WO1995013298A1 - Agrin receptor - Google Patents

Abstract

The present invention relates to agrin receptors, peptides associated with agrin receptors. The invention also relates to the production of antibodies to the agrin receptor. The instant invention provides for the use of agrin receptor in diagnostic assays, testing assays and labelling assays. The instant invention discloses specific amino acid sequences for agrin receptor subunits.

Description

TITLE: Agrin Receptor

Background of the Invention

Field of the Invention This invention relates to certain agrin binding receptors which are capable of binding to agrin in a calcium dependent fashion. This invention also relates to substantially pure oligopeptides from these isolated agrin receptors which can be characterized by the molecular weight and association with agrin binding. The invention also relates to the use of the isolated protein in assays and assay kits for the detection of compounds which can bind agonistically or antagonistically to the agrin receptor. The invention also relates to the use of the proteins of the invention for the screening of inhibitors or facilitators of binding of agrin to the agrin receptor. The invention also relates to the generation of polyclonal and monoclonal antibodies which will bind to the agrin receptor and the use of these antibodies as inhibitors, immunopurification agents, and labeling agents. The invention relates to the use of the agrin receptor portion as a means for affinity purifying agrin and agrin related proteins from in vitro and in vivo sources. The invention also relates to the use of agrin receptor protein as a specific inhibitor of agrin binding in vitro and in vivo. The invention relates to the use of agrin receptor and the identification of binding domains on agrin protein. The invention also relates to specific amino acid sequences that are part of agrin receptor subunits.

Description of the Related Art

Agrin, agrin related proteins and agrin receptors play a vital role in directing the formation and specialization of pre- and post-synaptic neuromuscular junctions

(Nastuk et al., 1991, Neuron 7: 807-818). Agrin and agrin related proteins have been characterized as being involved in the clustering of AChRs (acetylcholine receptors), acetylcholinesterase (AChE), and other post synaptic components on the surface of cultured myotubes (Fallon et al., 1985, Nature 315: 571-574). Agrin has been found associated with both normal and damaged synaptic basal lamina, which is capable of directing the specialization of both pre- and post-synaptic neuromuscular junctions

(Reist et al., 1987, /. Cell Biol. 105: 2457-2469). Agrin-related molecules have been found in motor neuron cell bodies and at synapses in the developing embryo (Fallon and Gelfman, 1989, J, Cell Biol 108: 1527-1535; Magill-Solc and McMahan, 1989, 7. Cell Bio. 107: 1825-33).

Agrin is best characterized by the induction of AChR clustering on the myotubule surface, however evidence indicates that this is not the result of agrin mediated cross linking of the AChR on the cell surface, but rather via intracellular signal transduction events via a unique receptor (Nastuk et al., 1991, Neuron 7: 807- 818). Agrin-induced AChR aggregates arise from the lateral migration of preexisting receptors, and extracellular calcium is required (Godfrey et al., 1984, J. Cell Bio. 99: 615-627; Henderson et al., 1984, J. Neurosci. 4: 3140-3150). Agrin induced clustering can be inhibited by phorbol esters, metabolic energy inhibitors, or infection with Rous sarcoma virus (Wallace, 1988, /. Cell Biol. 107: 267-278; Anthony et al, 1988, /. Cell Biol. 106: 1713-1721 ).

Agrin is a single chain, multidomain extracellular matrix protein of about 200 kD (reviewed by Nastuk and Fallon, 1993, Trends in Neurosci. 16: 72-76). Stable, bioactive proteolytic breakdown products of the molecule purified from Torpedo electric organ are 150 kD and 95 kD. The amino-terminal half of the agrin protein consists of nine-tandem repeats, each repeat homologous to Kazel-type protease- inhibitor domains. There is also approximately 39% homology with the Bl, B2 and S- laminin chains, which may implicate this area in binding to the basal lamina component entactin (nidogen). The carboxy-terminal half of the protein has homology to the cysteine-rich motifs of epidermal growth factor (EGF). The carboxy half roughly corresponds with the 95 kD portion, and can mediate all of agrin 's known functions. There are three alternative exon splice sites, two of which can have a single or no exon included, and one which can have one, two or no exon included. This can potentially give rise to 16 agrin isoforms, but since the alternative exons are small, they remain 98% identical. While the 95 kD form of agrin is able to induce AChR clustering, a truncated form of the protein called the 70 kD form will not induce clustering of AChR. (Nitkin et al., 1987, J. Cell Biol. 105: 2471-2478; Smith et al., 1992, Mol. Cell. Neurosci. 3: 406-417). A putative agrin receptor was first demonstrated to exist, by the present inventors, on chicken myotube plasma membrane showing agrin binding in the presence and absence of AChRs. The distribution of the agrin receptors was not dependent on the presence of AChR (Nastuk et al., 1991, Neuron 7: 807-818). Dissociated mouse embryonic (ED 9-10) spinal cord cells displayed a subpopulation of neurons which also demonstrated agrin binding (Nastuk and Fallon, 1991, Soc. Neurosci. Absti-. 17: 219).

It has been postulated that agrin or agrin related molecules may play a role in the development of pathological conditions which exhibit motomeuron degeneration. Agrin appears crucial to the proper formation of the neuromuscular junction, which occurs when a nerve terminal contacts a muscle fiber, and to the reestablishment of injured synaptic connections . The proper formation of such an intricate structure during development requires the localized cell surface specialization and accumulation of nicotinic acetylcholine receptors (nAChRs) and acetylcholinesterase (AChE) at the synapse beneath the nerve terminal (Patthy and Nikolics, 1993, Trends in Neurosci. 16: 76-81). When damage occurs, reinnervation of the muscle is topographically specific, preferentially reforming connections at original synaptic sites (McMahon and Slater, 1984, J. Cell Bio., 98: 1453-73). The biological activity of agrin is not limited to AChR aggregation on myotubues. It has been shown that a dozen molecules undergo redistribution in response to agrin (Nastuk and Fallon, 1993, Trends in Neurosci. 16: 72-76). High concentrations of neuro transmitter receptors characterize both neuromuscular junctions as well as neuron-neuron synapses in the brain and peripherial nervous system. There is evidence that agrin protein and mRNA is expressed by cells of the brain, and can be detected by western and northern blotting. Agrin mRNA is found not only in motor neurons, but also in Purkinje cells and retinal ganglion cells that project exclusively within the central nervous system (Smith et al., 1992 , Molec. Cell. Neurosci. 3: 406-417). Thus specific isolation, purification, and characterization of agrin receptor protein would be critical for the understanding and manipulation of the formation of complex synaptic junctions in the CNS (central nervous system) and PNS (peripherial nervous system). The purified agrin receptor would be invaluable for the demonstration of the presence of compounds which could bind agonistically, or antagonistically with the receptor. The purified receptor protein allows for the generation of antibodies which will bind to the protein and thereby allow for the labeling, identification, and purification of agrin receptor proteins from a variety of tissues from any creatures with a nervous system.

Since the agrin receptor and protein may also be involved in the development of other tissues as well. The agrin receptor protein can be used as a means for screening tissues for the presence or absence of agrin, as an indicator of pathological conditions, or as a means for manipulation of tissue interactions dependent on agrin. The agrin receptor protein allows for the identification of substances which will bind to the agrin receptor. The agrin receptor protein allows for the identification of substances which will enhance or inhibit the binding of agrin or agrin-related proteins with the agrin receptor.

Summary of the Invention

This invention provides for a substantially purified oligopeptide which is characterized by the ability to bind agrin, having a molecular weight of about 190 kD as measured by SDS polyacrylamide gel electrophoresis. In another embodiment, the substantially purified protein is characterized by the ability to bind agrin, and has one polypeptide of a molecular weight of about 190 kD and a second polypeptide of a molecular weight of about 50 kD, as measured by SDS polyacrylamide gel electrophoresis.

The invention also encompasses substantially purified oligopeptides or fragments thereof, of the protein identified as the agrin receptor which is characterized by the ability to inhibit the binding of agrin or agrin related proteins to the agrin receptor.

The invention encompasses the use of substantially purified oligopeptides or fragments thereof, of the protein identified as the agrin receptor, as inhibitors of the clustering activity associated with agrin binding at pre- and postsynaptic membranes. This clustering including those structures listed in table I.

The invention encompasses the use of substantially purified oligopeptides or fragments thereof, of the protein identified as the agrin receptor, as activators of the clustering activity associated with agrin binding at pre- and postsynaptic membranes. This clustering including those structures listed in table I. In a preferred embodiment the protein and fragments thereof are glycosylated.

The agrin receptor of the invention may be isolated and purified from many sources, among these would be nervous tissues of any vertebrate or invertebrate animal. In a preferred embodiment the agrin receptor is isolated from Torpedo electric organ. The receptor protein will allow for the determination of amino acid residue sequence information for the agrin receptor protein. This sequence information will allow for the determination of nucleic acid sequences which encode and translate the agrin receptor. Thus one embodiment of the agrin receptor of the instant invention is produced using recombinant DNA technology where by an expression vector is used to express, translate, and produce the peptides of the instant invention. This expression can take place in bacterial, yeast, mammalian, plant, fungal, or insect cells or variations thereof. The expression vector may be used for the introduction into embryonic cells for the production of transgenic animals, plants or insects, or for gene therapy.

The instant invention provides amino acid sequences which are part of the agrin receptor subunits as listed in Table III. Thus the instant invention provides for the construction of natural or syntheic peptides of the disclosed sequences, as well as for recombinant DNA vectors for the expression of peptides from corresponding DNA sequences, or the generation of antisense RNA from corresponding inverse DNA sequences. The present invention also provides for methods of screening for compounds which can inhibit binding of agrin to the agrin receptor, comprising the steps of combining an amount of protein of receptor with an amount of compound to be tested, followed by the administration of an amount of agrin, where one of the components of the assay is labeled so that a quantitative determination of binding can be made.

The present invention also provides methods of screening compounds for binding with the agrin receptor comprising the steps of combining an amount of receptor with an amount of compound to be tested, where one of the components is labeled so that a qualitative measurement of binding can be made. Such screening can be accomplished so as to identify potential neuro-toxins, neuro-trophic compounds, agonists, antagonist, or enzymes which may be a part of the processes of agrin binding and the effects of such binding to the receptor. The present invention also encompasses the production of antibodies which are immunoreactive with the agrin receptor protein. In a preferred embodiment, the antibodies are monoclonal antibodies. A particular embodiment is the monoclonal antibody, mAb-AgRl, which is produced by the cell line AgRl. The present invention thus encompass the use of agrin protein to produce antibodies which will immunoreact with the agrin protein and inhibit the specific binding of the monoclonal antibody mAb-AgRl with the agrin receptor.

The present invention thus encompasses cell lines which can express antibodies that are able to bind to the agrin receptor. A specific embodiment of this is the cell line AgRl which produces the monoclonal antibody mAb-AgRl. The antibodies of the present invention can bind to the various epitopes of the agrin receptor, these epitopes encompassing protein quaternary, tertiary and secondary structure epitopes which are based upon the primary structure of the agrin receptor protein. The antibodies of the present invention can recognize carbohydrate epitopes associated with the agrin receptor protein and the functional derivatives of the carbohydrate moieties generated by natural or induced modification.

Thus the present invention also provides for methods for stimulating or inhibiting the activity of the agrin receptor consisting of the administration of an effective amount of the monoclonal antibody. Also provided is a method for stimulating or inhibiting the aggregation of AChR (acetylcholine receptors) on neural tissues comprising the administration of an effective amount of mAb-AgRl.

The antibodies of the instant invention can be used for immunopurification, immunolabling, immunodiagnostic and immunoassay procedures for the detection and isolation of agrin receptor proteins. These methods can be embodied in kits consisting of aliquoted antibody solution and control solution for the quantification of agrin receptor proteins, or agrin content of various tissues.

The present invention also teaches the use of a 70 kD agrin protein which is capable of binding to agrin receptors without stimulating AChR aggregation. This being a useful means of binding agrin to target membranes with agrin receptors, without subsequent AChR clustering.

Thus one with ordinary skill in the art will be able to use the teaching of the present specification to formulate the various embodiments described and leading naturally from the instant invention. The following examples are meant to be illustrative of the present invention, and are in no way intended to limit the scope of the instant invention.

Brief Description of the Drawings Figure 1 illustrates the Ligand Overlay Assay of Torpedo electric organ postsynaptic membranes. Far left lane: lane 1 Silver stain of immunoaffinity (mAb-AgRl) purified agrin receptor polypeptides: Note the 190 kD and 50 kD proteins. Right panel autoradiographs; lane 2, immunoaffinity purified receptor probes on overlay blot; lane

3, crude postsynaptic membranes (from sucrose gradient) probed with agrin overlay. Lane 4, 5: same sample as in lane 3, overlay in the absence of calcium (lane 4), overlay in the absence of agrin (lane 4).

Figure 2 illustrates SDS-PAGE analysis of the purified 190 kD and 50 kD polypeptides.

Figure 3 (a) is a diagram showing the elution profile of TEO agrin binding protein with inset showing analysis of specific fractions by SDS-PAGE and silver-staining. 3

(b) is a graph showing the agrin binding activity is retained even when AChR is depleted.

Figure 4 is a visualized blot of intact and alkaline stripped membrane proteins isolated from TEO and a correlation with agrin binding as measured using ¹²⁵I-labeled anti- agrin monoclonal antibodies.

Figure 5 is a graph illustrating the calcium sensitive, saturable binding of agrin to synaptic membrane receptors as measured by RIA.

Figure 6 (a) is a graph illustrating the calcium sensitive agrin binding by membrane receptors as measured by RIA as intact or stripped membranes. 6 (b) is a graph demonstrating that Agrin binds to rat brain synaptosomes, and is calcium sensitive, as shown by RIA.

Figure 7 is a diagram of the agrin protein and the relationship between the 90 kD and

70 kD fragments.

Figure 8 is a western blot demonstrating the binding of the 90 kD and 70 kD fragments to agrin receptors on isolated membranes.

Figure 9 is a diagram illustrating the relationship between the Agrin Receptor Subunit sequences and the human dystroglycan precursor protein. Detailed Description of the Invention

Example 1 Preparation and Isolation of Torpedo californica Membrane Proteins

Postsynaptic membranes from Torpedo californica electric organ (TEO) were isolated and stripped of peripheral membrane proteins and solubilized in 25 mM octyl-β-D-glucopyranoside (OG). Membrane fractions from TEO were prepared as described (Burden et al., 1983, Cell 35: 687-92) with modifications.

Briefly, electric organs were dissected from Torpedo californica (Pacific Biomarine) that had been anesthetized with 3-aminobenzoic acid ethyl ester (MS 222) and pithed. The tissue was immediately frozen in liquid nitrogen, homogenized in 25 ml buffer A (400 mM NaCl, 50 mM Tris, 10 mM EDTA, 10 mM EGTA, 10 ug/ml phenylmethylsulfonyl chloride (PMSF), 1.06 mg/ml diisopropyl fluorophosphate, 0.02% sodium azide, pH 7.4). The homogenate was centrifuged twice for 10 min at 6,000 x g and the resulting supernatant then centrifuged for 1 hr at 100,000 x g. The pellet was resuspended in buffer B (10 mM NaH2PO4, 1 mM EDTA, 1 mM EGTA, 100 ug/ml PMSF, 0.02% sodium azide, pH 7.4), homogenized, layered onto a continuous 25%-40% sucrose gradient, and centrifuged for 6 hr at 100,000 x g. Acetylcholine receptor (AChR)-rich fractions determined by SDS-PAGE (sodium dodecyl sulfate-poly acrilamide gel electrophoresis) , were then pooled. The four AChR subunits were identified by comigration with purified AChR generously provided by A. Karlin. Protein concentrations, determined according to Bradford (1976, Anal. Biochem. 72: 248-254), ranged from 0.4 to 0.6 mg/ml for the AChR-rich pools. Membranes from Torpedo liver were prepared exactly as described above.

Alkaline stripping (pH 11.0, 1 hr) of peripherial membrane proteins was performed as described (Neubig, et al., 1979, P.N.A.S. USA 76: 690-694). Briefly, the isolated membranes were resuspended in water, and the pH was adjusted to pH 11.0 with 0.1 M NaOH and kept on ice for 1 hr. The membranes were then centrifuged at 100,000 x g for 1 hr to recover the membranes. After treatment, the pellets were resuspended in the original volume of buffer B and used for further analysis. In some experiments, membranes were treated at pH 12.0 using the same protocol. Material extracted at pH 11 or pH 12 was analyzed for agrin binding and glycoprotein composition as described below. Membranes aggregated irreversibly after treatment at higher pH, and were not examined for agrin binding.

Figure 1 illustrates the Ligand Overlay Assay of Torpedo electric organ postsynaptic membranes. The far left lane (lane 1) shows the silver-stain of immunoaffinity purified agrin receptor polypeptides from the mAb-AgRl immunoaffinity column. This purified product resolved into two bands of about 190 kD and 50 kD (arrows). Autoradiography of the agrin overlay assay shows that agrin bound (¹²⁵I labeled-anti-agrin mAb detected) to the single 190 kD band (lane 2). Agrin bound to a single band in the presence of Ca^44" (lane 3), but not in the absence of Ca^"1-1" (lane 4). A second control lane received no agrin (lane 5). Molecular weight standards are from the top, 200, 116, 97, 68, 45, and 29 kD and are indicated by the plain lines. The protein content of unpurified postsynaptic membranes is illustrated in Figure 2 (lane 1, silver stain) and contains multiple proteins.

For identification of glycosylated peripherial membrane proteins extracted by alkaline treatment, supernatants were neutralized with 20 mM HEPES, pH 7.2, made 1 mM CaCl2 and incubated with wheat germ agglutinin (WGA)-coupled to agarose beads (Vector Laboratories, Burlingame, CA). The beads were washed and boiled in SDS sample buffer. Eluted glycoproteins were separated on 7.5% SDS-PAGE gels, blotted onto nitrocellulose, and incubated with biotinylated WGA (5 ug/ml, Vector) followed by avidin-biotin-HRP (horse radish peroxidase). WGA-binding glycoproteins were visualized by chemiluminescence (ECL, Amersham).

Solubilization of postsynaptic membranes was accomplished, optimally, by incubating untreated or alkaline stripped membranes in 25 mM octyl-β -D- glucopyranoside in 50 mM NaCl, 0.1 mM PMSF in sucrose-free buffer B, pH 7.4 for 18 hours at 4°C. Detergent- treated membranes were centrifuged at 100,000 x g f or 3 hr and the supernatant collected for further analysis. The conditions for solubilization of the agrin receptor is distinct from that for AChR, which was found to be 34 mM octyl-β-D-glucopyranoside, 0 mM NaCl as determined by SDS-PAGE and ^{1 5}I-α- bungarotoxin binding.

Example 2 Determination of Agrin Receptor by SDS-PAGE/Ligand Overlay Assay and by Non-denaturing Conditions using Ligand Affinity Chromato raphy Agrin was purified over 3,000 fold from Torpedo electric organ as described previously (Nastuk et al., 1991, Neuron 7: 807-818). Agrin bioreactivity is expressed in Units (U)/ml, which refers to the amount of agrin required to induce half-maximal AChR clustering on myotubes cultured in 1 ml of media (Godfrey et al., 1984, J. Cell Biol. 99: 615-627). One unit is estimated to correspond with 10"¹² to 10^{~ 13} M agrin (Nitkin et al., 1987, J. Cell Biol. 105: 2471-2478).

Briefly, the activity of purified agrin was tested as follows. Membranes were pelleted and resuspended in HEPES buffered MEM (pH 7.2; MEM-H) with 10% horse serum/1% BSA and incubated for 30 min at 4°C with 3 U of purified agrin in 20 ul of membranes with gentle agitation. Incubations (in duplicate) were carried out in the presence or absence of 2 mM EGTA. Maximal agrin activity under these conditions was determined by incubating agrin without membranes. After incubation, pelleted membranes and 15 ul of each supernatant was used for overnight stimulation of chick myotube cultures. Myotubes were labeled with rhodamine-α-bungarotoxin and AChR clustering activity determined. The clustering ability of agrin effects many synapse components, these are summarized in table I, phase indicated in relation to agrin binding (from Nastuk and Fallon, 1993, Trends in Neurosci. 16: 72-76).

Table I Synaptic component Phase of π

Plasma membrane

Acetylcholine Receptor early

Acetylcholineseterase (globular) early

Butylcholinesterase (globular) early

Agrin receptor early

43 kD protein early

Extracellular Matrix

Heparan sulphate proteoglycan late

Laminin late

Muscle agrin late

Acetylcholinesterase (A 12 asymmetric) ?

Cvtoskeleton α-actinin late

Filamin late

Vinculin late

The presence of agrin receptors was demonstrated as follows. Alkaline- stripped postsynaptic membranes were resuspended in 150 U of agrin for 1 hr at 4°C in 50 mM NaCl, 5% glycerol, 10 mM NaPO4, pH 7.4 with 1 mM calcium or 1 mM EGTA. The membranes were solubilized in SDS sample buffer with no reducing agents, separated on 7.5% SDS-PAGE gels. The gels were then blotted onto nitrocellulose. The bound agrin was demonstrated by the binding of mAb 11D2, an anti-agrin monoclonal antibody (Nitkin et al., 1987, J. Cell Bio. 105: 2471-2478). The mAb 11D2 was visualized by incubating the blots with 11D2 and biotinylated horse anti-mouse IgG, and avidin-biotin-HRP (horse radish peroxidase) (Vector). Bound antibody was detected by chemiluminescence (ECL, Amersham), and indicated the presence of agrin bound to protein blotted to the nitrocellulose.

In other experiments, the solubilized proteins separated by SDS-PAGE were either stained with silver or blotted to nitrocellulose. The separated blotted proteins were then blocked with 5% non-fat dry milk. The indicated areas were then treated with purified TEO agrin in the presence of 1 mM Ca⁺⁺ or 1 mM EDTA. The blots were then incubated with ¹²⁵I-anti-agrin mAb 6D4 (Fallon et al., 1985, Nature 315: 571-4), washed, and dried; agrin-binding bands were then revealed by autoradiography.

Membrane proteins solubilized in mild, non-ionic detergent retained agrin binding properties, as indicated in Table I. Agrin binding fractions of gel-filtered membrane proteins (see Table I and Figure 1) were iodinated (lodogen method, Pierce). Labeled proteins were pre-incubated with agrin for 1 hr in the presence of 1 mM CΆ⁺⁺ or EGTA to form ligand-receptor complexes, and were then applied overnight to anti-agrin immunoaffinity columns (mAb-5Bl and mAb-6D4 cross- linked to Protein A agarose beads by the method of Schneider, C. et al., 1982, J. Biol. Chem. 257: 10766-10769) , washed, and eluted with 10 mM EGTA.

Figure 2 illustrates SDS-PAGE analysis of the purified receptor yielding 190 kD and 50 kD polypeptides (arrows). Lane 1 is a silver-stain of initial biochemical purification of TEO membrane proteins showing many proteins present. Lane 2 are the postsynaptic membranes after purification by immunoaffinity column (same as lane 1 of Figure 1). Lanes 3 and 4 shows the elutes from agrin affinity chromatography. The peak fractions from the gel-filtration column (Figure 3) were iodinated and applied to agrin affinity columns in the presence of calcium (lane 3) or the absence of calcium (lane 4). Lane 3 is an autoradiogram of the mAb-AgRl immunoprecipitate of the material eluted by EGTA from the agrin ligand- affinity column with calcium present. The column bound ¹²⁵I-labeled 190 kD and 50 kD polypeptides (arrows) in the presence but not in the absence of calcium. Molecular weight standards are the same as Figure 1. Lines to the right indicate (from the top) Na-ATPase and AChR γ, gamma, β and α subunits. Additional bands of 48, 55 and 65 kD also bound in a Ca⁺⁺ independent manner.

Example 3 Determination of Agrin Receptor under Non-denaturing Conditions using

ImmunoAffinitv Chromatography

Solubilized membrane proteins isolated as above were made 0.5 M in NaCl and also applied to immunoaffinity columns which were prepared with mAb-AgRl covalently linked to the column matrix (Schneiden et al., 1982, J. Biol. Chem. 257: 10766-10769). The columns were washed in 1.0 M NaCl. The 190 kD and 50 kD polypeptides coeluted from this column. This is shown in Figure 2. The agrin binding activity coπesponded with the elution of these polypeptides, and this is shown in Figure 3. The calcium sensitive eluates from the agrin ligand affinity columns (from example 2) containing the 190 kD and 50 kD polypeptides were then applied to mAb- AgRl immunoaffinity columns. The final eluate from these immunoaffinity columns was found to contain the same 190 kD and 50 kD polypeptides, and to retain agrin binding activity as demonstrated by the ligand overlay assay. As shown with the ligand overlay assays of the crude membrane preparations, agrin bound to the 190 kD purified polypeptide, and this is shown in Figure 2.

Postsynaptic membrane from Torpedo californica electric organ were prepared as described previously. Membrane proteins were alkaline-treated as described below, to remove peripheral membrane proteins, solubilized in 25 mM OG, and fractionated on Superose 6. Agrin binding was quantitated by a solid phase, ligand-binding RIA (described below), under sub-saturating conditions.

Figure 3 (a) is a diagram showing the elution profile of TEO agrin binding protein. The 190 kD and 50 kD polypeptides migrate as a complex during gel filtration chromatography. Agrin binding activity and OD28O of membrane proteins fractionated by gel-filtration. Shown as an inset is SDS-PAGE and silver-stain of mAb-AgRl immunoprecipitates of the indicated fractions. Note there is only a single peak of agrin binding activity in these membranes. A 190 kD and 50 kD protein co- eluted with this peak of agrin binding. Peak fractions of AChR and Na, K-ATPase are indicated by the arrowheads at fraction #29, and #33, respectively. Column fractions were applied to a mAb-AgRl immunoaffinity column (mAb crosslinked to Protein-A agarose beads), washed, and eluted with 0.1 M diethylamine (pH 11.5).

Figure 3 (a) shows the agrin binding activity as eluting in a single peak with the 190 kD/50 kD complex. This peak eluted 1 to 2 fractions earlier (larger stokes radius) than the elution of the AChR complex, which has a native Mr of 250 kD (data not shown). Results are the means ± S.E.M. of 4 experiments. AChR concentration was determined by ¹²⁵I-α-bungarotoxin binding and confirmed by SDS-PAGE; Na/K- ATPase was revealed by SDS-Page and silver staining. Insert immunoaffinity eluates were separated by SDS-PAGE and silver stained.

The agrin binding was retained, even with the depletion of AChR by binding to affinity columns. This is direct biochemical evidence that the agrin receptor and the AChR are different molecules. Briefly, solubilized postsynaptic membranes were applied to α-bungarotoxin affinity columns, to bind AChR, and the flow-through levels of AChR and agrin receptors was measured. These columns remove >95% of the AChR from the solubilized membranes without depleating the agrin receptor. It is also noteworthy that the membranes used for this test were not alkaline stripped, and thus had a normal complement of peripherial membrane proteins, including cytoskeletal elements. Moreover, the AChR and the agrin receptor could be separated from the AChR under mild detergent, salt, and pH, conditions which favor the preservation of macromolecular complexes. This data suggests that the AChRs and the agrin receptors are not tightly associated with one another in the membrane.

Figure 3 (b) is a bar graph showing the results of the use of α-bungarotoxin affinity columns, and the flow-through levels of AChR and agrin receptors. Solubilized membranes were incubated with buffer only (start), or with 10^"8 M biotinylated-α-bungarotoxin (AChR depleted), or without α-bungarotoxin (control), followed by strptavidin-agarose beads. More than 95% of the AChR was removed as measured by SDS-PAGE and scanning densitometry of the AChR α-subunit. Agrin receptor binding of agrin was measured by RIA. The agrin receptor does not co- precipitate with AChRs.

Example 4 Characterization of Agrin Receptor

Both the 190 kD and the 50 kD polypeptides are integral membrane proteins since they are resistant to alkaline extraction (Figure 2), and have carbohydrate modifications demonstrating sensitivity to glycosidases. Alkaline treatment (pH 11) of Torpedo membranes is known to extract peripherial membrane proteins such as 43 kD, 58 kD, and dystrophin-like polypeptides (Neubig et al, 1979, JV.Λ.S. USA 76: 690-694; Yeadon, et al., 1991, J. Cell Biol. 115: 1069-1076). However, these stripped membranes retain integral membrane proteins with hydrophobic domains, such as AChR. Our results show that alkaline-stripped postsynaptic membranes fully retain agrin binding activity (Figures 4 and 5). Even with extraction at pH 12, the same glycoproteins are extracted under both conditions, and the membranes retained agrin binding activity. Figure 4 is a visualized blot of intact and alkaline stripped membrane proteins isolated from TEO and a correlation with agrin binding as measured using Horseradish peroxidase (HRP)-labeled anti-agrin monoclonal antibodies.

Figure 4 demonstrates alkaline stripping extracts peripherial membrane proteins as shown in lanes 1 vs. 2. The stripped membranes retaining integral membrane proteins after treatment at pH 11, lane 2. Intact TEO membrane is shown in lane 1 , all proteins visualized by coomassie blue staining. The well characterized 43 kD and 300 kD dystrophin-like cytoplasmic peripherial membrane proteins are removed by this treatment (arrows). The position of the four AChR subunits are indicated, δ, γ, β, α. Western blot probed with WGA of proteins extracted at pHl 1 (lane 3) and extracted at pH 12 (lane 4) show that the same set of peripherial membrane glycoproteins are extracted by both treatments. Figure 5 is a bar graph illustrating the agrin binding to intact and pH 1 1 stripped membranes, in a calcium sensitive manner. Equal volumes of intact or alkaline stripped membranes were assayed by agrin binding RIA (as described in example 6). The results show that agrin binding is quantitatively recovered following membrane alkaline stripping, with complete retention of calcium sensitivity.

Table II shows the results of lectin affinity chromatography depletion of the agrin receptor. Various lectins were used and the residual actin binding activity measured. Alkaline stripped, detergent solubilized postsynaptic membranes were incubated with the indicated lectin-substituted agarose beads. After incubation, the level of agrin binding activity in the supernatants was determined by RIA. This demonstrates that the agrin receptor is most likely glycoprotein, and sensitive to the carbohydrate structure.

Table π Lectin % Residual Agrin Binding

Agarose Beads 100 ± 2.1

Peanut agglutinin 4.1 ± 3.4

Wheat germ agglutinin 5.2 ± 3.2

Ricinus Cornmunis 1 10.2 ± 4.1

Concanvalin A 14.5 ± 5.0

Dolichos biflorus 33.0 ± 4.8

S oybean agglutinin 78.6 ± 3.0

Ulex europeaeus 87.2 ± 4.5

Pisum sativum 96.3 ± 3.6

Example 5 Production of mAb-AgRl

Monoclonal antibody to the agrin receptor was generated using standard fusion tehchniques, with the following modification. Mice were immunized by injection of alkaline- stripped postsynaptic membranes from TEO in RIBI adjuvant (Ribi, 1984, J. Biol. Resp. Mod. 3: 1-9) subcutaneously at the ankle. After 6 injections at 5-7 day intervals the mice were sacrificed and the popliteal an inguinal lymph nodes harvested and fused with the Agl08 myeloma cell line. Hybridoma supernatants were screened for antibodies that immunoprecipitated the agrin receptor. Briefly 30 U of agrin were mixed with 15 ul of solubilized postsynaptic membranes to form agrin-agrin receptor complexes. The hybridoma supernatants were then added, incubated overnight, and then applied to sepharose bound anti-mouse Ig. Agrin/agrin-receptor/anti-agrin receptor complex was thus precipitated associated with the sepharose beads. The agrin remaining in the supernatant was measured by a two site RIA where anti-agrin mAb-6D4 was immobilized on microwells, agrin (supernatant) added, and the captured agrin quantitated by incubation with iodinated anti-agrin mAb-5Bl. A reduction in the amount of agrin in the supernatant indicated that the supernatnant contained the desired anti-agrin receptor monoclonal antibodies. The cell line producing mAb-AgRl has been deposited with the American Type Culture Collection (ATCC, Rockville, M.D.) under the Budapest Treaty and identified as mouse myeloma/B-cell hybridoma, anti-AgR-MAb-3B3, with the assigned designation number HB 11530, on January 21, 1994.

Example 6 Agrin Receptor RIA

In order to demonstrate agrin binding to TEO membranes, a RIA (radio immunoassay) was developed which used agrin receptor or membranes bound to assay wells, the capture of agrin, and the detection of captured agrin using labelled monoclonal anti-agrin antibodies.

Briefly, intact or solubilized membranes were plated onto polylysine-coated wells (Removawell 96-well strips, Dynatech). The plates were washed with two changes of MEM-H and blocked with MHB with 10% horse serum. Wells were then incubated in agrin diluted in MHB (200 U/ml, unless otherwise noted) for 30 min, washed, and incubated a further 30 min in 1 ug/ml ¹²⁵I-anti-agrin-monoclonal antibody 6D4. Nonspecific antibody binding was determined by the addition of 100 ug/ml unlabelled 6D4 during the incubation in iodinated antibody, and was less than 10% of specific values. The wells were washed twice in MEM-H then immersed in two changes of HBSS with 0.1% BSA for 5 min each, dried, separated and counted. Three replicate wells per condition were analyzed in each experiment, data are expressed ± S.E.M..

To quantitate the effects of calcium on agrin binding to AChR-rich membranes, the initial blocking step, the agrin incubation, and the subsequent wash in MEM-H were performed with or without 2 mM EGTA. In some experiments, the peptides GRGDSP (100 uM; Gly-Arg-Gly-Asp-Ser-Pro; Telios) and LRE (300 uM; Leu-Arg-Glu; from J. Sanes) were added during the agrin binding step, although neither peptide inhibited binding. In this RIA approach, the binding of exogenous agrin to the membranes is measured as a function of bound radiolabelled anti-agrin antibody. Figure 5 -3- shows that the calcium-dependent agrin binding to TEO membranes is saturable. In the absence of calcium, binding is reduced by more than 5- fold, and rises linearly with increasing ligand concentration. No agrin binding was observed on Torpedo liver membranes, with or without the presence of calcium. Half- maximal agrin binding is observed at an agrin concentration of about 250 U/ml, this coπesponding to < 10^{~ 10} M agrin. This low ligand concentration suggests a high affinity binding to the agrin receptor. Figure 5 is a graph showing the binding of agrin to membrane receptors as measured by RIA (described in example 6) and demonstrates saturation. In the absence of calcium, agrin binding is not saturable.

Figure 6 (a) is a graph which demonstrates the calcium sensitive binding of agrin to the membrane receptors as measured by RIA. Agrin binding to postsynaptic membranes was measured under varying calcium concentrations. Optimal agrin binding is observed at 1-2 mM calcium, which agrees with the calcium optima for agrin's AChR clustering activity.

Figure 6 (b) is a graph which demonstrates that agrin binds to adult rat brain synaptosomes in a calcium dependent manner. Rat brain synaptosomes were prepared according to Watkins et al., (1984, J. Physiol..London 153: 35P). Synaptosomes were immobilized on poly-lysine coated microwells and agrin binding was determined as above. These data indicate that agrin receptors are expressed in the rat brain.

In order to clarify if the 50 kD polypeptide was capable of binding agrin directly, or was associated with the 190 kD protein, gel filtration size-exclusion choromatography was performed to separate the 50 kD protein solubilize in non-ionic detergent. Although the majority of the 50 kD polypeptide eluted with the 190 kD protein, some eluted in a later peak. No agrin binding to the isolated 50 kD polypeptide was observed. OG-solubilized membrane proteins from alkaline-treated TEO membranes were separated on Superose 6. Agrin binding activity was determined by solid phase RIA . The 50 kD polypeptide band exhibited negligible agrin binding in the ligand overlay assay and isolated 50 kD polypeptide showed no agrin binding activity either.

Example 7 Agrin Binding Domain of 95 kD and 70 kD Polypeptides Purification of agrin protein yielded two major polypeptide fragments referred to as 95 and 70 kP, based on the apparent Mr under reducing conditions. These fragments have identical amino termini, with the 70 kP polypeptide being truncated at its carboxyl end. We find that both the 95 kP and the 70 kP fragments bind to the TEO membranes as measured by RIA. This result shows that at least one of the receptor binding domains of agrin is located in the region delineated by the 70 kP fragment. It is known that the 95 kP fragment induces AChR clustering of myotubes, while the 70 kP fragment does not (Nitkin et al., 1987, J. Cell Biol. 105: 2471-2478). Therefore, a domain within the 70 kD fragment is necessary for receptor binding but not sufficient alone to induce AChR clustering. Figure 7 is a schematic diagram of the structure of agrin. The amino terminal

1/3 of Torpedo agrin (shaded portion) has not been sequenced; a consensus domain orgainization for this region based on rat and chick sequence is shown for illustrative purposes. The locations of the 95 kD and 70 kD agrin polypeptides derived from full length agrin are indicated. Also shown are the known sites of alternative exon splicing (arrowheads; the number of amino acids encoded by each insert is indicated above them). Motifs predicted by sequence analysis: filled box indicate signal sequence; hatched region indicate follistatin; label LNIII indicates the laminin domain III; ST indicates the serine-threonine rich sequence; E indicates ΕGF like region; LN-A indicates type A laminin.

Figure 8 shows the identification of agrin polypeptides which bind to postsynaptic membranes. Shown is a western blot probed with anti-agrin antibody 11D2 (lanes 1-4) or an iπelevant IgG (lanes 5-8). Lanes 1, 5: starting agrin preparations of 150 U. Lanes 2-8: pellets of alkaline stripped postsynaptic membranes (25 ug) after incubation with: no agrin (lane 2, 6); or agrin (150 U) in the presence of calcium (lane 3, 7), or the absence of calcium (lane 4, 8). Both the 95 kD and the 70 kD agrin polypeptide fragments bind postsynaptic membranes in a calcium dependent manner. It is also noteable that the starting membrane preparations showed a lack of detectable agrin.

Example 8 Peptide Microsequencing

Immunoaffinity-purified agrin receptor subunits (approximately 20 ug of total protein) were electrophoresed on a discontinuous SDS-PAGE separating gel (upper half 5%-7%, lower half 11% acrylamide). The polypeptides were electroblotted for 3 hr. at 2 A onto Immobilon PVDF membranes (Millipore) in 10% methanol, 25 mM Tris, 192 mM glycine (pH 8.3). After visualization with Ponceau S, the 190 kDa- and 50 kDa-bearing bands were excised, minced, and subjected to in situ enzymatic digestion with 0.5 ug of endoproteinase Lys-C (190 kDa polypeptide) or 0.25 ug of trypsin (50 kDa polypeptide), essentially as described (Fernandez, et al., 1994, Anal. Biochem. 218:112-117). The released peptides were separated with a Hewlett Packard 1090M HPLC system equipped with a UV diode aπay detector, using a Brownlee Aquapore C8 microbore column (1 x 250 mm) with 0.1% TFA as solvent A and 0.08% TFA in acetonitrile-water (70:30 v/v) as solvent B. A linear gradient was developed from 0 to 55% in solvent B over a period of 90 min at a flow rate of 50 ul/min. The peptide peaks were collected manually in 1.5 ml microfuge tubes. Selected fractions from each digest were subjected to automated Edman degredation on an Applied Biosystems 477 A sequencer equipped with a Model 120A In-line PTH analyzer. The following sequences were obtained for each subunit, (Table HI below). Table HI

190 kDa Subunit

1. LKEQQLLSESSWVKFQSTXQ [SEQ ID NO. 1]

2. HEYFMHAAXKGGLTAVXXFE [SEQ ID NO. 2] 3. GLAQAFGDRXSSTVTLLAIS [SEQ ID NO. 3]

50 kDa Subunit

4. XVLVXXIGXTLPLEXQ [SEQ ID NO. 4]

5. SLGSQLADADGRPTPAFT [SEQ ID NO. 5] 6. GVPIIFADELDDSKPPPSSSVXLII [SEQ ID NO. 6]

The native agrin receptor is a heteromeric complex of two membrane glycoproteins (190 kDa and 50 kDa) that are related to the dystrophin-associated glycoproteins α- and β-dystroglycan respectively. The sequences obtained show that the peptides are related to the dystroglycans. A sequence alignment with the human dystroglycan precursor sequence [SEQ ID NO. 7] is shown in Figure 9.

SEQUENCE LISTING

(1) GENERAL INFORMATION:

(i) APPLICANT:

NAME: Worcester Foundation for Experimental Biology

STREET: 222 Maple Avenue

CITY: Shrewsbury

STATE: Massachusetts

COUNTRY: USA

POSTAL CODE (ZIP) : 01545

TELEPHONE: 508-842-8618

TELEFAX: 508-842-9632

NAME: Justin Fallon

STREET: 62 Massachusetts Ave.

CITY: Harvard

STATE: MA

COUNTRY: USA

POSTAL CODE (ZIP) : 01451-0266

NAME: Mark A. Bowe

STREET: 3 Ureco Terrace

CITY: Worcester

STATE: MA

COUNTRY: USA

POSTAL CODE (ZIP) : 01536

NAME: Beth A. McKechnie

STREET: 6 Kerrie Circle

CITY: Franklin

STATE: MA

COUNTRY: USA

POSTAL CODE (ZIP) : 02038

NAME: Mary A. Nastuk

STREET: 13 Dean St.

CITY: Hudson

STATE: MA

COUNTRY: USA

POSTAL CODE (ZIP) : 01749

NAME: John D. Leszyk

STREET: 8 Highland View Dr.

CITY: Sutton

STATE: MA

COUNTRY: USA

POSTAL CODE (ZIP) : 01590

[ii) TITLE OF INVENTION: AGRIN RECEPTOR

(iii) NUMBER OF SEQUENCES: 7

(iv) COMPUTER READABLE FORM:

(A) MEDIUM TYPE: Floppy disk

(B) COMPUTER: IBM PC compatible

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

(D) SOFTWARE: Patentin Release #1.0, Version #1.25 (EPO)

(vi) PRIOR APPLICATION DATA:

(A) APPLICATION NUMBER: US 08/149,966

(B) FILING DATE: 10-NOV-1993 (2) INFORMATION FOR SEQ ID NO: 1:

(i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 20 amino acids

(B) TYPE: amino acid

(C) STRANDEDNESS: single

(D) TOPOLOGY: linear (ii) MOLECULE TYPE: peptide

(xi) SEQUENCE DESCRIPTION: SEQ ID NO: 1:

Leu Lys Glu Gin Gin Leu Leu Ser Glu Ser Ser Trp Val Lys Phe Gin 1 5 10 15

Ser Thr Xaa Gin 20 (2) INFORMATION FOR SEQ ID NO: 2:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 20 amino acids

(B) TYPE: amino acid (C) STRANDEDNESS: single

(D) TOPOLOGY: linear

(ii) MOLECULE TYPE: peptide (xi) SEQUENCE DESCRIPTION: SEQ ID NO: 2:

His Glu Tyr Phe Met His Ala Ala Xaa Lys Gly Gly Leu Thr Ala Val 1 5 10 15 Xaa Xaa Phe Glu

20

(2) INFORMATION FOR SEQ ID NO: 3 : (i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 20 amino acids

(B) TYPE: amino acid

(C) STRANDEDNESS: single

(D) TOPOLOGY: linear

(ii) MOLECULE TYPE: peptide

(xi) SEQUENCE DESCRIPTION: SEQ ID NO: 3: Gly Leu Ala Gin Ala Phe Gly Asp Arg Xaa Ser Ser Thr Val Thr Leu 1 5 10 15

Leu Ala lie Ser 20

(2) INFORMATION FOR SEQ ID NO: 4:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 16 amino acids (B) TYPE: amino acid

(C) STRANDEDNESS: single

(D) TOPOLOGY: linear (ii) MOLECULE TYPE: peptide

(xi) SEQUENCE DESCRIPTION: SEQ ID NO: 4:

Xaa Val Leu Val Xaa Xaa lie Gly Xaa Thr Leu Pro Leu Glu Xaa Gin 1 5 10 15

(2) INFORMATION FOR SEQ ID NO: 5: (i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 18 amino acids

(B) TYPE: amino acid

(C) STRANDEDNESS: single

(D) TOPOLOGY: linear

(ii) MOLECULE TYPE: peptide

(xi) SEQUENCE DESCRIPTION: SEQ ID NO: 5: Ser Leu Gly Ser Gin Leu Ala Asp Ala Asp Gly Arg Pro Thr Pro Ala 1 5 10 15

Phe Thr

(2) INFORMATION FOR SEQ ID NO: 6:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 25 amino acids (B) TYPE: amino acid

(C) STRANDEDNESS: single

(D) TOPOLOGY: linear

(ii) MOLECULE TYPE: peptide

(xi) SEQUENCE DESCRIPTION: SEQ ID NO: 6:

Gly Val Pro lie lie Phe Ala Asp Glu Leu Asp Asp Ser Lys Pro Pro 1 5 10 15

Pro Ser Ser Ser Val Xaa Leu lie lie 20 25

(2) INFORMATION FOR SEQ ID NO: 7 :

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 445 amino acids

(B) TYPE: amino acid

(C) STRANDEDNESS: single (D) TOPOLOGY: linear

(ii) MOLECULE TYPE: protein

(ix) FEATURE: (A) NAME/KEY: Protein

(B) LOCATION: 1..445

(D) OTHER INFORMATION: /label= HUMDAGI

/note= "human dystroglycan precursor 451-895' (ix) FEATURE:

(A) NAME/KEY: Region

(B) LOCATION: 90..109 (D) OTHER INFORMATION: /note= "190 kDa fragment homology"

(ix) FEATURE:

(A) NAME/KEY: Region (B) LOCATION: 182..201

(D) OTHER INFORMATION: /note- "190 kDa fragment homology"

(ix) FEATURE:

(A) NAME/KEY: Region (B) LOCATION: 123..142

(D) OTHER INFORMATION: /note= "190 kDa fragment homology"

(ix) FEATURE:

(A) NAME/KEY: Region (B) LOCATION: 204..219

(D) OTHER INFORMATION: /note= "50 kDa fragment homology"

(ix) FEATURE:

(A) NAME/KEY: Region (B) LOCATION: 226..243

(D) OTHER INFORMATION: /note= "50 kDa fragment homology"

(ix) FEATURE:

(A) NAME/KEY: Region (B) LOCATION: 301..324

(D) OTHER INFORMATION: /note= "Transmembrane"

(ix) FEATURE:

(A) NAME/KEY: Region (B) LOCATION: 345..369

(D) OTHER INFORMATION: /note= "50 kDa fragment homology"

(xi) SEQUENCE DESCRIPTION: SEQ ID NO: 7: Lys Lys Pro Arg Thr Pro Arg Pro Val Pro Arg Val Thr Thr Lys Val 1 5 10 15

Ser lie Thr Arg Leu Glu Thr Ala Ser Pro Pro Thr Arg lie Arg Thr 20 25 30

Thr Thr Ser Gly Val Pro Arg Gly Gly Glu Pro Asn Gin Arg Pro Glu 35 40 45

Leu Lys Asn His lie Asp Arg Val Asp Ala Trp Val Gly Thr Tyr Phe 50 55 60

Glu Val Lys lie Pro Ser Asp Thr Phe Tyr Asp His Glu Asp Thr Thr

65 70 75 80 Thr Asp Lys Leu Lys Leu Thr Leu Lys Leu Arg Glu Gin Gin Leu Val

85 90 95

Gly Glu Lys Ser Trp Val Gin Phe Asn Ser Asn Ser Gin Leu Met Tyr 100 105 110

Gly Leu Pro Asp Ser Ser His Val Gly Lys His Glu Tyr Phe Met His 115 120 125

Ala Thr Asp Lys Gly Gly Leu Ser Ala Val Asp Ala Phe Glu lie His 130 135 140 Val His Arg Arg Pro Gin Gly Asp Arg Ala Pro Ala Arg Phe Lys Ala 145 150 155 160

Lys Phe Val Gly Asp Pro Ala Leu Val Leu Asn Asp lie His Lys Lys 165 170 175 lie Ala Leu Val Lys Lys Leu Ala Phe Ala Phe Gly Asp Arg Asn Cys 180 185 190

Ser Thr lie Thr Leu Gin Asn lie Thr Arg Gly Ser lie Val Val Glu 195 200 205

Trp Thr Asn Asn Thr Leu Pro Leu Glu Pro Cys Pro Lys Glu Gin lie 210 215 220

Ala Gly Leu Ser Arg Arg lie Ala Glu Asp Asp Gly Lys Pro Arg Pro 225 230 235 240

Ala Phe Ser Asn Ala Leu Glu Pro Asp Phe Lys Ala Thr Ser lie Thr 245 250 255

Val Thr Gly Ser Gly Ser Cys Arg His Leu Gin Phe lie Pro Val Val 260 265 270

Pro Pro Arg Arg Val Pro Ser Glu Ala Pro Pro Thr Glu Val Pro Asp 275 280 285

Arg Asp Pro Glu Lys Ser Ser Glu Asp Asp Val Tyr Leu His Thr Val 290 295 300 lie Pro Ala Val Val Val Ala Ala lie Leu Leu lie Ala Gly lie lie 305 310 315 320

Ala Met lie Gin Tyr Arg Lys Lys Arg Lys Gly Lys Leu Thr Leu Glu 325 330 335

Asp Gin Ala Thr Phe lie Lys Lys Gly Val Pro lie lie Phe Ala Asp

340 345 350 Glu Leu Asp Asp Ser Lys Pro Pro Pro Ser Ser Ser Met Pro Leu lie

355 360 365

Leu Gin Glu Glu Lys Ala Pro Leu Pro Pro Pro Glu Tyr Pro Asn Gin 370 375 380

Ser Val Pro Glu Thr Thr Pro Leu Asn Gin Asp Thr Met Gly Glu Tyr

385 390 395 400

Thr Pro Leu Arg Asp Glu Asp Pro Asn Ala Pro Pro Tyr Gin Pro Pro 405 410 415

Pro Pro Phe Thr Val Pro Met Glu Gly Lys Gly Ser Arg Pro Lys Asn 420 425 430 Met Thr Pro Tyr Arg Ser Pro Pro Pro Tyr Val Pro Pro

435 440 445

Claims

e Claim:

1. A substantially purified oligopeptide which is characterized by the ability to bind agrin, having a molecular weight of about 190 kD as measured by SDS polyacrylamide gel electrophoresis.

2. A substantially purified protein of Claim 1 associated with a second polypeptide of a molecular weight of about 50 kD, as measured by SDS polyacrylamide gel electrophoresis.

3. A substantially purified oligopeptide or fragment thereof of the protein of

Claim 1 which is characterized by the ability to inhibit the binding of agrin to the protein of Claim 1.

4. The protein of Claim 1, isolated from Torpedo electric organ.

5. The protein of Claim 2, isolated from Torpedo electric organ.

6. The protein of Claim 1, in glycosylated form.

7. The protein of Claim 2, in glycosylated form.

8. A method of screening for compounds which can inhibit binding of agrin to the protien of Claim 1 comprising the steps of contacting an amount of protein or fragment therof of Claim 1 with an amount of compound to be tested, followed by the administration of an amount of agrin, where one of the components of the assay is labeled so that a quantitative determination of binding can be made.

9. A method of screening compounds for binding with the protien of Claim 1 comprising the steps of contacting an amount of protein or fragment therof of Claim 1 with an amount of compound to be tested, where one of the components is labeled so that a qualitative measurement of binding can be made.

10. A monoclonal antibody, mAb-AgRl, which is produced by the cell line AgRl.

11. The cell line AgRl.

12. A monoclonal antibody that inhibits the specific binding of the monoclonal antibody mAb-AgRl with the protein or fragment thereof of Claim 1.

13. A method for inhibiting binding of agrin to agrin receptor consisting of the administration of an effective amount of the protein or fragment thereof of Claim 1.

14. A method for inhibiting the aggregation of neural tranmitter receptors on neural tissues comprising the administration of an effective amount of the protein or fragment thereof of Claim 1.

15. A method as in Claim 15 where the neurotransmitter receptor is the AChR (acetyl choline receptors).

16. A method for inhibiting the clustering activity associated with agrin binding to pre- and postsynaptic membranes consisting of the administration of an effective amount of the protein or fragment thereof of Claim 1.

17. A method for enhancing the clustering activity associated with agrin binding to pre- and postsynaptic membranes consisting of the administration of an effective amount of the protein or fragment thereof of Claim 1.

18. A method for the detection of agrin receptor consisting of the steps of contacting an amount of the monoclonal antibody mAb-AgRl with a sample to be tested, and determining the amount of mAb-AgRl bound.

19. The protein of Claim 1 produced by recombinant DNA technology.

20. The protein of Claim 1 expressed by a vector in one of the group consisting of bacterial cells, yeast cells, plant cells, fungal cells, insect cells, reptile cells, avian cells, amphibian cells, mammalian cells and hybrid cells.

21. A replicating DNA vector capable of expression of the protein of Claim 1.

22. A cell line capable of expressing the protien of Claim 1.

23. A peptide comprising the amino acid sequence LKEQQLLSESSWVKFQSTXQ. [SEQ ID NO. 1]

24. A peptide comprising the amino acid sequence

HEYFMHAAXKGGLTAVXXFE. [SEQ ID NO. 2]

25. A peptide comprising the amino acid sequence

GLAQAFGDRXSSTVTLLAIS. [SEQ ID NO. 3]

26. A peptide comprising the amino acid sequence

XVLVXXIGXTLPLEXQ. [SEQ ID NO. 4]

27. A peptide comprising the amino acid sequence

SLGSQLADADGRPTPAFT. [SEQ ID NO. 5]

28. A peptide comprising the amino acid sequence

GVPIIFADELDDSKPPPSSSVXLII. [SEQ ID NO. 6]

29. A protein of Claim 1 which contains one of the amino acid sequences selected from the group consisting of

1. LKEQQLLSESSWVKFQSTXQ [SEQ ID NO. 1]

2. HEYFMHAAXKGGLTAVXXFE [SEQ ID NO. 2]

3. GLAQAFGDRXSSTVTLLAIS [SEQ ID NO. 3]

30. A 50 kDa protein of Claim 2 which contains one of the amino acid sequences selected from the group consisting of

1. XVLVXXIGXTLPLEXQ [SEQ ID NO. 4]

2. SLGSQLADADGRPTPAFT [SEQ ID NO. 5]

3. GVPIIFADELDDSKPPPSSSVXLII [SEQ ID NO. 6]