation. Based on different in vitro approaches, we found that the substrate-bound glycoprotein selectively inhibits the fibronectin-dependent: (1) neurite outgrowth from dorsal root ganglion neurons (strongly expressing alpha5beta1 integrin and the disialoganglioside GD3) by a ganglioside-sensitive signaling mechanism; and (2) migration of primary myoblasts and other non-neuronal cells in a ganglioside-independent manner. Our findings suggest the functional role of TN-R in PNS pattern formation during distinct stages of axon pathfinding and skeletal muscle innervation.

Galectin-3, a member of the galectin family of beta-galactoside-specific lectins has been found to be expressed by subsets of dorsal root ganglion (DRG) neurons during development and in adulthood. Here we show that (i) after 3-7 days in vitro, DRG neurons derived from neonatal mice express galectin-3 intra- and extracellularly and (ii) lectin expression requires the presence of nerve growth factor (NGF). After 3 days in vitro, a higher number of DRG neurons expressed galectin-3 in the presence of NGF (65 +/- 7%) than in the presence of brain-derived neurotrophic factor (BDNF, 30 +/- 3%) or neurotrophin-3 (NT-3, 34 +/- 3%). After 7 days in vitro, these numbers dropped to 51 +/- 3% (for NGF), 0% (for BDNF) and 8 +/- 4% (for NT-3), respectively. Our findings provide first evidence for the contribution of a neurotrophin to the neuronal expression of galectins and suggest an NGF/TrkA-mediated expression of galectin-3 by early postnatal DRG neurons.

We have recently demonstrated that the beta-galactoside-specific lectin galectin-3 is expressed by microglial cells in vitro, but not by normal resting microglia in vivo. In the present study, we have analyzed the expression of galectin-3 by microglia under traumatic conditions in vivo using two experimental rat models which substantially differ in the severity of lesion related to a breakdown of the blood-brain barrier (BBB) and the occurrence of inflammatory processes. These two features are absent after peripheral nerve lesion and present after cerebral ischemia. Here we show that, following facial nerve axotomy under conditions allowing (nerve anastomosis) or not subsequent regeneration (nerve resection), galectin-3 is not expressed by microglia in the corresponding facial nucleus 1-112 days after lesion. Galectin-3 is also absent in microglia at sites of a defective BBB in the normal brain, such as the circumventricular organs. Following experimental ischemia (i.e., permanent occlusion of the middle cerebral artery), in contrast, galectin-3 becomes strongly expressed by activated microglia as early as 48 hours after trauma, as determined by immunohistochemistry and Western blot analysis. Our findings suggest that the expression of galectin-3 by microglia in vivo correlates with the state of microglial activation.

Tenascin-R (TN-R), a matrix glycoprotein of the central nervous system (CNS), has been implicated in a variety of cell-matrix interactions involved in the control of axon growth, myelination and cell adhesion to fibronectin during development and regeneration. While most of the functional analyses have concentrated exclusively on the role of the core protein, the contribution of TN-R glycoconjugates present on many potential sites for N- and O-glycosylation is presently unknown. Here we provide evidence that TN-R derived from adult mouse brain expresses chondroitin sulfate (CS) glycosaminoglycans (GAGs), i.e. C-6S and C-4S, that are recognized by the CS/dermatan sulfate-specific monoclonal antibodies 473 HD and CS-56. Using ligand-binding, cell adhesion and neurite outgrowth assays, we show that TN-R-linked CS GAGs (i) are involved in the interaction with the heparin-binding sites of fibronectin and are responsible for TN-R-mediated inhibition of cell adhesion to a 33/66-kD heparin-binding fibronectin fragment or to FN-C/H I and FN-C/H II peptides, known to participate in fibronectin binding to cell surface proteoglycans; and (ii) partially contribute to the interaction between TN-R and TN-C which, however, does not lead to an interference with TN-R- and TN-C-mediated inhibition of neurite outgrowth when the two molecules are offered as a mixed substrate in culture. Our findings suggest the functional implication of TN-R-linked CS GAGs in matrix interactions with fibronectin and TN-C that are likely to contribute to a modulation of cellular behavior and the macromolecular organization of matrix components in the developing or injured adult CNS.

In the central nervous system (CNS), tenascin-R (TN-R) is mainly expressed by oligodendrocytes and in white matter tracts. Here, we have examined the molecular association of TN-R with CNS myelin by incubation of myelin membranes (MM) purified from adult mouse brain under different ionic conditions. By Western blot analysis, the 160 kDa isoform was the main TN-R component detectable in MM as a dimer which became degraded to monomers of 160 kDa and major fragments of 125 and 80 kDa in the absence of protease inhibitors. In the presence of chelating agents, TN-R was completely extracted from MM. Calcium ions promoted the dissociation of TN-R while zinc or copper blocked it. TN-R release from MM was sensitive to heat suggesting the involvement of calcium-dependent myelin protease(s) in this process. In addition, 1,10-phenanthroline (a metalloprotease blocker) partially inhibited TN-R release in the presence of calcium ions. We conclude that divalent metal ions stabilize the association of TN-R with CNS myelin and upon damage, the protein can be released and degraded by endogenous proteases, suggesting the implication of myelin-derived TN-R in axon growth inhibition and demyelinating diseases.

Galectin-3 is a member of the galectin family of beta-galactoside-specific animal lectins. Here we show that galectin-3 is constitutively expressed in 15 out of 16 glioma cell lines tested, but not by normal or reactive astrocytes, oligodendrocytes, glial O-2A progenitor cells and the oligodendrocyte precursor cell line Oli-neu. Galectin-3 is also expressed by one oligodendroglioma cell line, but not by primitive neuroectodermal tumor and 4 neuroblastoma cell lines tested so far. In all galectin-3 expressing cell lines, the lectin is predominantly, if not exclusively, localized intracellularly and carries an active carbohydrate recognition domain (shown for C6 rat glioma cells). Moreover, in contrast to primary astrocytes, glioma cells do not or only weakly adhere to substratum-bound galectin-3, probably reflecting an unusual glycosylation pattern. Our findings indicate that the expression of galectin-3 selectively correlates with glial cell transformation in the central nervous system and could thus serve as a marker for glial tumor cell lines and glial tumors.

The number of animal lectins, basically defined upon their interaction with specific carbohydrate structures, is growing considerably during the last few years. Among these proteins the recently identified subfamily of I-type lectins consists of mainly transmembranous glycoproteins belonging to the immunoglobulin superfamily. Most of the I-type lectins participate in cell adhesion events, as are the different sialoadhesins recognizing sialylated glycan structures, which represent the best characterized subgroup. I-type lectins are abundant in the nervous system and have been implicated in a number of morphogenetic processes as fundamental as axon growth, myelin formation and growth factor signaling. In the present review, we summarize the structural and functional properties of I-type lectins expressed in neural tissues with a main focus on the sialoadhesin myelin-associated glycoprotein, the neural cell adhesion molecule and the fibroblast growth factor receptors.

Galectin-3 is a member of the galectin family and belongs to a group of soluble beta-galactoside-binding animal lectins. The molecule is expressed by neural and nonneural cells intra- (cytoplasm and nucleus) as well as extra-cellularly (plasma membrane and extracellular space). By using an in vitro cell-substratum adhesion assay, we have addressed the question whether galectin-3 present in the extracellular milieu may support the adhesion and/or neurite outgrowth of neural cells in a manner analogous to cell adhesion molecules. Galectin-3 was immobilized as a substratum and various cell types, N2A (neuroblastoma), PC12 (pheochromocytoma), and TSC (transformed Schwann cells) cell lines, neural cells from early postnatal mouse cerebellum, and dorsal root ganglion neurons from newborn mice were allowed to adhere to the lectin. Here we show that all cell types studied specifically adhered to galectin-3 by the following criteria: 1) the number of adherent cells was dependent on the galectin-3 concentration used for coating; 2) adhesion of cells to galectin-3, but not to collagen type I or laminin was inhibited by polyclonal antibodies to galectin-3; 3) upon addition of asialofetuin (a polyvalent carrier of terminal beta-galactosides) to the cell suspension prior to the adhesion assay, cell adhesion to galectin-3 was inhibited in a dose-dependent manner; and 4) cell adhesion to galectin-3 was abolished by treatment of cells with endo-beta-galactosidase. In addition, the adhesion of dorsal root ganglion neurons to galectin-3 could be inhibited by lactose. Notably, substratum-bound galectin-3 promoted the outgrowth of neurites from dorsal root ganglia explants and this neurite outgrowth promoting activity could be inhibited by polyclonal antibodies to galectin-3.

We have previously described two proteoglycans from human sciatic nerve which are immunochemically related to the chondroitin sulphate proteoglycans versican and decorin. The chondroitin sulphate of the versican-like molecule and the core protein of the decorin-like molecule have been found previously to be up-regulated after lesioning the adult mouse sciatic nerve. To investigate if the versican- and decorin-like molecules are involved in cell-extracellular matrix interactions, we studied the effect of both molecules on cell adhesion. The versican- and decorin-like molecules, substrate-coated in a mixture with fibronectin, but not with laminin or collagen types I or IV, inhibited the adhesion of several cell lines, neonatal dorsal root ganglion neurons and Schwann cells. The inhibitory activity was concentration-dependent and mediated by the chondroitin sulphate. Furthermore, when different proteoglycans were incubated with fibronectin, only the versican- and decorin-like molecules and the chondroitin sulphate proteoglycan aggrecan, but not the heparan sulphate proteoglycan perlecan, were able to inhibit fibronectin-mediated cell adhesion. The versican- and decorin-like molecules, substrate-coated alone or in a mixture with fibronectin or laminin, were at most slightly inhibitory to neurite outgrowth from PC12 phaeochromocytoma cells and neonatal dorsal root ganglion neurons. In a solid-phase ligand-binding assay the versican- and decorin-like molecules interacted with fibronectin, but not with laminin or collagen types I and IV. Binding of the versican-like molecule to fibronectin and inhibition of cell adhesion by this molecule was mediated via the heparin and cell-binding domains of fibronectin. These observations suggest that binding of the two proteoglycans to fibronectin is involved in the modulation of adhesion of cells to fibronectin.

Janusin (formerly designated J1–160/180) is an extracellular matrix glycoprotein highly homologous to tenascin, consisting of two major molecular forms of 160 and 180 kD expressed by oligodendrocytes and in myelin. Janusin expression is upregulated during myelination and in the adult it remains expressed at lower levels. It is also present at the node of Ranvier, where myelin, axon, and astrocytic process are in close contact. To gain an understanding of the regulatory mechanisms which may under-lie expression of janusin, the differentiation stage-dependent expression of janusin was studied in cultures enriched in mouse oligodendrocytes and their precursor cells. Expression of janusin by these cells was highest on both A2B5+ and O4+/O1− oligodendroglial precursor cells and a subset of myelin associated glycoprotein-positive (MAG+) oligodendrocytes. Hardly any of the more differentiated O1+ or O10+ oligodendrocytes expressed janusin. Expression of janusin was influenced by co-culture with astrocytes or neurons. Astrocytes or astrocytic-conditioned culture supernatants elevated the expression of janusin by the more differentiated oligodendrocytes (O1+ or MAG+ cells), while its expression by oligodendroglial precursor cells was relatively unchanged. Platelet-derived growth factor, but not basic fibroblast growth factor, also elevated the expression of janusin by O1+ or O10+ oligodendrocytes. In contrast, co-culture with neurons originating from dorsal root ganglia or spinal cord decreased the expression of cell-bound janusin by oligodendrocytes and their precursor cells. These observations indicate that expression of janusin on these cells in culture is susceptible to opposing regulatory influences from astrocytes and neurons. Such influences may modulate the temporal and spatial distribution of janusin in the developing and adult central nervous system. © 1993 Wiley-Liss, Inc.

We have analyzed the expression of the oligodendrocyte-derived extra-cellular matrix molecule janusin (previously termed J1-160/180) in the retina and optic nerve of developing and adult mice using indirect light and electron microscopic immunocytochemistry, immunoblot analysis, and enzyme-linked immunosorbent assay. In the optic nerve, janusin is not detectable in neonatal and only weakly detectable in 7-day-old animals. Expression is at a peak in 2- or 3-week-old animals and subsequently decreases with increasing age. In the retina, expression increases until the third postnatal week and then remains at a constant level. In immunocytochemical investigations at the light microscopic level, janusin was found in the myelinated regions of the nerve with spots of increased immunoreactivity possibly corresponding to an accumulation of the molecule at the nodes of Ranvier. At the electron microscopic level, contact sites between unmyelinated axons, between axons and glial cells, and between axons and processes of myelinating oligodendrocytes were immunoreactive. Cell surfaces of astrocytes at the periphery of the nerve and forming the glial-limiting membrane, in contrast, were only weakly immunopositive or negative. In cell cultures of young postnatal mouse or rat optic nerves, oligodendrocytes and type-2 astrocytes, but not type-1 astrocytes were stained by janusin antibodies. In the oligodendrocyte-free retina, janusin was detectable in association with neuronal cell surfaces, but not with cell surfaces of Muller cells or retinal astrocytes. Our observations indicate that expression of janusin in the optic nerve and in the retina is developmentally differentially regulated and that other cell types, in addition to oligodendrocytes, express the molecule. Since the time course of janusin expression in the optic nerve coincides with the appearance of oligodendrocytes and myelin and since janusin is associated with cell surfaces of oligodendrocytes and outer aspects of myelin sheaths and is concentrated at nodes of Ranvier, we suggest that janusin is functionally involved in the process of myelination.

The oligodendrocyte-derived extracellular matrix protein J1-160/180 displays repellent substrate properties toward neurons. In a search for neuronal ligands mediating the response to J1-160/180, we have identified the F3/11 cell surface protein, a glyco-phosphatidylinositol-anchored member of the immunoglobulin superfamily. F3/11 mediates the initial recognition between a J1-160/180 substrate and cerebellar neurons or F3-transfected CHO cells. In cerebellar neurons, the F3/11-J1-160/180 interaction induces a repulsion consisting of the loss of substrate adhesion with time in culture and inhibition of neurite outgrowth. Antibody blocking experiments show that the avoidance response of neurites at J1-160/180 substrate borders is also mediated by F3/11. Active cell-cell and cell-substrate repulsion is considered a major mechanism governing the extent and directionality of axonal growth, but the ligand-receptor interactions involved have remained unknown. Our results show that F3/11 mediates the neuronal response to the repellent molecule J1-160/180 and may thus be involved in signal transduction leading to cell repulsion.

J1-160 and J1-180 are developmentally late appearing J1 extracellular matrix glycoproteins derived from oligodendrocytes. They prevent adhesion of neurons (but not of astrocytes or fibroblasts) when offered as a substrate in mixture with laminin (Pesheva et al., J. Cell Biol., 109, 1765 - 1778, 1989). In the present study we have examined the influence of divalent cations on the inhibitory substrate properties of J1-160/180 glycoproteins towards adhesion of neurons. By metal chelate affinity chromatography, we show that J1-180, but not J1-160, binds Ca2+, while both J1 components are capable of binding Zn2+ and other divalent metal ions. Divalent cation binding was observed by gel filtration, aggregation assays with coated latex beads and electron microscopic examination to elicit aggregation of the molecules. Divalent cation binding also affects their non-permissive substrate properties towards neurons from early postnatal mouse cerebellum. Without divalent cations, J1-160 and J1-180 are inhibitory for substrate adhesion of neurons independently of the adhesive substrate present (laminin or poly-l-lysine). This effect is neutralized when J1-180 is preincubated with Ca2+ or Zn2+ prior to coating as substrate. In contrast, preincubation with Ca2+ ions does not affect the inhibitory substrate properties of J1-160 under these conditions. These observations show that J1-160/180 molecules may undergo self-aggregation in a divalent cation-dependent mechanism, which correlates with the neutralization of their inhibitory effect on neuronal adhesion. The aggregation state of the molecules may thus influence the process of myelination by a homophilic binding mechanism and determine the effectiveness of neurite extension during central nervous system development and under traumatic conditions in the adult.

We have studied the functional involvement of J1-160 and J1-180 in the interaction between oligodendrocytes and neurons, astrocytes, or L cells in short- and long-term adhesion assays using monoclonal antibodies directed against topographically distinct epitopes on the molecules. Whereas antibodies to mouse liver membranes and monoclonal antibody 597 do not interfere with neuron-oligodendrocyte or astrocyte-oligodendrocyte adhesion after 30 min of coculture, antibodies 596, 619, and 620 interfere with astrocyte to oligodendrocyte and neuron to oligodendrocyte adhesion. The adhesion of L cells to oligodendrocytes is not affected by the antibodies. When neurons or astrocytes are cultured on oligodendrocytes for more than 30 min, monoclonal antibody 619 continues to reduce adhesion of astrocytes to oligodendrocytes after 1 and 2 h. However, during this time period the antibody affects neuron to oligodendrocyte adhesion in a different manner. It does not interfere with adhesion of neurons to oligodendrocytes at 1 h and enhances the adhesion of neurons to oligodendrocytes after 2 h of coculture. After 6 and 24 h of coculture, antibody 619 does not affect the adhesion of neurons or astrocytes to oligodendrocytes, suggesting that other adhesive mechanisms are predominant at later times of interaction. At all times studied, neurons and astrocytes adhered well to the oligodendrocytes. To study the influence of the J1 molecules on neuronal interactions in the absence of other oligodendrocyte-derived cell surface components, purified J1-160 was coated as a substrate and neuron attachment was measured as a function of time. Two hours after plating neurons adhered well to J1-160, as they did to laminin, while cell detachment was subsequently observed from J1-160, but not from laminin. These results implicate J1-160 and J1-180 in a recognition process between oligodendrocytes and neurons or astrocytes, but not fibroblasts. This recognition process appears to merge into adhesion or stabilization of cell contacts for astrocytes and destabilization of cell interactions or repulsion for neurons. It is likely that these two opposite effects in cell behavior elicited by the J1 molecules result from differential intracellular responses to a cell surface trigger possibly mediated by different cell surface receptors and/or different consequences in intracellular signaling networks.

Cell surface receptors for extracellular matrix components have recently been characterized as integral membrane complexes with common features in their structural and functional properties. We have investigated the expression of the mammalian fibronectin receptor in the mouse nervous system using immunocytological and immunochemical methods. The fibronection receptor was detectable on immature oligodendrocytes and immature and mature astrocytes in culture, while central nervous system neurons did not reveal detectable levels of fibronectin receptor at the developmental stages studied. In the peripheral nervous system both glia and neurons were found to express the fibronectin receptor. The receptor complex in both peripheral and central nervous system has an apparent molecular weight of approximately 140 kD under reducing conditions and resolves into two or three distinct protein bands under nonreducing conditions. The fibronectin receptor expresses the L2/HNK-1 epitope that is characteristic of several adhesion molecules, including L1, N-CAM, the myelin-associated glycoprotein, and J1 and thus is another member of the L2/HNK-1 family of adhesion molecules. The L2/HNK-1 carbohydrate epitope is expressed differently and independently of the fibronection receptor protein backbone in that it is detectable in neonatal brain but not in adult brain. Our observations attribute a functional role to the fibronectin receptor and its L2/HNK-1 carbohydrate epitope during development and maintenance of cell interactions in the central and peripheral nervous systems.

On the basis of recent evidence that the carbohydrate structures designated L2/HNK-1 and L3 are shared by several neural adhesion molecules including L1, N-CAM, the myelin-associated glycoprotein and J1, we have suggested that other members of the L2/HNK-1 and L3 families are adhesion molecules. Here we show by Western blot analysis that integrin, the cell surface receptor for the extracellular matrix constituents fibronectin and laminin in chicken, also expresses the L2/HNK-1 and L3 epitopes and thus is another family member proven to be involved in adhesion.

The glia-derived J1 extracellular matrix glycoproteins have been referred to as J1-160/J1-180 (the developmentally late appearing lower molecular weight group) and J1-200/J1-220 (the developmentally early appearing higher molecular group immunochemically related to tenascin). Members of the two groups show distinct cross-reactivities. To characterize the structural and functional differences between these J1 glycoproteins, two monoclonal antibodies were generated which recognize only the members of the lower molecular weight group. The two antibodies detect immunochemical similarities among the members of the lower molecular weight group, but do not react with J1/tenascin. J1-160 and J1-180 are specifically expressed by differentiated oligodendrocytes in culture and by myelin of the central nervous system and have not been found in the peripheral nervous system nor in any other organ of the adult mice tested. Electron microscopic examination of rotary-shadowed J1-160 and J1-180 reveals, respectively, dimeric and trimeric (tribrachion) kink-armed rodlike structures, which are linked by disulfide bridges. J1-160/J1-180 are nonpermissive substrates for the attachment and spreading of early postnatal small cerebellar neurons, astrocytes, and fibroblasts. In a mixture with laminin, J1-160/J1-180 are nonpermissive substrates for neurons, but not for astrocytes or fibroblasts. The repulsive effect toward neurons can be neutralized by one of the monoclonal antibodies, but not by the other. These observations are discussed in the context of cell interactions during regeneration in the mammalian nervous system.

The myelin-associated glycoprotein (MAG) can be obtained from adult mouse brain from detergent-lysates of a crude membrane fraction as a 96-100 kd form (detergent solubilized MAG), and from 100,000 g supernatants of homogenates as a 90-96 kd form (soluble MAG). The soluble form distributes into the Triton X-114-poor aqueous phase, while detergent-solubilized MAG predominantly enters the Triton X-114-rich phase. Both molecular forms bind to heparin in hypo- and isotonic buffers. Soluble MAG binds to several collagens (type G, I, II, III, IV, V, VI, IX) with a kd of 5.7 X 10(-8) M for collagen type IX and 2.0 X 10(-7) for collagen type IV. Binding of 125I-labeled MAG to collagen G can be completely inhibited by unlabeled MAG and collagen G, but not by heat-denatured collagen. MAG does not bind to itself, laminin, fibronectin, or the neural cell adhesion molecules L1 and N-CAM. Binding of MAG to collagen G is most effectively blocked by a high molecular weight dextran sulfate, heparan sulfate and heparin, with chondroitin sulfate and a low molecular weight dextran sulfate being less potent blockers. These findings are in agreement with previous observations on the localization of MAG in basal lamina and interstitial collagens of the sciatic nerve in situ.

Microglial activation in response to pathological stimuli is characterized by increased migratory activity and potential cytotoxic action on injured neurons during later stages of neurodegeneration. The initial molecular changes in the CNS favoring neuronofugal migration of microglia remain, however, largely unknown. We report that the extracellular matrix protein tenascin-R (TN-R) present in the intact CNS is antiadhesive for activated microglia, and its downregulation after facial nerve axotomy may account for the loss of motoneuron protection and subsequent neurodegeneration. Studies on the protein expression in the facial and hypoglossal nucleus in rats demonstrate that TN-R is a constituent of the perineuronal net of motoneurons and 7 d after peripheral nerve injury becomes downregulated in the corresponding motor nucleus. This downregulation is reversible under regenerative (nerve suture) conditions and irreversible under degenerative (nerve resection) conditions. In short-term adhesion assays, the unlesioned side of brainstem cryosections from unilaterally operated animals is nonpermissive for activated microglia, and this nonpermissiveness is almost abolished by a monoclonal antibody to TN-R. Microglia-conditioned media and tumor necrosis factor-alpha downregulate TN-R protein and mRNA synthesis by cultured oligodendrocytes, which are one of the sources for TN-R in the brainstem. Our findings suggest a new role for TN-R in neuronal protection against activated microglia and the participation of the latter in perineuronal net destruction, e.g., downregulation of TN-R.