WO1995026201A1 - Brevican, a glial cell proteoglycan - Google Patents

Abstract

The present invention provides substantially purified mammalian brevican, which is a chondroitin sulphate proteoglycan produced by glial cells. The invention provides, for example, the full-length 145 kDa form brevican as well as an 80 kDa active fragment of bovine brevican and of rat brevican. The invention also provides antibodies that are specific for a mammalian brevican and a cell line that produces a monoclonal antibrevican antibody. The invention further provides nucleic acid molecules encoding a mammalian brevican, vectors containing the nucleic acid molecules, host cells containing the vectors. The invention also provides nucleotide sequences that hybridize to a nucleic acid molecule encoding brevican. The invention also provides methods for detecting the presence of a glial cell in a cell sample suspected of containing a glial cell. The invention further provides methods for detecting a glial cell in a subject and methods for directing axonal growth or inhibiting axonal growth in a subject.

Description

BREVICAN, A GLIAL CELL PROTEOGLYCAN

BACKGROUND OF THE INVENTION

FIELD OF THE INVENTION

This invention relates generally to the fields of molecular biology and protein biochemistry and more specifically to a chondroitin sulphate proteoglycan that is expressed in glial cells.

BACKGROUND INFORMATION

The brain is a complex organ consisting of about 100 billion neurons as well as a variety of supporting cells including glial cells, which produce diverse factors that promote neuronal and neuritic growth. Proteoglycans are involved in a variety of developmental processes of the nervous system. Dynamic changes in the distribution of proteoglycans and glycosaminoglycans (GAG's) occur in several different histogenic processes of the central and peripheral nervous system. In addition, proteoglycans and

GAG's modulate the adhesion, migration and neurite extension of neural cells in vitro . Chondroitin sulfate and keratan sulfate proteoglycans may be crucial active components of the astroglial barriers against axonal outgrowth.

In order to exploit the developmental and reparative functions of the central nervous system, a need exists to identify the critical brain CSPG's involved in regulating these activities. The present invention satisfies this need and provides additional advantages as well.

SUMMARY OF THE INVENTION

The present invention provides substantially purified mammalian brevican, which is a chondroitin sulphate proteoglycan produced by glial cells. The invention provides, for example, the full-length 145 kDa form of bovine or rat brevican as well as an 80 kDa active fragment of brevican. In addition, the invention provides antibodies that are specific for mammalian brevican and a cell line that produces a monoclonal anti-brevican antibody. The antibodies are useful for identifying the presence of brevican in a cell sample and for purifying brevican from a sample containing brevican.

The invention also provides nucleic acid molecules encoding a mammalian brevican, vectors containing the nucleic acid molecules and host cells containing the vectors. The nucleic acid molecules are useful for producing substantially purified brevican or an active fragment thereof using methods of recombinant DNA technology. The invention further provides nucleotide sequences that hybridize to a nucleic acid molecule encoding brevican. The nucleotide sequences are useful as probes to detect the presence of a nucleic acid encoding brevican in a sample obtained from a cell.

The invention also provides methods for detecting the presence of a glial cell in a cell sample suspected of containing a glial cell. The method is useful for detecting, for example, gliosis at a site of nerve cell injury in a subject. The invention further provides methods for detecting a glial cell in a subject and methods for directing axonal growth or inhibiting axonal growth in a subject.

BRIEF DESCRIPTION OF FIGURES

Figure 1. SDS-PAGE analysis of total proteoglycan fraction from brain and the purified 80 kDa core protein. The total proteoglycan fraction was isolated by anion-exchange chromatography from the soluble brain extracts, digested with chondroitinase ABC, then analyzed in a 8-16% gradient gel. Undigested total proteoglycan fraction (lane 1); proteoglycan fraction digested with 50 mU (milliUnits; lane 2), 10 mϋ (lane 3) or 2.5 mϋ (lane 4) chondroitinase ABC; 2.5 iriU chondroitinase ABC without proteoglycan fraction (lane 5); HPLC-purified 80 kDa core protein (lane 6). 2.5 μg proteoglycan samples (measured as protein) were loaded in lanes 1 to 5. Protein were visualized by silver staining.

Figure 2. Specificity of the adsorbed multispecific antiserum and antibodies that were affinity- selected using clones #5 and #17. Immunoblot analysis of undigested (lane 1) and chondroitinase ABC-digested (lanes 2 to 5) total proteoglycan fraction. Samples were reacted using multispecific antiserum adsorbed with E. coli extracts (lane 1) or buffer C-eluates (lane 2) or using antibodies affinity-selected with clone #5 (lane 3), with clone #17 (lane 4) or with clone #13 (lane 5; control).

Figure 3. Complete cDNA sequence (SEQ ID NO: 1) and derived amino acid sequence (SEQ ID NO: 2) of bovine brevican core protein. The amino acid sequence of brevican was derived from the nucleotide sequence of two overlapping cDNA clones (#5, nucleotide positions 1 to 2145, and #17, nucleotide positions 1360 to 3257; SEQ ID NO: 1). The arrow indicates the putative signal peptide cleavage site and the arrowhead indicates the N-terminus of the 80 kDa protein. Amino acid sequences obtained from the N-terminus and from tryptic fragments of the purified 80 kDa protein are underlined. An "(X)" under a peptide sequence indicates an amino acid residue that could not be determined unequivocally. Ser-Gly and Gly-Ser dipeptides are boxed using a thin line and the three dipeptides most likely to be substituted by GAG chains are boxed using a thick line. A "•" indicates a potential N-linked sugar chain attachment site. Cysteine residues are circled. The polyadenylation signal is indicated in bold print. Figure 4. The predicted domain structure of mammalian brevican and comparison with other members of the aggrecan/versican family of proteoglycans. Boxes represent putative functional domains shared by all the members of the aggrecan/versican family.

Figure 5. Immunochemical characterization of brevican with specific antibodies. Total brain soluble extract (lane 1), buffer C-eluates from the DEAE-Sepharose column (lane 2) and undigested (lane 3) and chondroitinase ABC-digested (lane 4) total proteoglycan fraction were examined by immunoblot analysis using anti-brevican antibodies that were affinity-selected with clone #5.

Figure 6. Northern blot analysis of brevican RNA. Total RNA (10 μg/lane) from bovine brain (lane 1), heart (lane 2), lung (lane 3), spleen (lane 4), rat cerebral primary neurons (lane 5) and cerebellar primary astrocytes (lane 6) was fractionated by electrophoresis, transferred to a nylon membrane and hybridized with a probe consisting of either a ³²P-labelled 545 bp EcoRI fragment (lanes 1-4) or a 581 bp Pst I-Sal I fragment (lanes 5 and 6) of the nucleic acid of SEQ ID NO: 1. Prior to transfer, the gel was examined by ethidium bromide staining to ensure that equivalent amounts of total RNA were loaded in each lane (not shown) . Positions of RNA molecular size markers (Gibco-BRL) are indicated (kilobases) .

Figure 7. Comparison of the derived amino acid sequences for bovine brevican (SEQ ID NO: 2) and rat brevican (SEQ ID NO: 3).

The upper sequence (bBRE) is the amino acid sequence encoding bovine brevican (as shown in Figure 3).

The lower sequence (rBRE) represents the rat brevican amino acid sequence (SEQ ID NO: 3). In the lower sequence, a "."

(dot) indicates the amino acid is the same as shown in the upper bovine brevican sequence, a letter indicates the one letter amino acid code for a residue that is different from the corresponding residue in the bovine brevican sequence and a "-" (dash) indicates the amino acid has not yet been determined.

Figure 8. Immunolocalization of brevican in primary cultures of astrocytes and in cerebellar granule neuron-astrocyte co-cultures.

Panels A to D. Immunolocalization of brevican in primary cultures of astrocytes. Subconfluent monolayers of astrocytes were stained with anti-rat brevican IgG (panel B) or preimmune IgG (panel D) . Brevican immunoreactivity was visualized with rhodamine-conjugated goat anti-rabbit IgG (KPL) . Panels A and C are phase contrast views of the fields shown in panels B and D, respectively.

Panels E to G. Immunolocalization of brevican in cerebellar granule neuron-astrocyte co-cultures. Neurons were cultured on monolayers of astrocytes for 4 days, then double-stained with anti-brevican IgG (panel F) and fluorescein-conjugated tetanus toxin (panel G) . Brevican immunoreactivity was visualized with rhodamine-conjugated anti-rabbit IgG (KPL) . Panel E shows a phase contrast view of the fields shown in panels F and G.

DETAILED DESCRIPTION OF THE INVENTION

The present invention provides a substantially purified mammalian brevican, which is a chondroitin sulphate proteoglycan (CSPG) that is produced by glial cells. Unless specifically indicated otherwise, the term "brevican" is used in its broadest sense to include the full-length 145 kDa form of the protein as well as an active fragment such as the 80 kDa form of brevican (see Figure 3, SEQ ID NO: 2 , and Figure 1 , SEQ ID NO: 3), which also can occur in a glial cell. Brevican is referred to herein as a "protein," "core protein" and "proteoglycan." Although the specific meanings of these terms are well known in the art, since brevican can exist in each of these forms, the terms "protein," "core protein" and "proteoglycan" are used interchangeably.

As used herein, the term "substantially purified" means brevican protein that is in a form that is relatively free from contaminating lipids, proteins, nucleic acids or other cellular material normally associated with a protein in a cell. A substantially purified brevican protein can be obtained, for example, by using the purification methods described herein or by expressing a nucleic acid sequence encoding brevican. The term "substantially purified" also can refer to a nucleic acid sequence encoding brevican, as discussed below.

The present invention also provides an amino acid sequence for bovine brevican as shown in Figure 3 (SEQ ID NO: 2) and for rat brevican as shown in Figure 7 (SEQ ID NO: 3). The invention also provides amino acid sequences that are substantially the amino acid sequences shown in Figures 3 and 7 such as amino acid sequences of other mammalian brevicans such as human brevican. In general, an amino acid sequence having at least 65% sequence homology with the amino acid sequence of Figure 3 (SEQ ID NO: 2) or of Figure 7 (SEQ ID NO: 3) is considered substantially the same sequence. Thus, a mammalian brevican is characterized, in part, by having a greater homology with another mammalian brevican such as human brevican than with another member of the aggrecan/versican family of CSPG's.

As used herein, the term "substantially the amino acid sequence" includes a sequence having conservative amino acid substitutions for the amino acids shown in Figure 3 or Figure 7. A conservative amino acid change can include, for example, the substitution of one acidic amino acid for another acidic amino acid, of one hydrophobic amino acid for another hydrophobic amino acid or other conservative changes known in the art. A conservative amino acid substitution can be characterized, in part, by not significantly altering the secondary structure or the function of brevican as described herein. Conservative amino acid substitutions are found in rat brevican, for example, as compared to bovine brevican. In addition, an amino acid comprising a portion of the sequence shown in SEQ ID NO: 2 is considered to be encompassed within the invention (see below).

More than a dozen different proteoglycans are expressed in the developing and adult brain, including the chondroitin sulfate proteoglycans (CSPG's), which are distributed primarily in extracellular spaces of the brain. Some brain CSPG's, including neurocan, which is a developmentally regulated CSPG that shares functional domains with the cartilage CSPG aggrecan and the fibroblast CSPG versican, and NG2, which is a unique membrane-bound CSPG expressed in 02A glial progenitor cells, have been cloned (Rauch et al., J. Biol. Chem. 267:19536-19547 (1992), which is incorporated herein by reference; Nishiyama et al., J. Cell. Biol. 114:359-371 (1991)). Versican was cloned from a human fibroblast library and also is present in the brain (Zimmermann and Ruoslahti, EMBO J. 8:2975-2981 (1989). In addition, other brain CSPG's have been identified using biochemical and immunochemical methods. For example. Cat 301 is a neuronal cell surface CSPG that was identified using a monoclonal antibody and is related to aggrecan (Zaremba et al.. Neuron 2:1207-1219 (1989); Fryer et al., J. Biol. Chem. 267:9874- 9883 (1992). The overall structure of a mammalian brevican such as bovine or rat brevican is similar to the structures of aggrecan, versican and neurocan (see Tables I and II). Based on these similarities, brevican can be categorized as a member of the aggrecan/versican family of CSPG's. The aggrecan/versican family of CSPG's is characterized, in part, by having a hyaluronic acid-binding domain consisting of an immunoglobulin-like loop, two link protein-like tandem repeats, one copy of EGF-like repeat, a lectin-like domain and a complement regulatory protein-like domain (see Figure 4). In addition, each member of the aggrecan/versican family of CSPG's contains a central non- homologous domain of varying length.

Bovine brevican and rat brevican have approximately 90% homology in their conserved domains. In addition, brevican has 55-59% sequence identity with the conserved domains present in the other members of the aggrecan/versican family. In bovine brevican, the hyaluronic acid-binding domain immunoglobulin-like loop consists of amino acid positions 35 to 158, the two link protein-like tandem repeats consist of amino acids 159 to 353, the EGF-like repeat consists of amino acids 649 to 684, the lectin-like domain consists of amino acids 685 to 814 and the complement regulatory protein-like domain consists of amino acids 815 to 875 (see Figure 3; see, also, Figure 4). The central region of brevican (amino acids 354 to 648) has only minimal homology with the corresponding regions in the other aggrecan/versican CSPG's; this region is considered herein to be nonhomologous among the aggrecan/versican family of CSPG's.

The nonhomologous central region of mammalian brevican is much shorter than the corresponding regions in rat aggrecan, human aggrecan, human versican and rat neurocan (1213, 1491, 1693 and 595 amino acids, respectively) . In brevican, the central nonhomologous region contains a high concentration of glutamic acid, including a cluster of eight consecutive glutamic acid residues. Human versican contains a similar cluster of glutamic acid residues on the C-terminal side of the link protein-like domain at residues 400 to 408 (Zimmermann and Ruoslahti, EMBO J. 8:2975-2981 (1989), which is incorporated herein by reference) . The β amyloid precursor protein of Alzheimer's disease, which exists as a CSPG in certain cell lines, also contains a stretch of acidic residues (Kang et al.. Nature 325:733-736 (1987)). This cluster of acidic residues can be involved in specific biological functions such as binding cationic factors or minerals. For example, a glutamic acid cluster in bone sialoprotein can bind hydroxyapatite (Oldberg et al., Jj.. Biol. Chem. 263:19430-19432 (1988).

The bovine brevican core protein contains 11 Ser- Gly and Gly-Ser dipeptide sequences, which are associated with GAG attachment. An efficient GAG attachment site also may require acidic residues flanking the dipeptides. In bovine brevican, Ser-418, Ser-576 and Ser-588 (Figure 3) meet these requirements and, therefore, can be binding sites for GAG chains. The GAG attachment sites on the brevican core protein can be predicted based on known GAG attachment sequences present in other proteoglycans (Bourdon et al., Proc. Natl. Acad. Sci.. USA 84:3194-3198 (1987), which is incorporated herein by reference). However, brevican core protein, i.e., without GAG chains, also is present in brain. Immunoblot experiments using total brain extracts showed that brevican is a "part-time" proteoglycan (see Figure 5), similar to NG2 in the brain (Stallcup et al.. Cold Spring Harbor Symp. Quant. Biol. 48:761-774 (1983)).

Brevican core protein was isolated from adult bovine brain and from rat brain as an intact 145 kDa form and a cleaved 80 kDa active fragment thereof. The methods disclosed herein describe the purification of bovine brevican. However, the same methods were used to successfully obtain substantially purified rat brevican. The presence of cleaved forms of the various members of the aggrecan/versican family of CSPG's is not uncommon in brain. For example, a cleaved 150 kDa form of neurocan is present in adult rat brain (Rauch et al., J. Biol. Chem. 266:14785-14801 (1991). Neurocan cleavage is developmentally regulated and occurs in the nonhomologous central region, similar to brevican. Also, a 60 kDa N- terminal fragment of versican exists in the brain as the glial hyaluronic acid-binding protein (Zimmermann and Ruoslahti, 1989).

The invention provides active fragments of mammalian brevican. As used herein, the term "active fragment" means a protein consisting of less than the full length brevican shown, for example, in Figure 3 (SEQ ID NO: 2) . An active fragment of brevican can have an activity that is characteristic of brevican or can be an epitope that is specifically recognized by an anti-brevican antibody as described herein. The 80 kDa form of brevican, which is a cleaved form of the 145 kDa brevican, provides an example of an active fragment of brevican that can interact with a carbohydrate ligand in the brain. Since brevican consists of well characterized domains such as a hyaluronic acid binding domain and a lectin-like domain (see Figure 4), these isolated domains can be functional in isolated form and, therefore, are further examples of active fragments of brevican. Significantly, the 80 kDa brevican does not contain a hyaluronic acid binding domain, which is characteristic of other members of the aggrecan/versican family of CSPG's. Nevertheless, the 80 kDa form is considered an active fragment of brevican since it can interact, for example, with a carbohydrate ligand. An active fragment of mammalian brevican such as the 80 kDa fragment is characterized by having at least 80% homology with a portion of the amino acid sequence of SEQ ID NO: 2 or SEQ ID NO: 3 and by having an activity that is associated with brevican. The 80 kDa form of brevican is an example of an active fragment of brevican. In addition, an amino acid sequence having, for example, at least 80% homology with the amino acid sequence of residues 685 to 814 of SEQ ID NO: 2 and the ability to bind a lectin is considered an active fragment of brevican. An active fragment of brevican can be obtained using well known recombinant DNA methods and can be characterized using methods described herein.

The present invention also provides antibodies specific for a mammalian brevican. Active fragments of antibodies such as Fv, Fab and Fab'- fragments that are specifically reactive with brevican are included within the definition of an "antibody" as used herein. As used herein, the term "antibody" includes, for example, chimeric antibodies, which can be mouse/human chimeras such as humanized mouse monoclonal antibodies, bifunctional or heterofunctional antibodies, which can contain antigen binding sites from two or more antibodies, or CDR-grafted antibodies. Antibodies exhibiting a titer of at least about 1.5 x 10⁵ as determined, for example, by ELISA are useful in the present invention. An anti-brevican antibody can be characterized by its ability to bind a portion of a mammalian brevican protein, including a portion of the 80 kDa form of brevican, the 145 kDa form of brevican or both forms of brevican.

The antibodies of the invention can be produced and characterized as described herein or by any method known in the art. For example, an anti-bovine brevican antibody was obtained by affinity selection of a multispecific antiserum with a brevican-expressing phage and was characterized using immunoblot analysis (see Example II) . Polyclonal and monoclonal antibodies also can be produced by methods described in Harlow and Lane, Antibodies: A Laboratory Manual (Cold Spring Harbor Laboratory Press 1988), which is incorporated herein by reference. For example, an anti-rat brevican antibody was obtained by immunizing rabbits with a substantially purified rat 80 kDa form of brevican and affinity selecting anti-rat brevican antibodies against immobilized 80 kDa brevican (see Example IV). Thus, a mammalian brevican, including portions of the amino acid sequence of Figure 3 (SEQ ID NO: 2) or Figure 7 (SEQ ID NO: 3), can be used as an immunogen to generate such antibodies.

Antibodies also can be produced, for example, by a hybridoma cell line, by chemical synthesis or using recombinant methods (see Sambrook et al. , Molecular

Cloning: A laboratory manual (Cold Spring Harbor

Laboratory Press 1989), which is incorporated herein by reference) . Chimeric, humanized, CDR-grafted and bifunctional antibodies can be produced using methods well known to those skilled in the art (see, for example, Harlow and Lane, 1988, and Hilyard et al., In Protein Engineering: A practical approach (IRL Press 1992), which is incorporated herein by reference) .

An anti-brevican antibody can be useful to substantially purify brevican from a sample containing brevican. For example, an anti-brevican antibody can be attached to a solid support such as a gel chromatography matrix, added to a column and used to affinity purify brevican from a sample containing brevican that is added to the column. Methods for attaching an antibody to a solid support matrix and purifying an antigen by affinity chromatography are known in the art and described, for example, by Harlow and Lane (1988). An anti-brevican antibody can be detectably labelled using methods well known in the art (see, for example, Harlow and Lane, 1988; chap. 9; see, also. Example IV and Figure 8) . An antibody can be detectably labelled by attaching any of a variety of moieties, including biotin, an enzyme such as alkaline phosphatase, a fluorochrome or a radionuclide, such as technicium-99 or iodine-125. Methods for selecting and attaching a particular moiety are known in the art and described, for example by Harlow and Lane (1988).

Following contact of a labelled antibody with a sample, specifically bound antibody can be identified by detecting the particular moiety. If desired, a labelled second antibody can be used to identify specific binding of an unlabelled first antibody such as an anti-brevican antibody. A second antibody is specific for the particular class of the first antibody. For example, if an anti- brevican antibody is of the IgG class, a second antibody will be an anti-IgG antibody (see Figure 8). Such second antibodies can be produced as described above or can be purchased from commercial sources. The second antibody can be labelled as described above. When a sample is labelled using a second antibody, the sample is first contacted with a first antibody, then the sample is contacted with the labelled second antibody, which specifically binds to the first antibody and results in a detectably labelled sample.

The present invention also provides substantially purified nucleic acid sequences that encode mammalian brevican or an active fragment of brevican. As used herein, the term "substantially purified" means that the nucleic acid is relatively free from contaminating materials such as lipids, proteins, carbohydrates or cellular material normally associated with a nucleic acid in a cell. For example, a nucleic acid sequence that is chemically synthesized or is produced using recombinant DNA methods is considered substantially purified. Recombinant DNA methods for producing a substantially purified nucleic acid are well known in the art and include cloning a sequence or polymerase chain reaction (PCR) amplification of a sequence (see Sambrook et al., 1989; see, also, Erlich, PCR Technology: Principles and applications for DNA amplification (Stockton Press 1989), which is incorporated herein by reference) .

The invention provides a nucleic acid sequence encoding, for example, the 80 kDa form or the 145 kDa form of bovine brevican as shown in Figure 3 (SEQ ID NO: 1) or a nucleic acid sequence that is substantially the same nucleic acid sequence as shown in SEQ ID NO: 1. As used herein, the term "substantially the same nucleic acid sequence" means a sequence that can contain, for example different nucleotides than shown in Figure 3 but that, as a result of the degeneracy of the genetic code, encodes the same amino acid sequence as shown in SEQ ID NO: 2. A nucleic acid sequence that encodes substantially the same amino acid as shown in SEQ ID NO: 2 also is encompassed within the invention. Thus, a nucleic acid encoding, for example, rat brevican, which has substantially the same amino acid sequence as bovine brevican, is considered substantially the same nucleic acid sequence as shown in SEQ ID NO: 1. A nucleic acid sequence of the invention also can encode, for example, an active fragment of brevican, which can contain conservative amino acids changes as compared to a portion of SEQ ID NO: 2 or SEQ ID NO: 3.

The invention also provides vectors comprising a nucleic acid molecule encoding mammalian brevican and host cells containing the vector. Vectors are well known in the art and include, for example, cloning vectors and expression vectors, as well as plasmids or viral vectors (see, for example, Goedell, Methods in Enzymology- vol. 185 (Academic Press 1990), which is incorporated herein by reference) . Expression vector comprising a nucleic acid sequence encoding brevican can be particularly useful for expressing large amounts of brevican protein, which can be purified and used as an immunogen to raise anti-brevican antibodies or can be administered to a subject. , A baculovirus vector is an example of a vector that can be used to express large amounts of brevican. Expression vectors also can be useful for expressing an antisense nucleic acid, which is complementary to a nucleic acid sequence encoding brevican, or a ribozyme, which can be specific for brevican RNA. A vector comprising a nucleic acid sequence encoding mammalian brevican can further comprise a promoter or enhancer element, which can be constitutive or inducible and, if desired, can be tissue specific. Host cells also are known in the art and can be selected based on the particular vector.

The invention also provides nucleic acid sequences that can hybridize to a brevican-encoding nucleic acid sequence under stringent hybridization conditions. Such hybridizing sequences should be at least ten nucleotides in length and can be prepared, for example, by restriction endonuclease digestion of a cloned nucleic acid sequence encoding brevican or PCR amplification of a portion of the nucleic acid sequence shown in Figure 3 (SEQ ID NO: 1) or can be chemically synthesized using well known methods. The hybridizing nucleic acid sequences can be detectably labelled and used as probes or can be used as primers for PCR. Such PCR primers were useful, for example, to obtain cDNA sequences encoding rat brevican (not shown) . Methods for detectably labelling a nucleic acid are well known in the art (see, for example, Sambrook et al., 1989; see, also, Ausubel et al., Current Protocols in Molecular Biology (John Wiley & Sons 1987), which is incorporated herein by reference) . A unique strategy was used to clone the cDNA sequences encoding bovine brevican and rat brevican. Multiple cDNA clones encoding proteoglycan core proteins were isolated using a multispecific polyclonal antiserum that recognizes multiple core proteins. The multispecific antiserum was used to screen a phage expression library. A proteoglycan-rich fraction was obtained by DEAE anion- exchange chromatography of brain tissue homogenate and was used to obtain the multispecific antiserum. Chemically deglycosylated CSPG's were used for immunization so that the antiserum would recognize bacterially-expressed recombinant proteins produced in a λgtll cDNA library. Antibodies that reacted with nonproteoglycan components were removed by adsorbing the antiserum on various affinity resins coupled with material that did not bind the DEAE column or was weakly bound.

With regard to the bovine brevican, three groups of cDNA species were cloned from bovine brain using the disclosed strategy. Each of the cloned groups of cDNA encoded proteoglycan core proteins and one of these groups (clones #5 and #17) encoded the 145/80 kDa bovine brevican core protein as described herein. Affinity selection experiments confirmed that the 145 kDa protein is the full- length form of bovine brevican encoded by the isolated cDNA. As shown in Figure 2, antibodies that were affinity- selected against the protein expressed by clones #5 (lane 3) and #17 (lane 4) recognized both the 145 kDa and 80 kDa forms of brevican, whereas antibodies selected against an irrelevant clone (#13; lane 5) did not bind to brevican. Similar results were obtained for rat brevican (see Example IV, which describes results obtained using anti-rat brevican antibodies) .

The 145 kDa and 80 kDa molecular masses for the core protein brevicans were determined using SDS-PAGE. Similar molecular masses were observed for rat brevicans examined by SDS-PAGE and western blot analysis (see Example IV) . The experimentally determined molecular masses for bovine brevican are larger than the molecular weight calculated based on the derived amino acid sequence (97,194 and 56,087, respectively). However, the ratio of experimentally determined molecular masses of these proteins (1:0.55) is in good agreement with that from derived amino acid sequence (1:0.58). The derived amino acid sequence was determined based on the nucleic acid sequence of Figure 3 (SEQ ID NO: 1). An aberrant migration pattern has been reported for other proteoglycan core proteins, including neurocan (Rauch et al., (1992).

Northern blot analysis detected only a single transcript for bovine brevican (Figure 6) and rat brevican (not shown). Thus, the two different forms of the core protein are not the result of alternative splicing of brevican mRNA. Brevican mRNA is expressed in brain tissue, but not in lung, heart or spleen (Figure 6). Thus, brevican, like neurocan, is expressed primarily in the brain. The 3.3 kb bovine and rat brevican mRNA is much smaller than the other members of the aggrecan/versican family of CSPG's. For example, rat aggrecan has two mRNA transcripts of 8.2 kb and 8.9 kb, human versican has multiple transcripts ranging from 8-10 kb and rat neurocan has a single 7.5 kb transcript.

Brain CSPG's appear to regulate various processes such as axonal outgrowth during development (see Snow et al., Devel. Biol. 138:359-376 (1990) and Brittis et al.. Science 225:733-736 (1992), each of which is incorporated herein by reference) and maturation of synapses (Hockfield et al.. Cold Spring Harbor Symp. Quant. Biol. 15:505-514 (1990), which is incorporated herein by reference). The functional activities likely are due to the characteristic combination of structural domains that are conserved in brevican and the other members of the aggrecan/versican family of CSPG's. For example, the N-terminal domains of aggrecan and versican can bind hyaluronic acid and the lectin-like domain of aggrecan can bind carbohydrates (Saleque et al., Glvcobiology 3:185-190 (1993), which is incorporated herein by reference) .

Depending on the topologic orientation on a neural cell surface and in brain ECM, a CSPG can link hyaluronic acid and carbohydrate in the brain tissue. For example, a CSPG can associate with a neural cell surface through the lectin-like domain and the hyaluronic acid- binding domain can interact with hyaluronic acid in the extracellular space. In addition, the hyaluronic acid- binding domain can associate with cell surface hyaluronic acid captured by a putative "hyaluronic acid receptor." Similarly, there can be cell surface ligands for the lectin-like domain that are structurally related to the mucin-like ligands for L-selectin.

Since the aggrecan/versican family of CSPG's share N-terminal and C-terminal domains, the central nonhomologous regions and the number of GAG attachment sites can mediate the different physiologic functions associated with these proteoglycans. Aggrecan, versican, neurocan and brevican core proteins are estimated to have 100, 20, 7 and 3 GAG chains, respectively. Thus, these different CSPG's can introduce different amounts of chondroitin sulfate into a tissue, which can produce qualitative difference in the brain ECM such as tissue water content and the volume of extracellular spaces. Moreover, chondroitin sulfate can be an active component of the astroglial barriers for axonal outgrowth (Snow et al., 1990; Brittis et al., 1992). Since brevican is expressed at a high level in cultured astrocytes, brevican can function as the barrier molecule. Thus, the spatial and temporal expression of these CSPG's may play a critical role in the growth and guidance of axons, in some cases directing axonal growth and in others inhibiting growth. An understanding of the expression of brevican and other members of the aggrecan/versican family during development and repair of the nervous system can allow for the manipulation of the level of expression of these proteoglycans and, therefore, can provide a method for directing axonal growth.

Brevican is expressed primarily in glial cells as compared to neuronal cells in the central nervous system (see Example IV and Figure 8). Expression of brevican can regulate axonal extension during development and repair of the nervous system. For example, brevican is present along neurites, which provide an in vitro model for axon growth. Neurites are axon-like processes that form when neuronal cells are cultured under prescribed conditions (Nobel et al., J. Neurosci. 4:1892-1903 (1984); Neugebauer et al., J. Cell Biol. 107:1177-1187 (1988), each of which is incorporated herein by reference) . In normal development, glial cells can function as a substrate that forms a pathway for neuronal migration and axonal guidance. Similarly, monolayers of primary astrocytes in vitro mimic the in vivo function of glial cells and can act as a substrate for neurite outgrowth of neurons.

The invention provides methods for identifying the presence of a glial cell in a cell sample, comprising detecting the presence of brevican in the cell sample, where the presence of brevican indicates that a glial cell is present in the sample. As used herein, the term "cell sample" means, for example, a tissue explant such as a brain tissue explant, which can be placed in a culture medium and incubated in vitro (see Example IV). A cell sample also can be a biopsy sample obtained from a subject that is being examined, for example, for gliosis, which can occur after neuronal injury in the nervous system. In order to detect the presence of glial cells in a cell sample, the cell sample can be contacted with a reagent such as an antibody that specifically binds to brevican (see Example IV and Figure 8) . Specific binding of an anti-brevican antibody indicates that brevican and, therefore, glial cells are present in the cell sample. Where the cell sample consists of a tissue explant in culture, the presence of glial cells also can be identified using, for example, an enzyme-linked immunosorbent assay (ELISA) to detect and, if desired, quantitate brevican. Cells can be attached to a tissue culture plate, then contacted, for example, with a fluorescently labelled anti- brevican antibody (see Figure 8) . Specific binding of the antibody indicates that brevican and, therefore, glial cells are present in the cell sample. Where the cell sample is a biopsy sample obtained from a subject, the cell sample can be prepared for immunohistological examination and the presence of glial cells can be identified using an anti-brevican antibody as described above.

Brevican was detected in cultured primary astrocytes (Figure 8, panels A to D) and in a cerebellar granule neuron-astrocyte co-culture (Figure 8, panels E to G) . The pattern of brevican distribution in these cultured cells indicates that brevican can be involved in the neuron-glial cell interaction. The presence of brevican along neurites indicates that brevican can have a role in regulating axon extension or directing axon extension along a particular route. For example, brevican, like other CSPG's, can form an axon barrier, which tightly regulates the route of axon extension during development and prevents axon extension into particular regions of the developing brain. The presence of brevican along extending axons confirms the usefulness of brevican as a marker for axonal growth and extension in the nervous system. In addition, since brevican is more abundant in adult brain as compared to developing brain, brevican can be involved in maintaining axon structure in the nervous system.

A cell sample also can be processed to obtain, for example, RNA from the cells. The presence of RNA encoding brevican can be determined using well known methods such as northern blot analysis (Example IV) or methods of PCR analysis (see, for example, Ehrlich, 1989). Oiigonucleotide probes or PCR primers can be chemically synthesized based on the nucleic acid sequence for brevican as shown in Figure 3. A nucleic acid sequence encoding a mammalian brevican or a specific oiigonucleotide encoding a portion of a brevican can be detectably labelled using well known methods as described above.

The invention also provides of detecting the presence of a glial cell in a subject, comprising administering a detectably labelled reagent that binds to brevican present in the subject and detecting specific binding of the reagent. For example, the invention provides a method of diagnostic imaging, which can detect a region of gliosis in a subject. Gliosis can occur in response to an injury to the nervous system and can resulting in scarring, which can prevent the return of normal function following nerve injury. The presence of gliosis can be identified by detecting the presence of a relatively excessive amount of brevican in an injured area. In addition, the disclosed method can detect the presence of a relatively excessive amount of brevican, which can occur due, for example, to production of brevican by a glial cell cancer such as a gliosarcoma. The ability to detect the presence of normal or tumorigenic glial cells at an early stage of glial cell infiltration in an area can allow early clinical intervention if necessary.

For diagnostic imaging of glial cells that express brevican in a subject, a reagent such as an anti- brevican antibody can be detectably labelled with an appropriate radionuclide such as technicium-99. A pharmaceutical composition comprising the detectably labelled reagent and a pharmacologically acceptable carrier can be administered to a subject by various methods including, for example, intravenous injection or infusion. Pharmaceutically acceptable carriers are well known in the art and include aqueous solutions such as physiologically buffered saline or other solvents or vehicles such as glycols, glycerol, oils such as olive oil or injectable organic esters. A pharmaceutically acceptable carrier can contain physiologically acceptable compounds that act, for example, to stabilize an anti-brevican antibody. Physiologically acceptable compounds include, for example, carbohydrates, such as glucose, sucrose or dextrans, aritioxidants, such as ascorbic acid or glutathione, chelating agents, low molecular weight proteins or other stabilizers or excipients. The presence of a glial cell can be identified by detecting the localization of specifically bound reagent using known imaging techniques.

The present invention also provides methods of administering brevican to a subject to direct or inhibit axonal growth. As used herein, the term "direct axonal growth" refers to the biological activity of brevican to guide axonal growth in a particular direction into a region. As used herein, the term "inhibit axonal growth" refers to the biological activity of brevican to act as an axonal barrier, which prevents extension of an axon into a region.

An effective amount of a pharmaceutical composition comprising brevican and a pharmacologically acceptable carrier can be administered to a subject having a nerve injury in order to direct axonal growth in the site of injury. One skilled in the art would know that the choice of a pharmaceutically acceptable carrier can depend. for example, on the site of administration of the brevican and the extent of the injured area. An effective amount of a pharmaceutical composition comprising brevican is an amount that can provide a proper milieu for axonal growth. In general, the pharmaceutical composition will be administered locally to a site of injury through injection or by means of a mechanical device such as a subdermal pump.

A pharmaceutical composition comprising brevican and a pharmacologically acceptable carrier also can be administered to a subject in order to inhibit axonal growth. For example, a disease such as neurofibro atosis is characterized by pathological nerve growth, which results in the formation of neuromas and neurofibromas in a subject. An effective amount of a pharmaceutical composition comprising brevican can be administered to the site of a developing neuroma or neurofibroma and can inhibit nerve growth by forming an axonal barrier.

The following examples are intended to illustrate but not limit the present invention. Thus, while the examples generally describe methods for obtaining substantially purified bovine brevican and nucleic acids encoding bovine brevican, essentially the same methods were used to obtain substantially purified rat brevican and nucleic acids encoding rat brevican.

EXAMPLE I

PREPARATION OF A MULTISPECIFIC ANTISERUM FOR CHONDROITIN SULPHATE PROTEOGLYCAN CORE PROTEINS

This example provides a method for obtaining a multispecific antiserum that recognizes deglycosylated chondroitin sulphate proteoglycan (CSPG) core proteins. A. Isolation of total CSPG's from bovine brain extracts:

In order to raise rabbit antisera recognizing multiple CSPG core proteins, a total soluble proteoglycan fraction was isolated from fresh adult bovine brains and the proteoglycans were deglycosylated. Proteoglycans were extracted from brain tissue and fractionated essentially as described by Herndon and Lander, Neuron 4:949-961 (1990), which is incorporated herein by reference. Briefly, 100- 150 g (wet weight) of brain tissue were homogenized in 9 vol ice-cold buffer A (0.3 M sucrose, 4 mM Hepes, pH 8.0,

0.15 M NaCl) containing protease inhibitors (1 mM EDTA,

^'0.25 mg/ml N-ethylmaleimide, 0.4 mM phenylmethylsulfonyl fluoride) ) in a Polytron® homogenizer (Brinkman

Instruments; Westbury NY). Homogenates were centrifuged at 12,000 x g for 30 min, then the supernatant was removed and centrifuged at 378,000 x g for 60 min.

The supernatant obtained following the second centrifugation (soluble fraction) was further fractionated by anion exchange chromatography on a DEAE-Sepharose Fast Flow column (25 ml bed volume) (Pharmacia; Piscataway NJ) that had been preequilibrated with buffer C (50 mM Tris, 0.15 M NaCl, 0.1% Triton X-100, pH 8.0). The soluble fraction was loaded and the column was washed sequentially with buffer C, buffer D (50 mM Tris, 0.25 M NaCl, 0.1% Triton X-100, pH 8.0), buffer E (50 mM Tris, 6 M urea, 0.25 M NaCl, 0.1% Triton X-100, pH 8.0) and buffer F (50 mM sodium formate, 6 M urea, 0.2 M NaCl, 0.1% Triton X-100, pH 3.5). Eluates from each wash were saved. Following the final wash, the column was adjusted to pH 8.0 and proteoglycans were eluted using a linear gradient of 0.2 to 1 M NaCl in 50 mM Tris, pH 8.0, 0.5% CHAPS (3-((3- cholamidopropyl)dimethylammonio)-1-propane sulfonate) . This fraction primarily contains CSPG's, but also contains glycoproteins other than proteoglycans. The concentration of the isolated total CSPG's (measured as protein) was determined using the BCA protein assay kit (Pierce; Rockford, IL) . Samples of the isolated CSPG's (2.5 μg protein) were enzymatically deglycosylated with chondroitinase ABC in 100 mM Tris-HCl, pH 8.0, 30 mM sodium acetate at 37°C for 45 min. In some experiments, proteoglycans first were digested with 0.5 units Chondroitinase-ABC/mg protein as described above, then subsequently digested with N-glycanase (0.6 units/μg protein) in 10 mM Tris, pH 8.0, 3 mM sodium acetate, 50 mM 2-mercaptoethanol, 50 mM EDTA, 0.5% SDS at 37°C for 18 hr.

Aliquots of the isolated CSPG's also were chemically deglycosylated using trifluoromethanesulfonic acid (TMSA) as described by Edge et al., Anal. Biochem. 118:131-137 (1981), which is incorporated herein by reference. Following chemical deglycosylation, the samples were dialyzed against phosphate buffered saline. Chemically deglycosylated CSPG core proteins were used for immunization (see below), so as to avoid contaminating the immunogen with proteins in the enzyme preparations.

The total CSPG sample and the deglycosylated CSPG core proteins were treated with GAG lyase to minimize the diffuse bands that occur due to the presence of highly heterologous GAG chains. Following GAG lyase treatment, the samples were examined by sodium dodecyl sulfate- polyacrylamide gel electrophoresis (SDS-PAGE) using 8-16% gradient gels (Novex; San Diego, CA) under reducing or nonreducing conditions. Proteins were visualized by silver staining (Boehringer Mannheim; Indianapolis IN) and molecular weights were determined by comparison to prestained molecular weight standards (Gibco-BRL; Gaithersburg MD) .

As shown in Figure 1, the undigested CSPG's migrated as a diffuse smear, which is typical for most proteoglycans fractionated under nonreducing conditions (lane 1). Following digestion with various amounts of chondroitinase ABC, the smear completely disappeared and at least five major bands of 220, 145, 125, 80, and 60 kilodaltons (kDa) were observed (lanes 2-4). Chemically deglycosylated CSPG's showed a similar pattern of core protein migration, except that a few bands in the chemically deglycosylated sample migrated slightly faster than the enzymatically-digested sample (lane 5). The faster migration is due to the removal of carbohydrate chains other than glycosaminoglycans by TMSA (not shown) .

B. Production of a Multispecific Anti-CSPG Antiserum

The dialyzed, chemically deglycosylated CSPG core proteins were mixed with an equal volume of monophosphoryl lipid A + trehalose dicorynomycolate .adjuvant (RIBI ImmunoChem; Hamilton MT) and injected into rabbits according to the immunization protocol suggested by the manufacturer of the adjuvant. Each rabbit was immunized with 2 mg of core proteins and booster immunizations were administered at 4 wk intervals. Blood samples were collected 1 wk after each boost and were examined by immunoblot analysis as described by Harlow and Lane (1988) . Specific binding was identified using an alkaline phosphatase-conjugated goat anti-rabbit IgG second antibody (Bio-Rad; Cambridge MA) and the BCIP/NBT substrate (bromochloroindolyl phosphate/nitroblue tetrazolium; Promega) .

One antiserum was obtained from blood collected after the second booster immunization reacted strongly with the 145 kDa and 80 kDa proteins and weakly with the 185 kDa and 125 kDa proteins (not shown) . This antiserum, which also showed substantial reactivity to non-proteoglycan components present in fractions that did not bind to the DEAE anion exchange column, was selected for further use. In order to remove the undesirable non- proteoglycan binding activity, the antiserum was adsorbed against extracts from E. coli XLl-Blue and with non- proteoglycan eluates obtained during DEAE chromatography. The E. coli extract, which was prepared as described by Sambrook et al. (1989), and the buffer C, buffer D and buffer E eluates each were coupled to CNBr-activated Sepharose® 4B (Pharmacia) . The antiserum was adsorbed first with the E. coli extract affinity resin, then with affinity resins containing the non-proteoglycan eluates. Adsorption was performed by incubating 1 ml of antiserum with an affinity resin overnight at 4°C with constant mixing, after which unbound material was collected.

Specificity of the adsorbed antiserum was examined by immunoblot analysis as described above. The antiserum maintained its specificity for ;the 80 kDa, 145 kDa and 185 kDa core proteins following adsorption with the E. coli extracts and DEAE eluates, whereas most of the nonspecific reactivity against the non-proteoglycan components was removed (Figure 2, lane 2). This multispecific antiserum was used to screen a λgtll cDNA library as described below.

EXAMPLE II

ISOLATION AND CHARACTERIZATION OF A cDNA SEQUENCE ENCODING BREVICAN

This example provides methods for isolating a cDNA encoding brevican and for characterizing the cDNA sequence.

A. Isolation of cDNA encoding brevican:

A bovine brain λgtll cDNA library (Clontech; Palo

Alto CA) was screened using the immunoscreening method described by Sambrook et al. (1989). Approximately 8 X 10⁵ clones were screened with the multispecific anti-CSPG antiserum. Reactive clones were visualized using an alkaline phosphatase-conjugated goat anti-rabbit IgG (Bio- Rad) and the BCIP/NBT substrate (Promega; Madison WI) as described above.

After three rounds of immunosσreening, eight positive clones were selected and plaque-purified. The phage DNA was digested with EcoRI and the fragments were subcloned into pBluescript II KS (+) (Stratagene, San Diego, CA) . Southern blot analysis of the subcloned inserts revealed that three of the subclones (clones #5, #17 and #19) belonged to a single cross-hybridizing group. The DNA sequences of the three cross-reacting subclones were determined by the dideoxy chain termination method using [³⁵S]dATP and Sequenase version 2.0 (United States Biochemicals; Cleveland OH) . Deletion mutants were prepared by digestion with exonuclease III using the Erase- a-Base System Kit (Promega) .

The three cross-reacting cDNA clones contained inserts of 2.2 kilobases (kb) , 1.9 kb and 2.5 kb. Using restriction analysis and DNA sequencing, one of the clones (#19) was found to be an artifact and was discarded. The two remaining clones overlap by 0.8 kb and encode a 3.3 kb transcript that contains a 2736 nucleotide open reading frame (Figure 3; SEQ ID NO: 1). The open reading frame is flanked by a 5' untranslated region of 111 nucleotides and a 3' untranslated region of 412 nucleotides.

The full length cDNA sequence encodes a 912 amino acid protein having a calculated molecular mass of 99,510 Daltons (Da) (Figure 3; SEQ ID NO: 2). Inspection of the derived amino acid sequence of brevican revealed the protein contains various structural domains that are characteristic of the aggrecan/versican family of CSPG's (Figure 4), including a hyaluronic acid-binding domain consisting of an immunoglobulin-like loop (residues 35 to 158), two link protein-like tandem repeats (residues 159 to 353), one copy of EGF-like repeat (residues 649 to 684), a lectin-like domain (residues 685 to 814) and a complement regulatory protein (CRP)-like domain (residues 815 to 875). In addition, brevican contains a central domain (residues 354 to 648) that is not homologous among the members of the aggrecan/versican family of CSPG's. This nonhomologous domain in brevican core protein is much shorter than the central region in other proteoglycans.

B. Characterization of the proteins encoded by the brevican cDNA sequence

In order to determine the relationship of the two isolated cDNA clones to the individual core proteins detected by immunoblot analysis, anti-brevican antibodies were affinity-selected against the brevican fusion proteins encoded by clones #5 and #17 (see Argraves et al., J. Biol. Chem. 12922-12924 (1986), and Suzuki et al., Proc. Natl. Acad. Sci. USA 83:8614-8618 (1986), each of which is incorporated herein by reference) . Briefly, 1 X 10^s plaque- forming units of the two brevican cDNA-containing clones were inoculated onto a 10 cm agarose plate and incubated at 42°C for 4 hr. A nitrocellulose filter soaked with 10 mM isopropyl-β-D-thiogalactopyranoside was placed on the top of the agarose and incubation was continued for 5 hr at 37°C. The filter was removed from the plate, washed 3x with phosphate-buffered saline and cut into small pieces. The adsorbed multispecific antiserum was added to the filters and incubated at 4°C overnight. Bound antibodies were eluted with 5 mM glycine/HCl buffer, pH 2.9, then immediately neutralized with 1/10 vol 1 M Na₂HP0₄, pH 8.0.

The specificity of the affinity-selected antibodies was examined by immunoblot analysis of the chondroitinase ABC-digested total proteoglycan fraction.

As shown in Figure 2, the affinity-selected antibody (lanes 3 and 4) recognized both the 145 kDa and 80 kDa core proteins but no longer reacted with the 185 kDa protein (compare lane 2, using multispecific antiserum before affinity selection) . These results indicate that the two isolated cDNA clones encode both the 145 kDa and 80 kDa core proteins.

The affinity-selected anti-brevican antibodies reacted with undigested total proteoglycans on immunoblots and recognized a diffuse smear of about 200 kDa (Figure 5, lane 3). Following digestion of the total CSPG's with chondroitinase ABC, the smear was eliminated and two bands corresponding to the 145/80 kDa core protein was observed (Figure 5, lane 4). Further digestion of the sample with heparitinase did not affect the migration of these bands (not shown). Thus, like other members of the aggrecan/versican family, brevican carries chondroitin (and/or dermatan) sulfate groups but not heparan sulfate.

DEAE chromatography eluates that did not bind the affinity resin or weakly bound to the resin also were probed with the affinity-selected antibodies. Again, only the two bands migrating at 80 kDa and 145 kDa were detected (see Figure 5, lane 2; buffer C eluate). In total soluble extracts of the brain, the antibodies reacted with a diffuse smear of about 200 kDa as well as the 80 kDa and 145 kDa bands (Figure 5, lane 1). These results indicate that a substantial portion of the brevican in brain tissue does not have any GAG chain attached.

EXAMPLE III

PURIFICATION AND CHARACTERIZATION OF BREVICAN

This example provides a method for purifying brevican and for determining that the 80 kDa protein is an active fragment of the 145 kDa protein. The total proteoglycan soluble fraction (65 μg) was digested with chondroitinase ABC as described above, then fractionated by reverse phase-high performance liquid chromatography (RP-HPLC) on a Vydac C4 column (4.6 X 250 mm) (Vydac/The Separations Group; Hesperia CA) . The column was preequilibrated with 0.1% trifluoroacetic acid (TFA), then the deglycosylated proteoglycan sample was loaded and eluted over 80 min using a linear 20% to 60% acetonitrile gradient. The fraction containing the 80 kDa core protein was isolated and the amino acid sequences of the N-terminus and of thirteen internal tryptic peptides were determined by the automated Edman degradation method using a gas-phase sequencer (Applied Biosystems; Foster City CA) .

Internal peptides of the 80 kDa core protein were generated by carboxymethylating the purified 80 kDa core protein (Stone et al., in A Practical Guide to Protein and

Peptide Purification for Microsequencing (ed. P.

Matsudaira; Academic Press, Inc. 1989), which is incorporated herein by reference), then digesting the sample with sequencing-grade bovine pancreatic trypsin

(Boehringer Mannheim) at an enzyme-to-substrate ratio of

1:25 in 0.1 M Tris/HCl (pH 8.1) /l M urea at 37°C for 18 hr.

The resulting peptide fragments were loaded onto a Vydac C8 column (2.1 x 250 mm), which had been preequilibrated in 0.1% TFA, and were eluted using a 0-40% acetonitrile gradient (over 200 min) .

The N-terminus and the tryptic peptide fragments were sequenced as described above. Twelve N-terminal amino acids and amino acids sequences of the tryptic fragments were obtained. Each experimentally determined amino acid sequence was present in the amino acid sequence that was derived from the isolated cDNA sequence (see Figure 3) . These results confirm that the isolated cDNA sequence encodes the 80 kDa protein. The N-terminal sequence of the 80 kDa protein occurred at residues 401 to 412 of the derived amino acid sequence and all of the amino acid sequences determined from the tryptic peptide fragments occurred in the C- terminal half of the isolated cDNA open reading frame (see Figure 3) . The calculated molecular weight of the C- terminal region corresponding to the 80 kDa core protein (residues 401 to 912) is 56,087 Da; the calculated molecular weight of the putative full-length core protein (residues 23 to 912; excluding the signal peptide) is 97,194 Da. The calculated molecular weights are lower than the molecular masses determined by SDS-PAGE (80 kDa and 145 kDa, respectively) . However, aberrant migration of proteoglycan core proteins is not uncommon.

Potential attachment sites for N-linked sugar chains are present at Asn-130 and Asn-337 (Figure 3). Another site, Asn-Pro-Ser at residues 903 to 905, contains a Pro residue adjacent to the Asn and, therefore, is not likely to be a glycosylation site. In order to determine whether brevican contains N-linked sugars, chondroitinase ABC-digested total proteoglycans were treated with N- glycanase, then examined by immunoblotting using the affinity-selected antibodies. Following N-glycanase treatment, the 145 kDa band shifted downward by about 5 kDa, whereas the 80 kDa band was not affected (not shown) . These results indicate that the two potential GAG attachment sites present in the N-terminal half of the derived amino acid sequence shown in Figure 3.

The derived amino acid sequence contains 11 serine-glycine and glycine-serine dipeptide sequences, which can be GAG attachment sites (Bourdon et al., 1987; Zimmermann and Ruoslahti, 1989; Figure 3). Five of these dipeptides are in the central nonhomologous region; GAG chains also can occur in the central regions of other members of the aggrecan/versican family of proteoglycans. The entire central nonhomologous region of brevican is rich in glutamic acid, which accounts for 24% of the 295 amino acid sequence. In addition, both the 80 kDa and 145 kDa forms of brevican contain a cluster of acidic amino acids at position 462 to 477. Human versican contains a similar cluster of glutamic acid residues on the C-terminal side of the link protein-like domain, at residues 400 to 408 (Zimmermann and Ruoslahti, 1989). In contrast, rat neurocan does not have such an acidic cluster (Rauch et al., 1992). However, the β amyloid precursor protein of Alzheimer's disease exists as a CSPG in certain cell lines (Shioi et al., 1992) and also contains a stretch of acidic residues (Kang et al., 1987).

These results indicate that bovine brevican exists in two forms, a 145 kDa proteoglycan that contains N-linked GAGs, and an 80 kDa form. Brevican has characteristics that place it within the aggrecan/versican family of CGSP's. However, as shown above, brevican can exist in a variant form that lacks one or more of the characteristic domains of the aggrecan/versican family.

EXAMPLE IV

BREVICAN EXPRESSION IN CELLS

This example demonstrates that brevican is expressed primarily in glial cells.

Total RNA from bovine brain, heart, lung and spleen were obtained from Clontech. Total RNA from rat primary neurons and astrocytes was isolated using the guanidinium isothiocyanate method (Sambrook et al., 1989). Primary cultures of type I astrocytes were prepared from postnatal (day 0) rats as described by McCarthy and de Vellis, J. Cell Biol. 85:890-902 (1980), which is incorporated herein by reference. Primary neurons were isolated from forebrains of embryonic (day 15) rat fetuses as described by Stallcup and Beasley (Proc. Natl. Acad. Sci. , USA 82:1276-1280 (1985), which is incorporated herein by reference) and treated with 20 μM σytosine arabinoside for two days to eliminate nonneuronal cells. Purity of the cultures was verified by staining cells with fluorescein- conjugated tetanus toxin (Novabiochem; San Diego, CA) for neurons and with rabbit anti-human GFAP for astrocytes. These analyses showed that each culture was >95% pure.

Ten μg of total RNA samples were denatured with glyoxal, electrophoresed in a 1% agarose gel and transferred to a GeneScreen Plus* membrane (DuPont/NEN; Boston MA) . ³²P-labelled probes were prepared by the random primer method using the Random Primed DNA labelling kit (Boehringer Mannheim) . Hybridization was carried out at 60°C for 16 hr. Following hybridization, the membrane was washed in the presence of 1% SDS at 60°C for 30 min and exposed to Kodak XAR5 film.

Probes consisted of either a 545 bp EcoRI fragment of the cDNA of SEQ ID NO: 1 (nucleotide positions 1409-1953), which has minimal sequence homology with aggrecan, versican or neurocan ("brevican-specific probe"), or with a 581 bp Pst I-Sal I fragment (nucleotide positions 249-829), which is highly homologous between rat and bovine sequences ("homologous probe"). When northern blots of the bovine RNA samples were examined using the brevican- specific probe, a single 3.3 kb band was observed in brain RNA but not RNA from other bovine tissues (Figure 6). In order to identify the cellular origin of brevican in the brain, RNA was isolated from primary cultures of rat cerebral neurons and cerebellar astrocytes and examined using the homologous probe. This probe hybridized with a single band of 3.3 kb in total RNA isolated from the astrocyte culture (Figure 6, lane 6) but not with RNA from primary neurons (Figure 6, lane 5). These result indicate that brevican is produced primarily in the brain and that the glial cells in the brain are the source of brevican production.

The localization of brevican protein in rat brain cells also was examined by an immunofluorescence assay using anti-rat brevican antibodies. Anti-rat brevican antibodies were obtained by immunizing rabbits with HPLC- purified 80 kDa rat brevican (obtained using the methods described above) and selecting for anti-rat brevican- specific antibodies by affinity purification against the 80 kDa form of rat brevican. The specificity of the anti-rat brevican antibodies was confirmed by western blot analysis, which showed the anti-rat brevican antibodies bound to the 145 kDa and 80 kDa forms of rat brevican (not shown) .

The anti-rat brevican antibody bound to brevican on the surface of rat astrocytes. The binding appeared in a punctate pattern and immunostaining was slightly more intense at the cell boundary (Figure 8.B.). No staining was observed with preimmune IgG (Figure 8.D.). Anti-rat brevican antibody binding also was examined in a co-culture of neurons and astrocytes. Cerebellar granule neurons were obtained from P7 rats and were added to nearly confluent astrocyte cultures. The co-culture was double-stained with fluorescein-conjugated tetanus toxin, which is specific for neurons, and anti-brevican IgG. Strong brevican immunoreactivity was detected at contact sites between neuronal cell bodies and astrocytes (Figure 8.F.). Brevican also was detected along neurites. In conjunction with the results showing that rat neuronal cells do not express brevican mRNA, the results of the immunostaining analyses indicate that brevican is produced by astrocytes and is deposited primarily at contact sites with neurons, rather than at sites of astrocyte-astrocyte contact. Although the invention has been described with reference to the examples provided above, it should be understood that various modifications can be made without departing from the spirit of the invention. Accordingly, the invention is limited only by the claims.

SEQUENCE LISTING

(1) GENERAL INFORMATION:

(i) APPLICANT: LA JOLLA CANCER RESEARCH FOUNDATION

(ii) TITLE OF INVENTION: Brevican, A Glial Cell Proteoglycan (iii) NUMBER OF SEQUENCES: 3

(iv) CORRESPONDENCE ADDRESS:

(A) ADDRESSEE: Campbell and Flores

(B) STREET: 4370 La Jolla Village Drive, Suite 700 (C) CITY: San Diego

(D) STATE: California

(E) COUNTRY: USA

(F) ZIP: 92122

(v) COMPUTER READABLE FORM:

(A) MEDIUM TYPE: Floppy disk

(B) COMPUTER: IBM PC compatible

(C) OPERATING SYSTEM: PC-DOS/MS-DOS

(D) SOFTWARE: Patentln Release #1.0, Version #1.25

(vi) CURRENT APPLICATION DATA:

(A) APPLICATION NUMBER:

(B) FILING DATE: 27-MAR-1995

(C) CLASSIFICATION:

(viii) ATTORNEY/AGENT INFORMATION:

(A) NAME: Imbra, Richard J.

(B) REGISTRATION NUMBER: 37,643

(C) REFERENCE/DOCKET NUMBER: FP-LJ 1453

(ix) TELECOMMUNICATION INFORMATION:

(A) TELEPHONE: (619) 535-9001

(B) TELEFAX: (619) 535-8949

(2) INFORMATION FOR SEQ ID NO:l:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 3259 base pairs

(B) TYPE: nucleic acid

(C) STRANDEDNESS: double

(D) TOPOLOGY: linear

(i ) FEATURE:

(A) NAME/KEY: CDS

(B) LOCATION: 112..2848

(xi) SEQUENCE DESCRIPTION: SEQ ID NO:l:

GAATTCCGGG CCTCTTGGGA AGGGATGTGG GTCATGGAGG GGGCAAACTT CGCGGTGCCT 60

CACCCCCACC CCGGTCCTCA GTTGCCTGCC GTCCCTCCCT CATCCTGCAG C ATG GCC 117

Met Ala

1

CCA CTG TTC CTG CCC CTC CTG GCA ACC CTG GTT CTG GCC TGG ATC CCT 165 Pro Leu Phe Leu Pro Leu Leu Ala Thr Leu Val Leu Ala Trp lie Pro 5 10 15

GTG GCC TTG GCT GAT GCT CTG GAA GGA GAC AGC TCA GAG GAC AGG GCC 213 Val Ala Leu Ala Asp Ala Leu Glu Gly Asp Ser Ser Glu Asp Arg Ala 20 25 30

TTC CGC GTG CGC ATC GCG GGC GAC GCG CCG CTG CAG GGC GTG CTG GGC 261

Phe Arg Val Arg lie Ala Gly Asp Ala Pro Leu Gin Gly Val Leu Gly 35 40 45 50

GGC GCC CTC ACC ATC CCA TGC CAC GTT CAC TAC CTG CGG CCG TCG CCG 309 Gly Ala Leu Thr ie Pro Cys His Val His Tyr Leu Arg Pro ser Pro 55 60 65

AGC CGC CGG GCC GCG CAG GGC TCC CCG CGG GTT AAG TGG ACC TTC CTG 357 Ser Arg Arg Ala Ala Gin Gly ser Pro Arg Val Lys Trp Thr Phe Leu 70 75 80

TCC GGC GGC CGG GAG GCC GAG GTG CTA GTG GCG CGG GGC CTA CGC GTC 405

Ser Gly Gly Arg Glu Ala Glu Val Leu Val Ala Arg Gly Leu Arg Val 85 90 95

AAG GTG AGC GAG GCC TAC CGG TTC CGC GTG GCA CTG CCT GCT TAC CCG 453 Lys Val ser Glu Ala Tyr Arg Phe Arg Val Ala Leu Pro Ala Tyr Pro 100 105 110

GCG TCA CTC ACC GAC GTC TCC CTG GTG CTG AGT GAG CTG CGG CCC AAC 501 Ala Ser Leu Thr Asp Val Ser Leu Val Leu Ser Glu Leu Arg Pro Asn 115 120 125 130

GAC TCA GGC ATT TAC CGC TGC GAG GTC CAG CAC GGC ATC GAC GAC AGC 549 Asp Ser Gly He Tyr Arg Cys Glu Val Gin His Gly He Asp Asp Ser 135 140 145

AGC GAC GCG GTG GAG GTC AAG GTC AAA GGG GTC GTC TTT CTC TAC CGG 597 Ser Asp Ala Val Glu Val Lys Val Lys Gly Val Val Phe Leu Tyr Arg 150 155 160

GAG GGC TCT GCC CGC TAC GCT TTC TCC TTC GCT GGG GCC CAG GAG GCC 645 Glu Gly ser Ala Arg Tyr Ala Phe Ser Phe Ala Gly Ala Gin Glu Ala 165 170 175 TGT GCC CGC ATC GGA GCC CGA ATC GCC ACT CCG GAG CAG CTC TAT GCC 693

Cys Ala Arg He Gly Ala Arg He Ala Thr Pro Glu Gin Leu Tyr Ala 180 185 190

GCC TAC CTC GGG GGC TAT GAA CAG TGT GAC GCT GGC TGG CTG TCC GAC 741 Ala Tyr Leu Gly Gly Tyr Glu Gin Cys Asp Ala Gly Trp Leu Ser Asp 195 200 205 210

CAG ACC GTG AGG TAT CCC ATC CAG ACG CCA CGA GAG GCC TGT TAT GGA 789 Gin Thr Val Arg Tyr Pro He Gin Thr Pro Arg Glu Ala cys Tyr Gly 215 220 225

GAC ATG GAT GGC TTC CCT GGG GTC CGG AAC TAC GGA GTG GTC GAC CCC 837 Asp Met Asp Gly Phe Pro Gly Val Arg Asn Tyr Gly Val Val Asp Pro 230 235 240

GAT GAC CTC TAT GAT GTT TAC TGT TAT GCT GAA GAA CTA AAT GGA GAG 885 Asp Asp Leu Tyr Asp Val Tyr Cys Tyr Ala Glu Glu Leu Asn Gly Glu 245 250 255

CTG TTC CTG GGT GCC CCT CCA GAC AAG CTG ACC TTG GAG GAG GCG CGG 933 Leu Phe Leu Gly Ala Pro Pro Asp Lys Leu Thr Leu Glu Glu Ala Arg 260 265 270

ACA TAC TGC CAG GAG CGG GGT GCT AAG ATT GCA ACC ACC GGC CAG CTG 981 Thr Tyr Cys Gin Glu Arg Gly Ala Lys He Ala Thr Thr Gly Gin Leu 275 280 285 290

TAT GCA GCC TGG GAT GGT GGC CTG GAC CGC TGC AGC TCT GGC TGG CTG 1029 Tyr Ala Ala Trp Asp Gly Gly Leu Asp Arg Cys Ser Ser Gly Trp Leu 295 300 305

TCT GAT GGC AGT GTG CGC TAC CCC ATC GTC ACC CCC AGC CAG CGC TGT 1077 Ser Asp Gly Ser Val Arg Tyr Pro He Val Thr Pro Ser Gin Arg Cys 310 315 320

GGT GGG GGC CTC CCT GGT GTC AAG ACT CTC TTC CTC TTC CCC AAC CAG 1125 Gly Gly Gly Leu Pro Gly Val Lys Thr Leu Phe Leu Phe Pro Asn Gin 325 330 335

ACT GGC TTC CCC AAC AAG CAC AGC CGC TTC AAC GTC TAC TGC TTC CGA 1173 Thr Gly Phe Pro Asn Lys His Ser Arg Phe Asn Val Tyr Cys Phe Arg 340 345 350

GAC TCT GCC CAG CCT TCT GCC ATC CCT GAG GCA GCC AAC CCA GCC TCT 1221 Asp Ser Ala Gin Pro Ser Ala He Pro Glu Ala Ala Asn Pro Ala Ser 355 360 365 370

CAC CTG GCC TCT GAT GCA CTG GAA GCC ATT GTC ACA GTG ACT GAG ACC 1269 His Leu Ala Ser Asp Ala Leu Glu Ala He Val Thr Val Thr Glu Thr 375 380 385

CTG GAG GAA CTG AAG CTG CCC CAG GAA GCT GTG GAA AGC GAG TCC CGA 1317 Leu Glu Glu Leu Lys Leu Pro Gin Glu Ala Val Glu Ser Glu Ser Arg 390 395 400

GGA GCC ATC TAT TCC ATC CCC ATT ATA GAG GAT GGA GGT GGT GGG AGC 1365 Gly Ala He Tyr Ser He Pro He He Glu Asp Gly Gly Gly Gly Ser 405 410 415

TCC ACT CCA GAA GAC CCA GCA GAG GCC CCT AGA ACC CTC CTA GAA TTC 1413 Ser Thr Pro Glu Asp Pro Ala Glu Ala Pro Arg Thr Leu Leu Glu Phe 420 425 430

GAA ACC CAA TCC ATT GTG CCT CCA TTG GGG TCC TCA GAA GAG GAA GGC 1461 Glu Thr Gin Ser He Val Pro Pro Leu Gly Ser Ser Glu Glu Glu Gly 435 440 445 450 AAG GTG TTG GAG CAA GAA GAG AAA TAC AGG GGT GAA GAA GAG AAA GAA 1509 Lys Val Leu Glu Gin Glu Glu Lys Tyr Arg Gly Glu Glu Glu Lys Glu 455 460 465

GAG GAA GAA GAA GAG GAG GAG GTG GAG GAT GAG GCC CTG TGG GCC TGG 1557 Glu Glu Glu Glu Glu Glu Glu Val Glu Asp Glu Ala Leu Trp Ala Trp 470 475 480

CCC AGT GAG CTC AGC AGC CTG GAC CCA GAG GCC CCT CTC CCC ACT GAG 1605 Pro Ser Glu Leu Ser Ser Leu Asp Pro Glu Ala Pro Leu Pro Thr Glu 485 490 495

CCC GTT CCA GAG GAG TCA CTC ACC CAG GCA TCG CCT CCA GTG AGG GCT 1653 Pro Val Pro Glu Glu Ser Leu Thr Gin Ala ser Pro Pro Val Arg Ala 500 505 510

GCC CTC CAG CCT GGT GTA TCA CCA CCA CCC TAT GAT GAG CCA GAG GCT 1701 Ala Leu Gin Pro Gly Val Ser Pro Pro Pro Tyr Asp Glu Pro Glu Ala 515 520 525 530

CCC AGG CCT CCA AGG GTC CTT GGA CCA CCC ACC AAG ACC CTG CCC ACT 1749 Pro Arg Pro Pro Arg Val Leu Gly Pro Pro Thr Lys Thr Leu Pro Thr 535 540 545

CCT AGG GAG GGG AAC CTG GCA TCC CCC CCA CCT TCC ACT CTG GTT GGG 1797 Pro Arg Glu Gly Asn Leu Ala Ser Pro Pro Pro Ser Thr Leu Val Gly 550 555 560

GCA AGA GAG ATA GAG GAG GAG ACT GGG GGT CCT GAG CTC TCT GGG GCC 1845 Ala Arg Glu He Glu Glu Glu Thr Gly Gly Pro Glu Leu Ser Gly Ala 565 570 575

CCT CGA GGA GAG AGT GAG GAG ACA GGA AGC TCC GAG GAT GCC CCT TCC 1893 Pro Arg Gly Glu Ser Glu Glu Thr Gly Ser Ser Glu Asp Ala Pro Ser 580 585 590

CTG CTT CCA GCC ACA CGG GCC CCT GGG GAT ACC AGG GAT CTG GAG ACC 1941 Leu Leu Pro Ala Thr Arg Ala Pro Gly Asp Thr Arg Asp Leu Glu Thr 595 600 605 610

CCC TCT GAA GAG AAT TCC AGA AGA ACT GTC CCA GCA GGG ACT TCA GTG 1989 Pro Ser Glu Glu Asn Ser Arg Arg Thr Val Pro Ala Gly Thr Ser Val 615 620 625

CGT GCC CAG CCA GTG CTG CCC ACT GAC AGT GCC AGC CGT GGT GGA GTG 2037 Arg Ala Gin Pro Val Leu Pro Thr Asp Ser Ala Ser Arg Gly Gly Val 630 635 640

GCC GTG GCC CCC TCA TCA GGT GAC TGT GTC CCC AGC CCC TGC CAC AAT 2085 Ala Val Ala Pro Ser ser Gly Asp Cys Val Pro Ser Pro Cys His Asn 645 650 655

GGC GGG ACA TGC TTG GAG GAG GAG GAG GGG GTC CGC TGC CTG TGT TTG 2133 Gly Gly Thr Cys Leu Glu Glu Glu Glu Gly Val Arg Cys Leu cys Leu 660 665 670

CCT GGC TAT GGG GGG GAC CTG TGC GAT GTT GGC CTC CAC TTC TGC AGC 2181 Pro Gly Tyr Gly Gly Asp Leu Cys Asp Val Gly Leu His Phe Cys Ser 675 680 685 690

CCC GGT TGG GAC GCC TTC CAG GGT GCC TGC TAC AAG CAC TTT TCT GCC 2229 Pro Gly Trp Asp Ala Phe Gin Gly Ala Cys Tyr Lys His Phe Ser Ala 695 700 705

CGA AGG AGC TGG GAG GAG GCG GAG AAC AAG TGC CGG ATG TAC GGC GCG 2277 Arg Arg Ser Trp Glu Glu Ala Glu Asn Lys Cys Arg Met Tyr Gly Ala 710 715 720 CAC CTG GCC AGC ATC AGC ACG CCG GAG GAA CAG GAC TTC ATC AAC AAT 2325 His Leu Ala Ser He Ser Thr Pro Glu Glu Gin Asp Phe He Asn Asn 725 730 735

CGA TAC CGG GAG TAC CAG TGG ATC GGG CTC AAT GAC AGG ACC ATC GAA 2373 Arg Tyr Arg Glu Tyr Gin Trp He Gly Leu Asn Asp Arg Thr lie Glu 740 745 750

GGG GAT TTC CTG TGG TCA GAT GGC GTC CCC CTG CTC TAT GAG AAC TGG 2421 Gly Asp Phe Leu Trp Ser Asp Gly Val Pro Leu Leu Tyr Glu Asn Trp 755 760 765 770

AAC CCT GGG CAG CCA GAC AGC TAC TTC CTG TCC GGA GAG AAC TGC GTG 2469 Asn Pro Gly Gin Pro Asp Ser Tyr Phe Leu Ser Gly Glu Asn cys Val 775 780 785

GTT ATG GTG TGG CAC GAT CAG GGA CAA TGG AGT GAT GTT CCC TGC AAC 2517 Val Met Val Trp His Asp Gin Gly Gin Trp ser Asp Val Pro Cys Asn 790 795 800

TAC CAC CTG TCC TAC ACC TGC AAG ATG GGG CTG GTG TCC TGT GGG CCC 2565 Tyr His Leu Ser Tyr Thr Cys Lys Met Gly Leu Val Ser Cys Gly Pro 805 810 815

CCA CCA GAG CTG CCC CTG GCT GAA GTG TTT GGC CGC CCA CGG CTG CGC 2613 Pro Pro Glu Leu Pro Leu Ala Glu Val Phe Gly Arg Pro Arg Leu Arg 820 825 830

TAT GAA GTC GAC ACA GTG CTT CGT TAC CGG TGC CGG GAG GGG CTG ACC 2661 Tyr Glu Val Asp Thr Val Leu Arg Tyr Arg Cys Arg Glu Gly Leu Thr '

835 840 845 850

CAG CGC AAC CTA CCA TTG ATC CGC TGC CAG GAG AAT GGT CGC TGG GGG 2709 Gin Arg Asn Leu Pro Leu He Arg cys Gin Glu Asn Gly Arg Trp Gly 855 860 865

CTT CCC CAG ATC TCC TGT GTG CCC CGC AGG CCT GCT CGA GCT CTG CGC 2757 Leu Pro Gin He Ser cys Val Pro Arg Arg Pro Ala Arg Ala Leu Arg 870 875 880

CCA GTA GAG GCC CAA GAA GGA CGT CCG TGG AGG CTC GTG GGG CAC TGG 2805 Pro Val Glu Ala Gin Glu Gly Arg Pro Trp Arg Leu Val Gly His Trp 885 890 895

AAG GCA CGG CTG AAT CCC TCT CCC AAT CCT GCT CCA GGT CCC T 2848

Lys Ala Arg Leu Asn Pro Ser Pro Asn Pro Ala Pro Gly Pro 900 905 910

AAGGGGCAAA GCCCCAGGCA CTGCCCTCTG CCACCAAGCC TCCCTAGACC CATGCTCTCC 2908

ACCACGGGAA GTGACAGCAT GAGAGGGGTG CAGCTGAAGT CCACATGGTA GTGCCCAGAG 2968

GGCTTCTGGG AAATACTGGG TTGGCCAAAA ATTCGTTTGG CTTTTTCTGT AACATCAATA 3028

TCTGGGTGTC TCCAGTCCTG CCTTAGGCCC TCTCCCTCCT ATTCTGGGCC TCAGGGCCTT 3088

GAGTAGGTCT CTAAGTGCCT CAACTGCCCT CTCCTTGCCA GCCATCCTGT CCCCTCCATT 3148

CCCCTGGAGG CCCCTGTACC CACTCATTCT GTTTCCCAAG AGAATGGGTT TGCAGGATGG 3208

GGTACCTGTA AAACAAGCAG AAAATAAAGC TGCATTTGAG CCCCGGAATT C 3259 (2) INFORMATION FOR SEQ ID NO:2:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 912 amino acids

(B) TYPE: amino acid (D) TOPOLOGY: linear

(ii) MOLECULE TYPE: protein

(xi) SEQUENCE DESCRIPTION: SEQ ID NO:2:

Met Ala Pro Leu Phe Leu Pro Leu Leu Ala Thr Leu Val Leu Ala Trp 1 5 10 15

He Pro Val Ala Leu Ala Asp Ala Leu Glu Gly Asp Ser Ser Glu Asp 20 25 30

Arg Ala Phe Arg Val Arg He Ala Gly Asp Ala Pro Leu Gin Gly Val 35 40 45

Leu Gly Gly Ala Leu Thr He Pro cys His Val His Tyr Leu Arg Pro 50 55 60

Ser Pro Ser Arg Arg Ala Ala Gin Gly Ser Pro Arg Val Lys Trp Thr 65 70 75 80

Phe Leu Ser Gly Gly Arg Glu Ala Glu Val Leu Val Ala Arg Gly Leu 85 90 95

Arg Val Lys Val ser Glu Ala Tyr Arg Phe Arg Val Ala Leu Pro Ala 100 105 110

Tyr Pro Ala Ser Leu Thr Asp Val ser Leu Val Leu Ser Glu Leu Arg 115 120 125

Pro Asn Asp Ser Gly He Tyr Arg cys Glu Val Gin His Gly He Asp 130 135 140

Asp Ser Ser Asp Ala Val Glu Val Lys Val Lys Gly Val Val Phe Leu 145 150 155 160

Tyr Arg Glu Gly ser Ala Arg Tyr Ala Phe Ser Phe Ala Gly Ala Gin 165 170 175

Glu Ala Cys Ala Arg He Gly Ala Arg He Ala Thr Pro Glu Gin Leu 180 185 190

Tyr Ala Ala Tyr Leu Gly Gly Tyr Glu Gin Cys Asp Ala Gly Trp Leu 195 200 205

Ser Asp Gin Thr Val Arg Tyr Pro He Gin Thr Pro Arg Glu Ala Cys 210 215 220

Tyr Gly Asp Met Asp Gly Phe Pro Gly Val Arg Asn Tyr Gly Val Val 225 230 235 240

Asp Pro Asp Asp Leu Tyr Asp Val Tyr cys Tyr Ala Glu Glu Leu Asn 245 250 255

Gly Glu Leu Phe Leu Gly Ala Pro Pro Asp Lys Leu Thr Leu Glu Glu 260 265 270

Ala Arg Thr Tyr Cys Gin Glu Arg Gly Ala Lys He Ala Thr Thr Gly 275 280 285

Gin Leu Tyr Ala Ala Trp Asp Gly Gly Leu Asp Arg Cys ser Ser Gly 290 295 300 Trp Leu Ser Asp Gly Ser Val Arg Tyr Pro He Val Thr Pro ser Gin 305 310 315 320

Arg cys Gly Gly Gly Leu Pro Gly Val Lys Thr Leu Phe Leu Phe Pro 325 330 335

Asn Gin Thr Gly Phe Pro Asn Lys His Ser Arg Phe Asn Val Tyr Cys 340 345 350

Phe Arg Asp Ser Ala Gin Pro Ser Ala He Pro Glu Ala Ala Asn Pro 355 360 365

Ala Ser His Leu Ala Ser Asp Ala Leu Glu Ala He Val Thr Val Thr 370 375 380

Glu Thr Leu Glu Glu Leu Lys Leu Pro Gin Glu Ala Val Glu Ser Glu 385 390 395 400

Ser Arg Gly Ala He Tyr Ser He Pro He He Glu Asp Gly Gly Gly 405 410 415

Gly Ser Ser Thr Pro Glu Asp Pro Ala Glu Ala Pro Arg Thr Leu Leu 420 425 430

Glu Phe Glu Thr Gin Ser He Val Pro Pro Leu Gly Ser Ser Glu Glu 435 440 445

Glu Gly Lys Val Leu Glu Gin Glu Glu Lys Tyr Arg Gly Glu Glu Glu 450 455 460

Lys Glu Glu Glu Glu Glu Glu Glu Glu Val Glu Asp Glu Ala Leu Trp 465 470 475 480

Ala Trp Pro ser Glu Leu ser ser Leu Asp Pro Glu Ala Pro Leu Pro 485 490 495

Thr Glu Pro Val Pro Glu Glu Ser Leu Thr Gin Ala Ser Pro Pro Val 500 505 510

Arg Ala Ala Leu Gin Pro Gly Val Ser Pro Pro Pro Tyr Asp Glu Pro 515 520 525

Glu Ala Pro Arg Pro Pro Arg Val Leu Gly Pro Pro Thr Lys Thr Leu 530 535 540

Pro Thr Pro Arg Glu Gly Asn Leu Ala Ser Pro Pro Pro Ser Thr Leu 545 550 555 560

Val Gly Ala Arg Glu He Glu Glu Glu Thr Gly Gly Pro Glu Leu ser 565 570 575

Gly Ala Pro Arg Gly Glu Ser Glu Glu Thr Gly Ser ser Glu Asp Ala 580 585 590

Pro ser Leu Leu Pro Ala Thr Arg Ala Pro Gly Asp Thr Arg Asp Leu 595 600 605

Glu Thr Pro ser Glu Glu Asn Ser Arg Arg Thr Val Pro Ala Gly Thr 610 615 620

Ser val Arg Ala Gin Pro Val Leu Pro Thr Asp Ser Ala Ser Arg Gly 625 630 635 640

Gly Val Ala Val Ala Pro Ser Ser Gly Asp Cys Val Pro Ser Pro Cys 645 650 655 His Asn Gly Gly Thr Cys Leu Glu Glu Glu Glu Gly Val Arg Cys Leu 660 665 670

Cys Leu Pro Gly Tyr Gly Gly Asp Leu Cys Asp Val Gly Leu His Phe 675 680 685

Cys Ser Pro Gly Trp Asp Ala Phe Gin Gly Ala Cys Tyr Lys His Phe 690 695 700

Ser Ala Arg Arg ser Trp Glu Glu Ala Glu Asn Lys Cys Arg Met Tyr 705 710 715 720

Gly Ala His Leu Ala Ser He Ser Thr Pro Glu Glu Gin Asp Phe He 725 730 735

Asn Asn Arg Tyr Arg Glu Tyr Gin Trp He Gly Leu Asn Asp Arg Thr 740 745 750

He Glu Gly Asp Phe Leu Trp Ser Asp Gly Val Pro Leu Leu Tyr Glu 755 760 765

Asn Trp Asn Pro Gly Gin Pro Asp ser Tyr Phe Leu Ser Gly Glu Asn 770 775 780

Cys Val Val Met Val Trp His Asp Gin Gly Gin Trp Ser Asp Val Pro 785 790 795 800

Cys Asn Tyr His Leu Ser Tyr Thr Cys Lys Met Gly Leu Val Ser Cys 805 810 815

Gly Pro Pro Pro Glu Leu Pro Leu Ala Glu Val Phe Gly Arg Pro Arg 820 825 830

Leu Arg Tyr Glu Val Asp Thr Val Leu Arg Tyr Arg Cys Arg Glu Gly 835 840 845

Leu Thr Gin Arg Asn Leu Pro Leu He Arg Cys Gin Glu Asn Gly Arg 850 855 860

Trp Gly Leu Pro Gin He Ser Cys Val Pro Arg Arg Pro Ala Arg Ala 865 870 875 880

Leu Arg Pro Val Glu Ala Gin Glu Gly Arg Pro Trp Arg Leu Val Gly 885 890 895

His Trp Lys Ala Arg Leu Asn Pro Ser Pro Asn Pro Ala Pro Gly Pro 900 905 910

(2) INFORMATION FOR SEQ ID NO:3:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 908 amino acids

(B) TYPE: amino acid (D) TOPOLOGY: linear

(xi) SEQUENCE DESCRIPTION: SEQ ID NO:3:

Met He Pro Leu Leu Leu Ser Leu Leu Ala Ala Leu Val Leu Thr Gin 1 5 10 15

Ala Pro Ala Ala Leu Ala Asp Asp Leu Lys Glu Asp ser ser Glu Asp 20 25 30

Arg Ala Phe Arg Val Arg He Gly Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa 35 40 45

Xaa Xaa Gly Ala Leu Ala He Pro Cys His Val His His Leu Gin Pro 50 55 60

Pro Pro Ser Arg Arg Ala Ala Xaa Gly Xaa Xaa Arg Val Lys Trp Thr 65 70 75 80

Phe Leu Ser Gly Gly Pro Glu Val Glu Val Leu Val Xaa Xaa xaa Xaa 85 90 95

Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Leu Asp Ala

100 105 110

Tyr Pro Xaa Ser Xaa Xaa Xaa Xaa ser Leu Val Leu Ser Glu Leu Arg 115 120 125

Pro Asn Asp Ser Gly val Tyr Arg Cys Xaa Val Gin His Gly He Asp 130 135 140

Asp Ser ser Asp Ala xaa Glu Val Lys Val Xaa Gly Val Val Phe Leu 145 150 155 160

Tyr Xaa xaa Xaa Xaa Xaa Arg Tyr Ala Phe Ser Phe Xaa Gly Ala Gin 165 170 175

Glu Ala cys Ala Arg He Gly Ala Arg He Ala Thr Pro Pro Gin Leu 180 185 190

Tyr Ala Ala Tyr Leu Gly Gly Tyr Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa 195 200 205

Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Ala Cys

210 215 220

Tyr Gly Asp Met Asp Gly Tyr Pro Gly Val Arg Asn Tyr Gly Val Val 225 230 235 240

Gly Pro Asp Asp Leu Tyr Asp Val Tyr cys Tyr Ala Glu Asp Leu Asn 245 250 255

Gly Glu Leu Phe Leu Gly Ala Pro Pro Gly Lys Leu Thr Trp Glu Glu 260 265 270

Ala Arg Asp Tyr Cys Leu Glu Arg Gly Ala Gin He Ala Ser Thr Gly 275 280 285

Gin Xaa Tyr Ala Ala Trp Asn Gly Gly Leu Asp Xaa Cys ser ser Gly 290 295 300 Trp Leu Ala Asp Gly Ser Val Arg Tyr Pro He He Thr Pro Ser Gin 305 310 315 320

Arg Cys Gly Xaa Gly Leu Pro Gly Val Lys Thr Leu Phe Leu Phe Pro 325 330 335

Asn Gin Thr Gly Phe Pro Ser Lys Gin Asn Arg Phe Asn Val Tyr Cys 340 345 350

Phe Arg Asp Ser Ala His Pro Ser Ala Phe Ser Glu Ala Ser Ser Pro 355 360 365

Ala Ser Asp Gly Ala Ser Asp Gly Leu Glu Ala He Val Thr Val Thr 370 375 380

Glu Lys Leu Glu Glu Leu Gin Leu Pro Gin Xaa Xaa Val Glu Ser Glu 385 390 395 400

Ser Arg Gly Ala He Tyr ser He Pro He Thr Glu Xaa Gly Gly Gly 405 410 415

Gly Ser ser Thr Pro Glu Asp Pro Ala Glu Ala Pro Arg Thr Pro Leu 420 425 430

Glu Ser Glu Thr Gin Ser Xaa Xaa Pro Pro Thr Gly Ser Ser Xaa Xaa 435 440 445

Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa xaa xaa Xaa Xaa Xaa Xaa Xaa Xaa

450 455 460

Xaa Xaa xaa Xaa Xaa Xaa Xaa Xaa xaa Xaa Xaa Xaa Xaa xaa Xaa Xaa

465 470 475 480

Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Leu ser Ser Pro Leu Pro 485 490 495

Thr Gly Phe Gin Thr Glu His Ser Leu ser Gin Val Ser Pro Pro Ala 500 505 510

Gin Ala Val Leu Gin Leu Gly Ala Ser Pro Ser Xaa Xaa Xaa Xaa Xaa 515 520 525

Xaa Xaa Pro Arg Pro Pro Arg Val His Gly Pro Pro Ala Glu Thr Leu 530 535 540

Gin Pro Pro Arg Glu Gly ser Leu Thr ser Thr Pro Asp Gly Thr Leu 545 550 555 560

Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa

565 570 575

Xaa Xaa Xaa Xaa Xaa Glu Glu Ala Gly ser ser ser Leu Glu Asp Gly 580 585 590

Pro Ser Leu Leu Pro Ala Thr Trp Ala Pro xaa Gly Thr Arg Glu Leu 595 600 605

Glu Thr Pro Ser Gin Glu Lys ser Gly Arg Thr Val Leu Thr Gly Thr 610 615 620

Ser Val Gin Ala Gin Pro Val Leu Pro Thr Asp Ser Ala Xaa aa Xaa 625 630 635 640

Xaa Val Ala Val Ala Pro Ser ser Gly Asp cys He Pro Ser Pro Cys 645 650 655 His Asn Gly Gly Thr Cys Leu Glu Glu Lys Glu Gly Phe Arg Cys Leu 660 665 670

Cys Leu Pro Gly Tyr Gly Gly Asp Leu Cys Glu Val Gly Leu His Phe 675 680 685 cys Ser Pro Gly Trp Glu Ala Phe Gin Gly Ala Cys Tyr Lys His Phe 690 695 700

Ser Thr Arg Arg Ser Trp Glu Glu Ala Glu Ser Gin Cys Xaa Xaa Xaa 705 710 715 720

Gly Ala His Leu Thr Ser He Cys Thr Pro Glu Glu Gin Asp Phe Val 725 730 735

Asn Asp Arg Tyr Arg Glu Tyr Gin Ser He Gly Leu Asn Asp Xaa Xaa 740 745 750

Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa

755 760 765

Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa 770 775 780

Xaa xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa

785 790 795 800

Cys Asn Tyr His Leu Ser Tyr Thr cys Lys Met Gly Leu Val ser Cys 805 810 815

Gly Pro Pro Pro Gin Leu Pro Leu Ala Gin He Phe Gly Arg Pro Arg 820 825 830

Leu Ala Xaa Xaa Val Asp Thr Val He Arg Tyr Arg Cys Arg Ala Gly 835 840 845

Leu Ala Gin Arg Asn Leu Pro Leu He Arg Cys Gin Glu Asn Gly Leu 850 855 860

Trp Xaa Xaa Asp Gin He Ser Cys Val Pro Arg Arg Pro Ala Arg Ala 865 870 875 880

Leu Arg Ser He Thr Ala Pro Glu Gly Pro Pro Gly Gin Leu Pro Arg 885 890 895

Xaa Xaa Lys Ala Leu Leu Thr Pro Pro Ser Ser Leu 900 905

Claims

We claim:

1. Substantially purified mammalian brevican protein.

2. The protein of claim 1, comprising substantially the amino acid sequence shown in Figure 3

(SEQ ID NO: 2) .

3. The protein of claim 1, comprising substantially the amino acid sequence of positions 354 to 648 of the amino acid sequence shown in Figure 3 (SEQ ID NO: 2).

4. An active fragment of the protein of claim 1, comprising substantially the amino acid sequence of positions 401 to 912 of the amino acid sequence shown in Figure 3 (SEQ ID NO: 2).

5. The active fragment of claim 4, comprising substantially the amino acid sequence of positions 401 to 648 of the amino acid sequence shown in Figure 3 (SEQ ID NO: 2) .

6. The protein of claim 1, comprising substantially the sequence of Figure 7 (SEQ ID NO: 3).

7. Substantially purified brevican of claim 1 having a derived molecular weight of about 97 kDa and a molecular mass of about 145 kDa as determined by SDS-PAGE.

8. The substantially purified brevican of claim 7, wherein said brevican is bovine brevican.

9. The substantially purified brevican of claim 7, wherein said brevican is rat brevican.

10. Substantially purified brevican of claim 1 having a derived molecular weight of about 56 kDa and a molecular mass of about 80 kDa as determined by SDS-PAGE.

11. The substantially purified brevican of claim 10, wherein said brevican is bovine brevican.

12. The substantially purified brevican of claim 10, wherein said brevican is rat brevican.

13. An anti-brevican antibody.

14. The antibody of claim 13, wherein said antibody is specific for the 80 kDa form of brevican.

15. The antibody of claim 13, wherein said antibody is specific for the 145 kDa form of brevican.

16. The antibody of claim 13, wherein said brevican is bovine brevican.

17. The antibody of claim 13, wherein said brevican is rat brevican.

18. The antibody of claim 13, wherein said antibody is a monoclonal antibody.

19. A cell line producing the antibody of claim 18.

20. The cell line of claim 19, wherein said cell is a hybridoma.

21. A reagent, comprising the antibody of claim 13 and a detectable label.

22. A substantially purified nucleic acid molecule encoding brevican.

23. The nucleic acid molecule of claim 22, comprising substantially the nucleic acid sequence shown in Figure 3 (SEQ ID NO: 1).

24. The nucleic acid molecule of claim 22, comprising substantially the nucleic acid sequence of nucleotide positions 111 to 2850 as shown in Figure 3 (SEQ ID NO: 1) .

25. The nucleic acid molecule of claim 22, comprising substantially the nucleic acid sequence of nucleotide positions 176 to 2850 as shown in Figure 3 (SEQ ID NO: 1) .

26. The nucleic acid molecule of claim 22, comprising substantially the nucleic acid sequence of nucleotide positions 1312 to 2850 as shown in Figure 3 (SEQ ID NO: 1) .

27. The nucleic acid of claim 22, wherein said brevican is rat brevican.

28. A vector comprising the nucleic acid molecule of claim 22.

29. A host cell comprising the vector of claim 28.

30. A nucleotide sequence, comprising at least ten nucleotides that hybridize under stringent conditions to the nucleic acid sequence shown in Figure 3 (SEQ ID NO: 1).

31. A probe, comprising the nucleotide sequence of claim 30 and a detectable label.

32. A method for detecting a glial cell in a cell sample suspected of containing a glial cell, comprising the steps of:

a. obtaining the cell sample;

b. contacting said sample with a reagent that can specifically bind to brevican, wherein said contact is under conditions that allow specific binding of said reagent to said brevican; and

c. identifying the presence of a glial cell by detecting the presence of said specifically bound reagent to brevican.

33. The method of claim 32, wherein said reagent is an antibody.

34. A method for detecting the presence of a glial cell in a cell sample suspected of containing a glial cell, comprising the steps of:

a. obtaining the cell sample;

b. extracting nucleic acids from said cells;

c. contacting said nucleic acids with a reagent that can specifically bind to a nucleic acid sequence encoding brevican, wherein said contact is under conditions that allow the specific binding of said reagent to said nucleic acid sequence encoding brevican; and d. identifying the presence of said glial cell by detecting the presence of said specifically bound reagent to said nucleic acid sequence encoding brevican.

35. The method of claim 34, wherein said nucleic acid is RNA.

36. The method of claim 34, wherein said reagent is the nucleotide sequence of claim 30.

37. The method of claim 34, wherein said reagent is the probe of claim 31.

38. A method of identifying the presence of a glial cell in a subject, comprising the steps of:

a. administering the antibody of claim 13 to the subject; and

b. detecting specific binding of said anti- brevican antibody to brevican in said subject, wherein said specific binding identifies the presence of a glial cell.

39. A method of directing axonal growth in a subject, comprising administering an effective amount of the substantially purified brevican of claim 1 to the subject.

40. A method of inhibiting axonal growth in a subject, comprising administering an effective amount of the substantially purified brevican of claim 1 to the subject.