Monoclonal antibodies (mAbs) against the major constituents of cartilage extracellular matrix, aggrecan and link protein, were screened by indirect immunofluorescence on frozen sections of bovine spinal cord. Antibodies against aggrecan and link protein gave rise to very similar perineuronal labeling in spinal cord gray matter. Aggrecan and link protein reactivities were seen in other regions of the central nervous system (CNS), although their distributions were not always coincident. Pretreatment of the tissue section with Streptomyces hyaluronidase, which is hyaluronate‐specific, led to the loss of both reactivities. On Western blots, anti‐aggrecan mAbs reacted with a large chondroitin sulfate proteoglycan. The chondroitinase‐treated CNS proteoglycan co‐migrated with the chondroitinase‐ and keratanase‐treated cartilage proteoglycan. In CNS tissue homogenates, the addition of Streptomyces hyaluronidase brought about the release of the proteoglycan from the tissue. Anti‐link protein mAbs were reactive with two species in the bovine CNS, the mobilities of which were very similar to those of the cartilage link proteins. The release of these species from the tissue required hyaluronidase. A rabbit antiserum against aggrecan was used to identify a similar proteoglycan in the rat CNS. In spinal cord‐derived cell cultures, the labeled material was associated with astrocytes. An aggrecan cDNA hybridized to a 9.5 kb mRNA in the rat CNS. We conclude that the perineuronal matrix consists, in part, of a hyaluronate‐bound aggrecan‐like proteoglycan and link proteins, and that the former is produced by astrocytes. © 1995 Wiley‐Liss, Inc.

Glial hyaluronate-binding protein (GHAP) is a 60 kDa glycoprotein with an amino acid sequence identical to that of the hyaluronate-binding region of versican, a large fibroblast aggregating proteoglycan found in the brain. Both GHAP and versican were identified by immunoblot in bovine brain extracts prepared only minutes after death. Human recombinant collagenase, stromelysin, mouse gelatinase and gelatinases isolated from human brain by affinity chromatography digest versican and give rise to a polypeptide with electrophoretic mobility identical to GHAP. Immunoblot analysis, peptide mapping and C-terminal amino acid sequencing indicate that the polypeptide generated by digestion with human brain gelatinases is identical to GHAP. We suggest that GHAP is a naturally occurring versican degradation product.

In the rat, intracerebral injection of bacterial hyaluronidase resulted in the almost complete disappearance of hyaluronic acid (HA) and glial hyaluronate‐binding protein (GHAP) from cerebral hemispheres, brain stem, and cerebellum (but not from optic nerves and chiasm) starting 2–3 hr after the injection. HA and GHAP reappeared throughout the brain in characteristic patches 2–3 days after the injection. The patches gradually became confluent and after 12 days the brain appeared virtually normal. In normal rat optic nerve, staining for HA and GHAP ceased abruptly in the region of the lamina cribrosa. The retina was completely negative. HA and GHAP disappeared from hyaluronidase‐injected optic nerve, chiasm, and contralateral optic nerve. In hyaluronidase‐injected crushed optic nerves, regenerated axons were able to grow for short distances (about 500μm) into the distal stump undergoing Wallerian degeneration. No such growth was observed in saline‐injected controls. © 1993 Wiley‐Liss, Inc.

The localization of versican, a large hyaluronate-binding fibroblast proteoglycan, was studied in rat prenatal and postnatal development. In adult rat white matter and cerebellum, the distribution of versican was identical to that previously reported for brain-specific glial hyaluronate-binding protein (GHAP). Versican was also found in gray matter where it formed characteristic coats around large neurons. It was also found in peripheral tissues, namely, kidney medulla, myotendinous junctions, and endoneurial and endomysial sheaths. In rat embryo the most notable finding was the presence of large amounts of versican immunoreactive material in precartilaginous mesenchyma. In embryonal CNS, versican was mainly confined to the marginal zone on the surface of the cerebral hemispheres. Versican expression mainly occurred postnatally in brain and spinal cord. In spinal cord white matter, versican immunoreactivity was already present in 3-day-old rats and preceded the appearance of GHAP, which was first detected on day 13 after the onset of myelination. Versican expression was markedly delayed in gray matter. The characteristic perineuronal coats were first observed on day 21 in the cerebral cortex. It is concluded that, with the exception of hyaluronate, brain extracellular matrix (ECM) is mainly produced postnatally and that the ECM protein produced by brain cells, most likely astrocytes, is similar to that produced by precartilaginous mesenchyma.

Rabbit antisera against tenascin, a large extracellular matrix protein, in conjunction with monoclonal antibodies of mouse origin against versican, a large hyaluronate-binding proteoglycan, were used to make a comparative study of the distribution of the two antigens in the same cryostat sections by double immunofluorescence. In the central nervous system, tenascin was invariably associated with versican, but the reverse was not true, in that versican was also found where tenascin was not detectable, particularly in gray matter. There were major species differences in the distribution of tenascin in the central nervous system. In the cow, tenascin was found in cerebral and spinal cord white matter and in the granule cell layer of the cerebellum. In the human brain, tenascin was found in cerebral white matter but not in the cerebellum. In the rat, tenascin was mainly confined to brain periventricular layer and spinal cord white matter. During the development of the cerebellum of the rat, the tenascin immunoreactivity decreased, and a lower molecular weight band appeared (J1-160/180/restrictin?) and persisted throughout adulthood. Tenascin expression was a relatively late event in the development of the rat central nervous system, immunoreactivity being first observed after birth. In the rat embryo, tenascin was found to co-localize with versican in precartilaginous mesenchyme and in connective tissue underlying epithelia. The colocalization of versican with tenascin suggests that versican may be the tenascin (cytotactin)-associated proteoglycan reported in the literature.

Hyaluronic acid (HA) plays the main structural role in the formation of brain extracellular matrix (ECM). The extracellular space appears empty by electron microscopy because HA is readily dissolved during the preparation of tissues for ultrastructural studies. The HA-binding proteins so far identified in brain ECM are versican, aggrecan and the glial HA-binding protein. Versican is a large fibroblast proteoglycan preferentially expressed in embryonic cartilage at the time of mesenchymal condensation. Glial HA-binding protein (GHAP) is probably a proteolytic product of versican corresponding to its HA-binding amino-terminal domain. It is mainly a white-matter protein, suggesting that the proteinase responsible for its cleavage from versican is normally activated in this location. Versican is found in both white matter and gray matter, where it forms pericellular coats around large neurons. Aggrecan, the aggregating proteoglycan of mature cartilage, co-localizes with versican in this location. In white matter, the localization of GHAP and versican is identical to that of the glial fibrillary acid protein, suggesting that both proteins are produced by astrocytes. An important difference between GHAP and versican is that GHAP but not versican is released from the tissues by hyaluronidase digestion, which suggests that versican is anchored to the cell membranes lining the extracellular space. GHAP was localized at the ultrastructural level in the granule cell layer of rat cerebellum, the only region of gray matter that is positive for GHAP in this species. Rats were perfused with aqueous fixatives containing cetylpyridinium chloride or tannic acid to prevent the solubilization of HA. GHAP is found throughout the extracellular space, the synaptic clefts being a notable exception. GHAP appears late in development, and the same is true for versican, the characteristic perineuronal coats first becoming apparent in the third postnatal week. It is suggested that a marked change occurs in the structure of brain ECM when HA-binding proteins first appear, and that the change is similar to that observed in prechondrogenic mesenchyme, i.e., reduction of the extracellular space and cell aggregation.

Hyaluronic acid was localized in acetone-fixed cryostat sections of brain and spinal cord obtained from adult, newborn and embryonal rat. The sections were incubated with glial hyaluronate-binding protein (GHAP) of human origin and the protein was visualized by indirect immunofluorescence with monoclonal antibodies raised to human GHAP and not staining rat brain by immunofluorescence. GHAP is a brain extracellular matrix (ECM) glycoprotein, approximately 60,000 molecular weight, which is structurally related to the HA-binding region of cartilage ECM proteins. The distribution of hyaluronate in adult brain white matter and cerebellar cortex was similar to that previously reported for GHAP. In both cases, the reaction product formed a mesh surrounding myelinated axons and granule cells. Hyaluronate was also found in parts of the brain that did not contain GHAP. A finely reticulated mesh was observed in the neuropil between cell bodies in cerebral cortex and basal ganglia. Scattered cortical neurons were surrounded by a rim of reactive material. Perineural staining was the rule rather than the exception in spinal cord anterior horn motoneurons, inferior olivary nucleus, large bulbar reticular neurons and dentate nucleus of cerebellum. The only part of the brain which appeared relatively free of hyaluronate was the molecular layer of the cerebellum. In newborn and embryonal rat, the densely packed cell bodies in cerebral gray matter, periventricular germinal layer and external granular layer of cerebellum were surrounded by hyaluronate. Small droplets of hyaluronate were observed in between the cylindrical epithelial cells lining the neural tube in 11 day embryos. Non-myelinated fiber tracts and the molecular layer of the developing cerebellum were relatively unstained. No hyaluronate was detected in the ependyma lining the cerebral ventricles and the central canal of the spinal cord.

The distribution of hyaluronate (HA) and chondroitin sulfate (CS) proteoglycan in the rat cerebral cortex was compared. For the localization of HA, the sections were incubated with human glial hyaluronate-binding protein (GHAP) and then reacted with monoclonal or polyclonal antibodies to GHAP. Polyclonal antibodies raised in rabbit were used for double-labeling experiments with monoclonal antibodies raised in mice and reacting with CS proteoglycans. Little reactivity was observed in rat cerebral cortex with polyclonal GHAP antibodies if the sections were not incubated with GHAP. Monoclonal antibodies to GHAP did not react with murine tissues. CS proteoglycans were localized in chondroitinase-digested sections with monoclonal antibodies reacting with the 4-sulfated oligosaccharide stubs formed by the digestion with chondroitinase ABC of CS side chains. In the rat cerebral cortex. the distribution of CS proteoglycans was similar to that reported by Bertolotto, A., Rocca, G. and Schiffer, D., J. Neurol. Sci., 100 (1990) 113-123, and his collaborators using the same antibodies. Many neurons mainly located in the upper and deep cortical layers were surrounded by CS immunoreactive material. Several (but not all) CS-positive neurons also stained for HA with an identical distribution except that in most instances the staining was confined to the periphery of the perikaryon and did not extend to the dendritic tree. The finding suggests that cerebral cortex CS proteoglycan is capable of interacting with HA.

CD44 is an integral membrane glycoprotein of approximately 90 kDa which has been implicated in the binding of hyaluronate to the cell surface. The expression of CD44 in astrocytes was investigated by means of indirect immunofluorescence on cultured cells. The vast majority of these cells were found to express CD44. Western blot analysis of these cells revealed a highly polydisperse species having an Mr corresponding to 74–86 kDa. In order to visualize hyaluronate-binding cells, living cultures were probed with fluorescein-conjugated hyaluronate (FI-HA). Some astrocytes were able to bind FI-HA, provided that they were first treated with hyaluronidase. Streptomyces hyaluronidase, which is hyaluronate-specific, was effective in exposing the hyaluronate-binding capacity of these cells. This leads one to conclude that hyaluronate is bound to the surface of these cells and that it masks their capacity to bind hyaluronate. Provided that they were first treated with hyaluronidase, the U-87 MG (glioblastoma-astrocytoma), U-373 MG (glioblastoma), and Hs 683 (glioma) cell lines were also able to bind FI-HA. The U-138 MG (glioblastoma) cell line was unable to bind FI-HA, with or without prior hyaluronidase treatment. A quantitative assay was developed with the use of [3H]hyaluronate ([3H]HA). This revealed the binding to be highly specific, inasmuch as the addition of unlabeled hyaluronate, but not other glycosaminoglycans, was effective in inhibiting the binding of the [3H]HA. An anti-CD44 monoclonal antibody, 50B4, was able to inhibit the binding of the [3H]HA to the U-373 MG cell line. In this cell line, then, CD44 functions as a hyaluronate receptor and one may infer that this is also the case in some atrocytes.

The localization of hyaluronic acid (HA), glial hyaluronate-binding protein (GHAP), and chondroitin sulfate (CS) proteoglycan was compared in cryostat sections of rat spinal cord. HA, GHAP, and CS proteoglycan were similarly distributed in white matter where they surrounded myelinated axons. In gray matter, large motoneurons were surrounded by a rim of reaction product in sections stained for HA and CS proteoglycan. GHAP immunoreactivity as well as HA had disappeared in hyaluronidase-digested sections, while CS proteoglycan immunoreactivity was not abolished under these conditions.

Glial hyaluronate-binding protein (GHAP) and a large aggregating chondroitin sulfate proteoglycan (Ag-Pg) similar to a fibroblast proteoglycan (versican) were localized in bovine, dog and cat central nervous system (CNS) gray matter by indirect immunofluorescence. The distribution of the two hyaluronate-binding proteins was identical with that of hyaluronate, an extracellular glycosaminoglycan. All substances formed a finely reticulated mesh in the neuropil with a condensation of the stain around large neurons. It is concluded that in gray matter, as in white matter, the extracellular matrix (ECM) contains hyaluronate-protein aggregates. We suggest that the hyaluronate-protein aggregates correspond to the pericellular network first described by Golgi.

The localization of hyaluronate was studied in the CNS of rat, goldfish and lamprey. Cryostat sections were incubated with glial hyaluronate-binding protein of human origin and stained by indirect immunofluorescence with glial hyaluronate binding protein antibodies not reaching with rat and fish. As previously reported for glial hyaluronate-binding protein and glial fibrillary acidic protein, hyaluronate and glial fibrillary acidic protein had a similar distribution in rat spinal cord and optic nerve, both substances forming ring-like structures around individual myelinated axons. A similar periaxonal distribution was observed in goldfish spinal cord and medulla, except that the rings were much wider, to accommodate the large goldfish axons. The glial fibrillary acidic protein-positive neuroglial tissue forming distinctive structures in goldfish vagal lobes also stained for hyaluronate. In both rat and goldfish spinal cord, motoneurons were surrounded by a hyaluronate coat. Goldfish optic nerve and lamprey spinal cord were hyaluronate-negative and, as previously reported, they stained for keratin but not for glial fibrillary acidic protein. The findings suggest that hyaluronate in CNS fibre tracts in a product of glial fibrillary acidic protein-positive neuroglia. They also suggest that the appearance of glial fibrillary acidic protein-positive neuroglia and the formation of a hyaluronate-bound extracellular matrix are related phenomena in phylogeny.

A large proteoglycan (365 kDa), identified with monoclonal antibodies raised against chondroitin sulfate, was isolated from human brain. The isolation required anion-exchange chromatography followed by gel filtration through a Sephacryl S-500 column. The proteoglycan bound specifically to [3H]hyaluronate (HA). The binding was not reduced by high salt concentrations (up to 4 M) and was inhibited at low pH (< 4.0). The binding was inhibited by the octamer and decamer (but not the hexamer) oligosaccharides of HA. Limited proteolysis of the proteoglycan gave rise to a relatively stable polypeptide (80 kDa). The amino-terminal sequence of the 80-kDa polypeptide was identical to the cDNA-derived amino-terminal sequence of versican, a large human fibroblast proteoglycan. A monoclonal antibody raised against bovine proteoglycans and recognizing the versican core protein reacted by immunoblotting with the proteoglycan isolated from human brain. The antibody was used to localize the proteoglycan in acetone-fixed cryostat sections of bovine spinal cord. The localization of the proteoglycan in the central nervous system was identical to that previously reported for glial hyaluronate-binding protein (GHAP), a 60-kDa glycoprotein of the brain extracellular matrix (ECM). However, a major difference was observed with respect to the sensitivity of the two antigens to hyaluronidase. As previously reported, GHAP was released from the tissue by hyaluronidase digestion, whereas the proteoglycan persisted under these conditions. We conclude that the protein-hyaluronate aggregates in brain ECM contain both GHAP and versican, that GHAP is only retained in the ECM by its interaction with hyaluronate, and that the proteoglycan is anchored in some other manner and probably connects cell surfaces with the ECM since it was not released by hyaluronidase digestion.

The effect of the substrate on neurofilament phosphorylation was studied in primary cultures of spinal cord and dorsal root ganglia dissociated from 15-day-old rat embryos. On polylysine and Primaria® substrates, spinal cord neurons formed aggregates connected by bundles of neurites. (Primaria® dishes have a modified plastic surface with a net positive charge.) On both polylysine and Primaria® substrates, spinal cord neurons were stained with neurofilament monoclonal antibodies reacting with phosphorylated epitopes appearing early in rat embryo development, i.e. soon after neurofilament expression. Conversely, immunoreactivity with antibodies recognizing late phosphorylation events was only observed on Primaria® substrates. As reported by many investigators, fibronectin and laminin were excellent substrates for dorsal root ganglia neurons in culture. However, on both laminin and fibronectin substrates immunoreactivity with antibodies recognizing late phosphorylation events, only occurred after several days in culture, at a time when non-neuronal cells (mainly astrocytes) had formed a confluent monolayer.

Over the past five years, we have made some progress in our studies on the composition of brain extracellular matrix. As in previous work on GFA protein, a major component of glial scars, the motivation for these studies was to find out why axons do not regenerate in mammalian CNS. In fact, we started doing research on brain extracellular matrix because the experimental evidence suggested that the glial scar per se, could not explain the riddle of CNS regeneration (Bignami et al., 1986).

Axonless horizontal cells in the outer plexiform layer of rat retina were studied with 19 monoclonal antibodies reacting with phosphorylated and non-phosphorylated epitopes of the two high molecular weight neurofilament proteins (NF 150K and NF 200K). With 6 antibodies, immunoreactivity was confined to the nerve fiber layer on the inner surface of the retina. Horizontal cells were not stained. Four antibodies in this group were axon-specific, while the remaining two stained motor and sensory neuron perikarya in rat spinal cord and dorsal root ganglia, respectively. Of the 13 antibodies which stained horizontal cells, 11 reacted with phosphorylated epitopes and failed to decorate motor neuron perikarya in the spinal cord, while in dorsal root ganglia, they stained a subpopulation of sensory neurons.

A glial hyaluronate-binding protein (GHAP) was isolated from bovine spinal cord and partially characterized. Bovine GHAP consisted of three immunologically related polypeptides with molecular masses of 76, 64, and 54 kDa and isoelectric points of 4.1, 4.2, and 4.4, respectively. Peptide mapping and partial amino acid sequencing showed that all three polypeptides derive from the same protein. The protein was localized immunohistochemically with rabbit antisera in the white matter surrounding the myelinated axons. Sugar analyses indicated that the three polypeptides are glycosylated and the sugar residues account for at least 30% of their weight. After enzymatic deglycosylation, the apparent molecular mass of the bovine GHAP was reduced to 43 kDa. The biochemical properties of bovine GHAP were compared to those of human GHAP. Initial peptide mapping indicated similarities between bovine and human GHAP. Partial amino acid sequencing of bovine GHAP showed a striking identity (up to 90%) with human GHAP and with the hyaluronate binding domain of the large human fibroblast proteoglycan, versican. Bovine and human GHAP were demonstrated to bind specifically to hyaluronic acid (HA) with one protein molecule binding to an average 17 disaccharide repeating units. The binding of bovine and human GHAP was inhibited by oligosaccharides of HA and specifically by the octamer. Salt concentrations of up to 1 M NaCl had very little effect on the binding of the GHAP to HA. The GHAP-HA interaction was pH dependent. Dissociation only took place at low pH (< 3.5). Analysis of several polypeptides derived from GHAP by limited proteolysis allowed us to conclude that one of the tandem repeated sequences is sufficient for HA binding and that the aminoterminal domain (which contains an immunoglobulin-like fold) is not involved in the GHAP-HA-binding event.