The fourth immunoglobulin-like domain of NCAM contains a carbohydrate recognition domain for oligomannosidic glycans implicated in association with L1 and neurite outgrowth

We have previously shown that the neural adhesion molecules L1 and NCAM interact with each other to form a complex which binds more avidly to L1 than L1 to L1 alone (Kadmon, G., A. Kowitz, P. Altevogt, and M. Schachner. 1990a. J. Cell Biol. 110:193-208). This cis-association between L1 and NCAM is carbohydrate-dependent (Kadmon, G., A. Kowitz, P. Altevogt, and M. Schachner. 1990b. J. Cell Biol. 110:209-218). In the present study, we report that L1 and NCAM bind to each other via oligomannosidic carbohydrates expressed by L1, but not by NCAM, as shown in several experiments: (a) complex formation between L1 and NCAM is inhibited by a mAb to oligomannosidic carbohydrates and by the oligosaccharides themselves; (b) NCAM binds to oligomannosidic carbohydrates; (c) within the L1/NCAM complex, the oligomannosidic carbohydrates are hidden from accessibility to a mAb against oligomannosidic carbohydrates; (d) the recombinant protein fragment of NCAM containing the immunoglobulin-like domains and not the fragment containing the fibronectin type III homologous repeats binds to oligomannosidic glycans. Furthermore, the fourth immunoglobulin-like domain of NCAM shows sequence homology with carbohydrate recognition domains of animal C-type lectins and, surprisingly, also with plant lectins. A peptide comprising part of the C-type lectin consensus sequence in the fourth immunoglobulin-like domain of NCAM interferes with the association between L1 and NCAM. The functional importance of oligomannosidic glycans at the cell surface was shown for neurite outgrowth in vitro. When neurons from early postnatal mouse cerebellum were maintained on laminin or poly-L-lysine, neurite outgrowth was inhibited by oligomannosidic glycans, by glycopeptides, glycoproteins, or neoglycolipids containing oligomannosidic glycans, but not by nonrelated oligosaccharides or oligosaccharide derivates. Neurite outgrowth was also inhibited by the peptide comprising part of the C- type lectin consensus sequence in the fourth immunoglobulin-like domain of NCAM. The combined results suggest that carbohydrate-mediated cis- associations between adhesion molecules at the cell surface modulate their functional properties.     

C-kinase manipulations disrupt activity-driven retinotopic sharpening in regenerating goldfish retinotectal projection

Regenerating optic axons initially branch over a wide area in tectum to form a crude retinotopic map. The map is sharpened, and retinotopically appropriate synapses are stabilized via NMDA receptors that detect, via summation of EPSPs, the coincident activity of neighboring ganglion cells that make synapses onto common tectal cells. Sharpening shares a number of properties with long-term potentiation (LTP) in hippocampus. This study tested whether protein kinase C (PKC) activation is necessary for sharpening as it is for LTP. Intracular (IO) or intracranial (IC) injections of kinase inhibitors or activators were made every other day from 19 to 37 days postcrush (sensitive period), and the projections formed were later recorded. Retinotopic sharpening was prevented by IC injection of the following agents: (1) general kinase inhibitors sphingosine and H7 (100-200 μM in fluid above brain), (2) active but not inactive phorbols (TPA, 1 μM), and (3) calphostin C (1 μM), a specific and irreversible PKC inhibitor. The mature projection on the opposite tectum, however, when examined was not unsharpened. Lack of sharpening was reflected in multiunit fields at each tectal point that averaged 27°–30° versus 11° in Ringers and inactive phorbol control regenerates. Intraocular injections of either TPA (1 μM), or calphostin C (1 μM) also prevented sharpening (26° and 32° multiunit fields), suggesting action on PKC axonally transported to the presynaptic terminals. Calphostin C had no noticeable effect on the firing patterns of retinal ganglion cells. The endogenous activator of PKC, arachidonic acid (AA), disrupted sharpening at 20 μM or higher (IC injection, 32° multiunit fields), while a control fatty acid, elaidic acid, had no effect. Although AA at 5 μM showed no effect, and diacylglycerol at 5 μM exhibited only small effects, together they produced a large synergistic effect (32° multiunit fields). Such synergy mirrors the synergy in the activation of several isoforms of PKC. Actual concentrations in the extradural fluid around the brain were assayed via injections of ³H-AA. Levels fell about sixfold after a day and by an additional fivefold the second day before the next injection. The results confirm that activity-driven retinotopic sharpening is very sensitive to manipulations of kinases, especially PKC. © 1994 John Wiley & Sons, Inc.     

Activity-driven sharpening of the retinotectal projection in goldfish: Development under stroboscopic illumination prevents sharpening

Blocking or synchronizing activity during regeneration of the retinotectal projection prevents both the sharpening of the retinotopic map recorded on tectum and the refinement of the structure of individual arbors within the plane of the map, and this refinement is triggered by N-methyl-d-aspartate (NMDA) receptors. We tested whether activity-driven refinement also occurs during development of the projection in larval and young adult goldfish. Shortly after hatching, larval goldfish were placed into tanks within light-tight chambers illuminated by a xenon strobe at 1 Hz for 14 h of each daily cycle. Fish were reared for 1.5–2 years, until large enough to record in our retinotectal mapping apparatus (6 cm length). Age- and size-matched controls had normal maps with multiunit receptive fields (MURFs) recorded at each tectal point of 10.8° (0.16 S.E.M., n = 5), whereas the strobe-reared fish had only roughly retino-topic maps with much enlarged MURFs averaging 26.7° (1.41 S.E.M., n = 5). This enlargement represents an abnormal convergence onto each tectal point, as the maps failed to sharpen during development. The arbors of individual retinal axons were stained with horseradish peroxidase (HRP) in larval fish and in adult strobereared and control fish. They were drawn with camera lucida from tectal whole mounts, and analyzed for spatial extent in the plane of the retinotopic map, order of branching, number of branch endings, depth of termination, and caliber of the parent axon. Arbors from larval fish (1–2 weeks) were small (approximately 50 × 40 μm) with less than 10 branches, occupied a single strata, and could not be separated into different classes by caliber of axon. The 87 arbors stained in control adult fish (6 cm long) were much like previously examined adult arbors, with those from fine, medium, and coarse axons averaging 115, 166, and 194 μm in extent, respectively, and having 17–24 branch endings. The 110 arbors from 12 strobe-reared fish were often abnormal. Although the fasciculation was normal, the extrafascicular routes were abnormal with reversing turns. The axons often had branches along their course, and these branches were scattered across a wider extent, rather than forming a distinct cluster. In contrast, neither the number of branches nor the depths of termination was significantly changed in any group. The coarse caliber arbors were most abnormal, being 64% longer and 30% wider than controls. The fine caliber arbors were also significantly larger by about 20%, but the medium caliber arbors were not enlarged. The enlarged arbors partially account for the unsharpened electrophysiological maps. Together the results show that during development, as well as during regeneration, the retinotectal map is subject to an activity-driven sharpening process. © 1993 John Wiley & Sons, Inc.     

Selective stabilization of retinotectal synapses by an activity-dependent mechanism

How does each ingrowing retinal fiber select the right spot in the overall retinotopic projection? Chemospecific surface interactions appear to be sufficient only to organize a crude retinotopic map on the tectum during regeneration of the optic nerve of goldfish. Precise retinotopic ordering is achieved via an activity-dependent stabilization of appropriate synapses, based on the correlated activity of neighboring ganglion cells of the same receptive field type in the retina. Four treatments have been found to block the sharpening process: 1) blocking activity of the ganglion cells with intraocular tetrodotoxin (TTX); 2) rearing in total darkness; 3) correlated activation of all ganglion cells via stroboscopic illumination in a featureless environment; 4) block of retinotectal synaptic transmission with alpha-bungarotoxin. These experiments support a role for normal visually driven activity in sharpening the diffuse projection, and demonstrate that the correlated activity of the optic fibers interacts within the postsynaptic cells, probably through the summation of excitatory postsynaptic potentials. Intraocular TTX experiments suggest that a similar mechanism may drive both the formation of ocular dominance patches in fish tectum and kitten visual cortex and the segregation of different receptive field types in the lateral geniculate nucleus. Thus, it may be a general mechanism whereby the diffuse projections of early development are brought to a mature level of organization.     

Formation of retinotopic connections: Selective stabilization by an activity-dependent mechanism

During regeneration of the optic nerve in goldfish, the ingrowing retinal fibers successfully seek out their correct places in the overall retinotopic projection on the tectum. Chemospecific cell-surface interactions appear to be sufficient to organize only a crude retinotopic map on the tectum during regeneration. Precise retinotopic ordering appears to be achieved via an activity-dependent stabilization of appropriate synapses and is based upon the correlated activity of neighboring ganglion cells of the same receptive-field type in the retina. Four treatments have been found to block the sharpening process: (a) blocking the activity of the ganglion cells with intraocular tetrodotoxin (TTX), (b) rearing in total darkness, (c) correlating the activation of all ganglion cells via stroboscopic illumination and (d) blocking retinotectal synaptic transmission with alpha-bungarotoxin (alphaBTX). These experiments support a role for correlated visually driven activity in sharpening the diffuse projection and suggest that this correlated activity interacts within the postsynaptic cells, probably through the summation of excitatory postsynaptic potentials (EPSPs). Other experiments support the concept that effective synapses are stabilized: a local postsynaptic block of transmission causes a local disruption in the retinotectal map. The changes that occur during this disruption suggest that each arbor can move to maximize its synaptic efficacy. In development, initial retinotectal projections are often diffuse and may undergo a similar activity-dependent sharpening. Indirect retinotectal maps, as well as auditory maps, appear to be brought into register with the direct retinotopic projections by promoting the convergence of contacts with correlated activity. A similar mechanism may drive both the formation of ocular dominance patches in fish tectum and kitten visual cortex and the segregation of different receptive-field types in the lateral geniculate nucleus. Activity-dependent synaptic stabilization may therefore be a general mechanism whereby the diffuse projections of early development are brought to the precise, mature level of organization.     

L1/HNK-1 Carbohydrate- and β1 Integrin-Dependent Neural Cell Adhesion to Laminin-1

Abstract: We have shown recently that mouse small cerebellar neurons adhere to a short amino acid sequence of the G2 domain of the laminin α1 chain via the cell surface-expressed HNK-1 carbohydrate. Therefore, we were interested in identifying glycoproteins carrying the HNK-1 carbohydrate at the cell surface of these neurons. Adhesion of small cerebellar neurons to laminin is partially dependent on Ca²⁺, Mn²⁺, and Mg²⁺, indicating the involvement of integrins, which were identified as β1, α3, and α6. They could be shown to bind to laminin by a β1-dependent adhesion mechanism. None of these subunits was found to carry the HNK-1 carbohydrate. HNK-1-immunoreactive glycoproteins were immunoprecipitated and shown to consist of predominantly one molecular species, which was identified as the neural cell recognition molecule L1. L1 was demonstrated to bind in a concentration-dependent and saturating manner to laminin. The binding could be partially inhibited by Fab fragments of monoclonal antibodies against the HNK-1 carbohydrate and against the Ig-like domains of L1. Furthermore, antibodies to the Ig-like domains of L1 and β1 integrin inhibited partially cell adhesion to laminin. Determination of the association of L1, β1 integrin, and the HNK-1 carbohydrate on the cell surface of live cerebellar neurons by antibody-induced patching and copatching revealed HNK-1 to be linked to L1, but less so to β1 integrin. However, only negligible association was found between L1 and β1 integrin. Furthermore, it could be shown that adhesion to laminin is mediated by L1/HNK-1- and β1 integrin-dependent mechanisms that act at least partially independent of each other.     

Expression of carbohydrate epitopes L2/HNK-1 and L3 in the larva and imago of Drosophila melanogaster and Calliphora vicina

Summary The carbohydrate epitopes L2/HNK-1 and L3 belong to two overlapping families of adhesion molecules in the vertebrate, and probably the invertebrate nervous systems. To investigate their pattern of expression during the development of insects, cryosections of late third instar larvae and imagoes of Drosophila melanogaster and Calliphora vicina were studied by indirect immunofluorescence using several monoclonal antibodies to the L2/HNK-1 and one monoclonal antibody to the L3 epitope. Each monoclonal antibody to the L2/HNK-1 epitope showed a different immunohistological staining pattern, which differed from that of the L3 monoclonal antibody. In both insect species the immunohistological staining patterns for the two carbohydrate epitopes were similar at the two developmental stages, with immunoreactivity not confined to the nervous system. In larvae, immunoreactivities of the monoclonal antibodies L2.334 and L3.492 were predominantly associated with the extracellular matrix as indicated by co-localization with laminin, particularly in the imaginal discs, while L2.349 revealed a more cell surface-associated distribution. In imagoes, immunoreactivities were detectable in most organs studied.     

Extracellular matrix molecules and cell adhesion molecules induce neurites through different mechanisms

It has recently become clear that both extracellular matrix (ECM) glycoproteins and various cell adhesion molecules (CAMs) can promote neurite outgrowth from primary neurons, though little is known of the intracellular mechanisms through which these signals are transduced. We have previously obtained evidence that protein kinase C function is an important part of the neuronal response to laminin (Bixby, J.L. 1989. Neuron. 3:287-297). Because such CAMs as L1 (Lagenauer, C., and V. Lemmon. 1987. Proc. Natl. Acad. Sci. USA. 84:7753-7757) and N-cadherin (Bixby, J.L. and R. Zhang. 1990. J. Cell Biol. 110:1253-1260) can be purified and used as substrates to promote neurite growth, we have now tested whether the response to CAMs is similarly dependent on protein kinase C. We find that inhibition of protein kinase C inhibits growth on fibronectin or collagen as well as on laminin. In contrast, C kinase inhibition actually potentiates the initial growth response to L1 or N- cadherin. The later "phase" of outgrowth on both of these CAMs is inhibited, however. Additionally, phorbol esters, which have no effect on neurite growth when optimal laminin concentrations are used, potentiate growth even on optimal concentrations of L1 or N-cadherin. The results indicate that different intracellular mechanisms operate during initial process outgrowth on ECM substrates as compared to CAM substrates, and suggest that protein kinase C function is required for continued neurite growth on each of these glycoproteins.     

Cranial and trunk neural crest cells use different mechanisms for attachment to extracellular matrices.

We have used a quantitative cell attachment assay to compare the interactions of cranial and trunk neural crest cells with the extracellular matrix (ECM) molecules fibronectin, laminin and collagen types I and IV. Antibodies to the beta 1 subunit of integrin inhibited attachment under all conditions tested, suggesting that integrins mediate neural crest cell interactions with these ECM molecules. The HNK-1 antibody against a surface carbohydrate epitope under certain conditions inhibited both cranial and trunk neural crest cell attachment to laminin, but not to fibronectin. An antiserum to alpha 1 intergrin inhibited attachment of trunk, but not cranial, neural crest cells to laminin and collagen type I, though interactions with fibronectin or collagen type IV were unaffected. The surface properties of trunk and cranial neural crest cells differed in several ways. First, trunk neural crest cells attached to collagen types I and IV, but cranial neural crest cells did not. Second, their divalent cation requirements for attachment to ECM molecules differed. For fibronectin substrata, trunk neural crest cells required divalent cations for attachment, whereas cranial neural crest cells bound in the absence of divalent cations. However, cranial neural crest cells lost this cation-independent attachment after a few days of culture. For laminin substrata, trunk cells used two integrins, one divalent cation-dependent and the other divalent cation-independent (Lallier, T. E. and Bronner-Fraser, M. (1991) Development 113, 1069-1081). In contrast, cranial neural crest cells attached to laminin using a single, divalent cation-dependent receptor system. Immunoprecipitations and immunoblots of surface labelled neural crest cells with HNK-1, alpha 1 integrin and beta 1 integrin antibodies suggest that cranial and trunk neural crest cells possess biochemically distinct integrins. Our results demonstrate that cranial and trunk cells differ in their mechanisms of adhesion to selected ECM components, suggesting that they are non-overlapping populations of cells with regard to their adhesive properties.     

Homophilic NCAM interactions interfere with L1 stimulated neurite outgrowth

The cell adhesion molecules NCAM and L1 are considered to play key roles in neuronal development and plasticity. L1 has been shown to interact with NCAM, possibly through NCAM binding to oligomannosidic glycans present in L1. We investigated the effect of recombinant immunoglobulin (Ig) modules of NCAM involved in homophilic NCAM binding, on L1 induced neurite outgrowth from PC12-E2 cells and found a complete inhibition of L1 induced neurite outgrowth after addition of Ig-modules 1, 2 and 3 of NCAM, suggesting that the ligation state of NCAM is crucial for normal L1 signaling.     

Activity sharpens the regenerating retinotectal projection in goldfish: sensitive period for strobe illumination and lack of effect on synaptogenesis and on ganglion cell receptive field properties

The regenerating optic nerve of goldfish first reestablishes a rough retinotopic map on the contralateral tectum and then sharpens it. Disruption of visual activity, either by blocking activity with intraocular tetrodotoxin (TTX; Schmidt and Edwards, 1983) or by synchronizing activity with strobe illumination (Schmidt and Eisele, 1985), disrupts the sharpening process: the map is correctly oriented but the multiunit receptive fields at each point average 25-40 degrees in diameter. In order to test whether strobe and TTX interfere with the same mechanism, we have tested whether their sensitive periods are the same, and whether strobe, like TTX treatment, does not affect either ganglion cell receptive field properties or synaptogenesis. In parallel studies, we exposed fish to 2 weeks of either strobe illumination or intraocular TTX beginning at various times after crush and determined via electrophysiological recordings that the periods of sensitivity were nearly identical. There was no effect of either treatment during the first 2 weeks (before the fibers arrive at the tectum), maximal disruption of sharpening between 14 and 50 days (the period of rapid synaptogenesis), decreasing disruption between 50 and 125 days, and no effect beyond that point or in the normal projection. In addition, long strobe exposures of up to 142 days produced no greater disruptions than shorter 2-3-week exposures, indicating no cumulative effect. The reestablishment of synaptic transmission in tectum, assayed by recording field potentials elicited by optic nerve shock, was not affected by stroboscopic illumination. Finally, individual ganglion cells, recorded intraretinally following long-term strobe exposure, had receptive fields that were normal both in size and in their characteristic responses to light-on, to light-off, or to both on and off. These findings support the hypothesis that strobe-like TTX prevents retinotopic refinement by preventing the correction of errors initially made by the ingrowing optic axons (Schmidt et al., 1988).     

Biosynthesis of HNK-1 glycans on O-linked oligosaccharides attached to the neural cell adhesion molecule (NCAM): the requirement for core 2 beta 1,6-N-acetylglucosaminyltransferase and the muscle-specific domain in NCAM

The HNK-1 glycan, sulfo-->3GlcAbeta1-->3Galbeta1-->4GlcNAcbeta1-->R, is highly expressed in neuronal cells and apparently plays critical roles in neuronal cell migration and axonal extension. The HNK-1 glycan synthesis is initiated by the addition of beta1,3-linked GlcA to N-acetyllactosamine followed by sulfation of the C-3 position of GlcA. The cDNAs encoding beta1,3-glucuronyltransferase (GlcAT-P) and HNK-1 sulfotransferase (HNK-1ST) have been recently cloned. Among various adhesion molecules, the neural cell adhesion molecule (NCAM) was shown to contain HNK-1 glycan on N-glycans. In the present study, we first demonstrated that NCAM also bears HNK-1 glycan attached to O-glycans when NCAM contains the O-glycan attachment scaffold, muscle-specific domain, and is synthesized in the presence of core 2 beta1,6-N-acetylglucosaminyltransferase, GlcAT-P, and HNK-1ST. Structural analysis of the HNK-1 glycan revealed that the HNK-1 glycan is attached on core 2 branched O-glycans, sulfo-->3GlcAbeta1-->3Galbeta1-->4GlcNAcbeta1-->6(Galbeta1-->3)GalNAc. Using synthetic oligosaccharides as acceptors, we found that GlcAT-P and HNK-1ST almost equally act on oligosaccharides, mimicking N- and O-glycans. By contrast, HNK-1 glycan was much more efficiently added to N-glycans than O-glycans when NCAM was used as an acceptor. These results are consistent with our results showing that HNK-1 glycan is minimally attached to O-glycans of NCAM in fetal brain, heart, and the myoblast cell line, C2C12. These results combined together indicate that HNK-1 glycan can be synthesized on core 2 branched O-glycans but that the HNK-1 glycan is preferentially added on N-glycans over O-glycans of NCAM, probably because N-glycans are extended further than O-glycans attached to NCAM containing the muscle-specific domain.     

High Expression of the HNK-1/L2 Carbohydrate Epitope in the Major Glycoproteins of Shark Myelin

The major 24- and 28-kDa glycoproteins in shark PNS and CNS myelin express high levels of the adhesion-associated HNK-1/L2 carbohydrate epitope. The 28-kDa protein, but not the 24-kDa protein, cross-reacts strongly with one of two anti-bovine P0 antisera not previously tested against fish myelin proteins. Shark PNS and CNS myelin also contains smaller amounts of high-molecular-weight HNK-1-positive proteins, including a prominent broad band in the 65-85-kDa range. Although myelin-associated glycoprotein (MAG) is well known to react with HNK-1 in some mammals, monoclonal and polyclonal anti-MAG antibodies did not react with the high-molecular-weight HNK-1-positive material in shark myelin, a result suggesting that it is not a MAG-like protein. The high expression of the HNK-1/L2 epitope in glycoproteins of shark myelin, including the major P0-related ones, suggests that this adhesion-related carbohydrate structure may have had an important role in the molecular evolution of the myelinating process.     

Determination of structural elements of the L2/HNK-1 carbohydrate epitope required for its function

The L2/HNK-1 carbohydrate epitope has been shown to carry an unusual 3-sulfoglucuronic acid linkedO-glycosidically through a neolactosyl-type back bone to a ceramide residue. Using monoclonal antibodies, the same or a closely related epitope has also been detectedN-glycosidically linked to glycoproteins, amongst them several neural cell adhesion molecules. We used synthetic glycolipids carrying sulfated or non-sulfated glucuronic acid attached to ceramide through glycans of different length to show that not only the sulfated glucuronic acid but also the neolactosyl-type backbone is essential for the recognition of the L2/HNK-1 carbohydrate by a monoclonal antibody, its binding to laminin and its role in neural cell migration and outgrowth of processes from neurons and astrocytes.Abbreviations mab monoclonal antibody - TLC thin layer chromatography - HRP horseradish peroxidase - glcA glucuronic acid - gal galactose - glcNAc N-acetyl-glucosamine - man mannose     

Avian neural crest cell attachment to laminin: involvement of divalent cation dependent and independent integrins.

The mechanisms of neural crest cell interaction with laminin were explored using a quantitative cell attachment assay. With increasing substratum concentrations, an increasing percentage of neural crest cells adhere to laminin. Cell adhesion at all substratum concentrations was inhibited by the CSAT antibody, which recognizes the chick beta 1 subunit of integrin, suggesting that beta 1-integrins mediate neural crest cell interactions with laminin. The HNK-1 antibody, which recognizes a carbohydrate epitope, inhibited neural crest cell attachment to laminin at low coating concentrations (greater than 1 microgram ml-1; Low-LM), but not at high coating concentration of laminin (10 micrograms ml-1; High-LM). Attachment to Low-LM occurred in the absence of divalent cations, whereas attachment to High-LM required greater than 0.1 mM Ca2+ or Mn2+. Neural crest cell adherence to the E8 fragment of laminin, derived from its long arm, was similar to that on intact laminin at high and low coating concentrations, suggesting that this fragment contains the neural crest cell binding site(s). The HNK-1 antibody recognizes a protein of 165,000 Mr which is also found in immunoprecipitates using antibodies against the beta 1 subunit of integrin and is likely to be an integrin alpha subunit or an integrin-associated protein. Our results suggest that the HNK-1 epitope on neural crest cells is present on or associated with a novel or differentially glycosylated form of beta 1-integrin, which recognizes laminin in the apparent absence of divalent cations. We conclude that neural crest cells have at least two functionally independent means of attachment to laminin which are revealed at different substratum concentrations and/or conformations of laminin.     

Characterization of soluble forms of NCAM

Neural cell adhesion molecule (NCAM) has been described as a family of membrane glycoproteins. However, soluble NCAM immunoreactivity has long been recognized. We here show that soluble NCAM is composed of two quantitatively major polypeptides of Mr 180,000 and 115,000 and two minor components of Mr 160,000 and 145,000. Soluble NCAM was immunochemically identical to membrane NCAM, was polysialylated and carried the HNK-1 epitope. It only constituted 0.8% of total NCAM in newborn rat brain. Soluble NCAM appeared in neuronal cell culture medium 15-30 min after the start of synthesis preceding accumulation of membrane-associated NCAM on the cell surface. This indicates that soluble NCAM contains a secreted component.     

Human Myelin/Oligodendrocyte Glycoprotein: A New Member of the L2/HNK-1 Family

Abstract— Myelin/oligodendrocyte glycoprotein (MOG) is a quantitatively minor component of CMS myelin. In this study, human MOG was found to express the L2/HNK-1 epitope on N-linked oligosaccharide structures. This carbohydrate epitope has been found previously in three other characterized human myelin glycoproteins: the my-elin-associated glycoprotein, P₀, and the oligodendrocyte-myelin glycoprotein. It seems, therefore, that the L2/HNK-1 epitope is expressed frequently in human myelin glycoproteins. Serial lectin affinity chromatography of ¹⁴C-glycopeptides indicated that MOG N -oligosaccharide structures are mainly of the complex type, accounting for 77.8% of total radioactivity. In contrast with myelin-asso-ciated glycoprotein and P₀, which express the L2/HNK-1 epitope on fucosylated structures, in MOG the epitope was detected on all glycopeptide fractions obtained by serial lectin affinity chromatography, although a preferential expression of the L2/HNK-1 epitope was observed on fucosylated structures. Finally, the data indicated that, as for other human myelin glycoproteins, only a subpopulation of MOG molecules expresses the L2/HNK-1 epitope.     

Region-specific upregulation of HNK-1 glycan in the PRMT1-deficient brain

BackgroundBrains express structurally unique glycans, including human natural killer-1 (HNK-1), which participate in development and high-order functions. However, the regulatory mechanisms of expression of these brain-specific glycans are largely unknown. We examined whether arginine methylation, another type of protein modification essential for neural development, impacts the expression of various glycans in the developing brain.MethodsWe analyzed several types of glycans, including the HNK-1 epitope, in the cerebellum and cerebral cortex from mice with nervous system-specific knockout of protein arginine methyltransferase 1 (PRMT1). We also analyzed the expression levels of glycosyltransferases responsible for HNK-1 and of HNK-1 carrier glycoproteins by quantitative RT-PCR and western blotting.ResultsAmong several glycans, expression of HNK-1 glycan was strikingly upregulated in the PRMT1-deficient cerebellum. Furthermore, such upregulation was found in the cerebellum but not in the cerebral cortex. Regarding the mechanisms, we demonstrated that the mRNA level and activity of the responsible glycosyltransferase (B3gat1) were elevated in the knockout cerebellum. We also showed that the expression of HNK-1 carrier glycoproteins such as neural cell adhesion molecule (NCAM), L1 and AMPA receptor subunit GluA2 were also increased in the PRMT1-deficient cerebellum.ConclusionsLoss of arginine methylation leads to an increase in HNK-1 glycan in the developing cerebellum but not in the cerebral cortex via upregulation of the biosynthetic enzyme and carrier glycoproteins.General significancePRMT1 is a novel regulator of HNK-1 glycan production in the cerebellum. Mechanisms involving crosstalk between glycosylation and arginine methylation are suggested to occur.     

A family of novel, acidic N-glycans in Bowes melanoma tissue plasminogen activator have L2/HNK-1-bearing antennae, many with sulfation of the fucosylated chitobiose core.

A family of about 20 novel acidic bi- and tri-antennary N-glycans, amounting to almost half those expressed on Bowes melanoma tissue-plasminogen activator (t-PA) were found to possess Galbeta1-->4GlcNAcbeta1-->, sulfated and sialylated GalNAcbeta1-->4GlcNAcbeta1--> or sulfated GlcAbeta1--> 3Galbeta1-->4GlcNAcbeta1--> antennae, of which those containing sulfated GlcA, depicting the L2/HNK-1 carbohydrate epitope, were preferentially located on the 6 arm. A proportion of the glycans were highly charged, because of multiple and variously distributed sulfation, some of which was located on the fucosylated chitobiose core. Multiple expression of the L2/HNK-1 epitope on a single glycan was observed. The most abundant compound was a biantennary glycan carrying sulfated GlcA on the 6-branched antenna and an alpha2-->6 sialylated GalNAc on the other. The N-glycosylation sequon containing Asn448, which is known to express all of the sulfate-carrying N-glycans contains, unusually, an arginine residue. An electrostatic interaction between this cationic amino acid and the core-sulfate group of the N-glycan is proposed to reduce mobility of the carbohydrate in the region of the t-PA active site. Because of the 'brain-type' nature of the N-glycans described in this neuro-ectodermal cell line, the possibility of neural t-PA interacting with the L2/HNK-1-recognizing molecule, laminin, of the central nervous system extracellular matrix is discussed.     

Laminin-1 is a novel carrier glycoprotein for the nonsulfated HNK-1 epitope in mouse kidney

The HNK-1 epitope has a unique structure comprising the sulfated trisaccharide (HSO(3)-3GlcAbeta1-3Galbeta1-4GlcNAc), and two glucuronyltransferases (GlcAT-P and GlcAT-S) are key enzymes for its biosynthesis. However, the different functional roles of these enzymes in its biosynthesis remain unclear. Recently, we reported that a nonsulfated form of this epitope, which is biosynthesized by GlcAT-S but not by GlcAT-P, is expressed on two metalloproteases in mouse kidney. In this study, we found that a novel glycoprotein carrying the nonsulfated HNK-1 epitope in mouse kidney was enriched in the nuclear fraction. The protein was affinity-purified and identified as laminin-1, and we also confirmed the N-linked oligosaccharide structure including nonsulfated HNK-1 epitope derived from laminin-1 by mass spectrometry. Curiously, immunofluorescence staining of kidney sections revealed that laminin-1 appeared not to be colocalized with the nonsulfated HNK-1 epitope. However, proteinase treatment strengthened the signals of both laminin-1 and the nonsulfated HNK-1 epitope, resulting in overlapping of them. These results indicate that the nonsulfated HNK-1 epitope on laminin-1 is usually embedded and masked in the robust basement membrane in tight association with other proteins. To clarify the associated proteins and the functional role of the carbohydrate epitope, we investigated the interaction between laminin-1 and alpha-dystroglycan through their glycans in mouse kidney using the overlay assay technique. We obtained evidence that glucuronic acid as well as sialic acid inhibited this interaction, suggesting that the nonsulfated HNK-1 epitope on laminin-1 may regulate its binding and play a role in maintenance of the proper structure in the kidney basal lamina.