EP1578915A2

EP1578915A2 - Methods for inhibiting proliferation of astrocytes and astrocytic tumor cells and for enhancing survival of neurons and uses thereof

Info

Publication number: EP1578915A2
Application number: EP02802862A
Authority: EP
Inventors: David E. Weinstein; Carlos Suarez; Marielba Zerlin
Original assignee: Albert Einstein College of Medicine
Current assignee: Albert Einstein College of Medicine
Priority date: 2001-11-07
Filing date: 2002-11-07
Publication date: 2005-09-28
Also published as: JP4503287B2; JP2005537779A; NZ532989A; EP1578915A4; AU2002363524B2; WO2003040333A3; WO2003040333A2; CA2466334A1

Abstract

The present invention provides methods for inhibiting proliferation of astrocytes and astrocytic tumor cells. The present invention further provides methods for treating a condition associated with a defect in astrocyte proliferation in a subject, and methods for treating a condition associated with astrocytic tumor cell proliferation in a subject. Additionally, the present invention discloses methods for enhancing survival of neurons. The present invention still further provides methods for treating neural degeneration in a subject. The present invention is also directed to pharmaceutical compositions, comprising NrS1 protein or nucleic acid and a pharmaceutically-acceptable carrier. Finally, the present invention provides a purified NrS1 protein, an agent that binds to the NrS1 protein, a kit comprising an agent that binds to the NrS1 protein, and a composition comprising the NrS1 protein and a carrier.

Description

METHODS FOR ΓNHTBITLNG PROLIFERATION OF ASTROCYTES

AND ASTROCYTIC TUMOR CELLS AND FOR ENHANCING SURVIVAL

OF NEURONS AND USES THEREOF

Background ofthe Invention

[ 0001] This year, and each year in the foreseeable future, 17,000 people in the United

States will be diagnosed with brain tumors. The majority of these tumors will be of astrocyte lineage, and most people diagnosed with these malignancies will die of their diseases. Brain tumors, or intracranial neoplasms, are found in about 2% of all routine autopsies. They are most common in early or middle adult life, but may occur at any age. Their frequency also appears to be increasing in the elderly (30).

[ 0002] Brain tumors invade and destroy normal tissue, producing such effects as impaired sensorimotor and cognitive function, increased intracranial pressure, cerebral edema, and compression of brain tissue, cranial nerves, and cerebral vessels (30). Metastases may involve the skull or any intracranial structure. The size, location, rate of growth, and histo logic grade of malignancy determine the seriousness of brain tumors. Nonmalignant tumors grow slowly, with few mitoses, no necrosis, and no vascular proliferation. Malignant tumors grow more rapidly, and invade other tissues. However, they rarely spread beyond the central nervous system (CNS), because they cause death by local growth. Drowsiness, lethargy, obtuseness, personality changes, disordered conduct, and impaired mental faculties are the initial symptoms in 25% of patients with malignant brain tumors (30).

[ 0003] While brain tumors, or intracranial neoplasms, are common, they are frequently misdiagnosed (30). Brain tumors may be classified by site (e.g., brain stem, cerebellum, cerebrum, cranial nerves, ependyma, meninges, neuroglia, pineal region, pituitary gland, and skull) or by histologic type (e.g., meningioma, primary CNS lymphoma, or astrocytoma) (30). Common primary childhood tumors are cerebellar astrocytomas and medulloblastomas, ependymomas, gliomas ofthe brain stem, and congenital tumors. In adults, primary tumors include meningiomas, schwannomas, and gliomas ofthe cerebral hemispheres (particularly the malignant glioblastoma multiforme and anaplastic astrocytoma, and the more benign astrocytoma and oligodendroglioma). Overall incidence of intracranial neoplasms is essentially equal in males and females, but cerebellar medulloblastoma and glioblastoma multiforme are more common in males (30). [ 0004] Gliomas are tumors composed of tissue representing neuroglia in any one of its stages of development (30). They account for 45% of intracranial tumors. Gliomas can encompass all ofthe primary intrinsic neoplasms ofthe brain and spinal cord, including astrocytomas, ependymomas, and neurocytomas. Astrocytomas are tumors composed of transformed astrocytes, or astrocytic tumor cells. Such tumors have been classified in order of increasing malignancy: Grade I consists of fibrillary or protoplasmic astrocytes; Grade II is an astroblastoma, consisting of cells with abundant cytoplasm and two or three nuclei; and Grades III and IV are forms of glioblastoma multiforme, a rapidly growing tumor that is usually confined to the cerebral hemispheres and composed of a mixture of astrocytes, spongioblasts, astroblasts, and other astrocytic tumor cells. Astrocytoma, a primary CNS tumor, is frequently found in the brain stem, cerebellum, and cerebrum. Anaplastic astrocytoma and glioblastoma multiforme are commonly located in the cerebrum (30).

[ 0005] Treatment of brain tumors is often multimodal, and depends on pathology and location ofthe tumors (30). For malignant gliomas, multimodal therapy, including chemotherapy, radiation therapy, and surgery, is used to try to reduce tumor mass. Regardless of approach, however, prognosis for patients suffering from these tumors is guarded: the median term of survival after chemotherapy, radiation therapy, and surgery is only about 1 year, and only 25% of these patients survive for 2 years. In view ofthe foregoing, it is imperative that new ways be developed for diagnosing, detecting, and treating malignant gliomas (30).

[ 0006] Astrocytes also have been implicated in pathologies produced by virtually all neural traumas, including CNS injury and neuronal cell death resulting from neurodegenerative disease. In the case of CNS injury, for example, resulting astrocytosis is thought to be a major contributor to the formation of a glial scar, which is believed to present a major barrier to productive neural regeneration (6). Therefore, a primary goal in the design of therapeutics for both CNS trauma and neurodegenerative diseases is the elucidation of mechanisms for limiting glial scar formation. [ 0007] Head injuries cause more deaths and disability than any other neurologic condition before age 50, and occur in more than 70% of accidents - the leading cause of death in men and boys less than 35 years of age. Mortality from severe injury approaches 50%, and is only modestly reduced by treatment. Damage may result from skull penetration or from rapid brain acceleration or deceleration, resulting in injury to surrounding tissue. Currently, there is no treatment for astrocytosis resulting from head trauma. [ 0008] Alzheimer's disease is a neurodegenerative disease characterized by a progressive, inexorable loss of cognitive function (30). The pathogenesis of Alzheimer's disease is associated with an excessive number of neuritic, or senile, plaques (composed of neurites, astrocytes, and glial cells around an amyloid core) in the cerebral cortex, and neurofibriUary tangles (composed of paired helical filaments). Approximately 4 million Americans suffer from Alzheimer's disease, at an annual cost of about $90 billion. The disease is about twice as common in women as in men, and accounts for more than 65% ofthe dementias in the elderly. While senile plaques and neurofibriUary tangles occur with normal aging, they are much more prevalent in persons with Alzheimer's disease. To date, a cure for Alzheimer's disease is not available, and cognitive decline is inevitable.

[ 0009] At present, there are no specific treatments for astrocytosis. In addition, while there are standard chemotherapeutic, radiotherapeutic, and surgical treatments for astrocytoma, these therapies are fraught with severe limitations, and are often palliative rather than curative. Accordingly, there is a great need to develop methods of treating astrocytomas, astrocytosis, and other conditions associated with a proliferation of astrocytes or astrocytic tumor cells. An understanding ofthe basic biology of neuron-glial interaction may provide insight into the elucidation of such treatment options.

Summary ofthe Invention

[ 0010] The present invention is based upon the discovery that CD81 on the surface of astrocytes modulates proliferation of astrocytes in neural tissue, and is not expressed in astrocytic tumor cells, and the discovery that NrSl on the surface of neurons is involved in bi-directional signaling to astrocytes and neurons following contact with CD81. On the basis of these findings, the present invention provides methods for enhancing survival of neurons. The methods comprise contacting the neurons with CD81 protein or a CD81 derivative, in an amount effective to enhance survival ofthe neurons.

[ 0011] The present invention is also directed to methods for treating neural degeneration in a subject in need of treatment. The methods comprise contacting neurons in the subject with

CD81 protein or CD81 derivative in an amount effective to enhance survival of neurons, thereby treating the neural degeneration.

[ 0012] The invention is additionally directed to methods for treating neural degeneration in a subject in need of treatment. The methods comprise activating NrSl in the subject. [ 0013] In other embodiments, the present invention is directed to methods for inhibiting proliferation of astrocytes. The methods comprise contacting the astrocytes with an amount of NrSl protein or NrSl derivative effective to inhibit proliferation ofthe astrocytes. [ 0014] Additionally, the invention is directed to methods for treating a condition associated with a defect in astrocyte proliferation in a subject in need of treatment. The methods comprise contacting astrocytes in the subject with NrSl protein or NrSl derivative in an amount effective to inhibit proliferation ofthe astrocytes, thereby treating the condition associated with a defect in astrocyte proliferation.

[ 0015] The present invention is additionally directed to methods for treating a condition associated with a defect in astrocyte proliferation in a subject in need of treatment. The methods comprise activating CD81 in the subject.

[ 0016] In additional embodiments, the invention is also directed to pharmaceutical compositions. The compositions comprise NrSl and a pharmaceutically-acceptable carrier. Other compositions ofthe invention comprise nucleic acids encoding NrSl and a pharmaceutically- acceptable carrier.

[ 0017] In further embodiments, the invention is directed to methods of determining whether a subject has an astrocytoma. The methods comprise assaying for CD81 expression in a diagnostic sample of cells of astrocytic lineage ofthe subject, where no detection of expression of CD81 in the cells of astrocytic lineage is diagnostic of an astrocytoma.

[ 0018] Finally, the present invention provides purified NrSl protein, agents that bind to the NrSl protein, kits comprising an agent that binds to the NrSl protein, and compositions comprising the NrSl protein and a carrier.

[ 0019] Additional objects ofthe present invention will be apparent in view ofthe description which follows.

Brief Description ofthe Drawings

[ 0020] FIGS. 1 A is a western blot, and FIGS. IB and 1C are fluorescence micrographs, that demonstrate that CD81 is expressed on the surface ofthe astrocyte. After the expression of CD81 message was identified by differential screen, protein expression was determined by Western blot analysis. (A) While astrocytes express ample CD81, the C6 glioma cell line is CD81 -negative. To localize the protein expression in the astrocyte, astrocytes were cultured either alone (B) or in the presence of neurons (C) for 48 h, then fixed and stained for the expression of CD81 on the cell surface, using the monoclonal antibody 2F7. The arrows in (B) point out the expression of CD81 along the astrocyte processes - a domain ofthe cell critical for neuronal interaction (10).

[ 0021] FIGS. 2A-2H are graphs and micrographs establishing that CD81 is a critical mediator of neuron-astrocyte interactions, by showing that anti-CD81 antibody Eatl effectively interferes with normal neuron-mediated astrocyte arrest. (A) In the presence of increasing concentrations of Eatl monoclonal antibody (mAB), there was a loss of neuron-mediated astrocyte proliferative arrest (closed bars); the antibody had no effect on astrocyte proliferation in the absence of neurons (open bars). (B) In contrast, 2F7 augmented neuronally-induced astrocyte proliferative arrest, such that, in the presence of this mAb and neurons, astrocyte proliferation was further reduced over the level seen with neurons alone. (C) Immunofluorescence studies of astrocytes stained with anti-GFAP antiserum in neuron-astrocyte co-cultures, in the presence of Eatl, showed a dependence on the Eatl epitope for normal responsiveness to neuronally-induced, astrocyte process formation. In contrast, blocking the Eat2 epitope had no effect on astrocyte responsiveness to neurons, as evidenced by the complex GFAP processes seen in these cultures (D) which had the same appearance as control co-cultures (E). While Abs Eatl and 2F7 had profound effects on astrocytic responses to neurons, they had no observable effects on neuronal survival or axonogenesis. Neuron-astrocyte co-cultures were also stained with the mAb TuJl, which recognizes a neuron-specific βlll tubulin isoform (3). The extent and quality of neurites were comparable in the presence of Eatl (F) and 2F7 (G), and in control co-cultures (H). [ 0022] FIGS. 3A-3C are fluorescence micrographs demonstrating that GST-CD81 binds to neurons, not astrocytes. Highly-enriched cultures of either neurons or astrocytes from P4 rat cerebellum were established, as previously described (28). The cells, which were plated at equivalent densities, were cooled on ice to prevent internalization, then incubated with 10 μg/ml of bacterially-expressed GST-CD81 for 1 h. The cells were fixed, and stained with a goat anti- GST antibody, followed by an Alexa red conjugated rabbit anti-goat secondary. FIG. 3A shows the absence of background staining (no primary antibody control). FIG. 3B shows the binding of the CD81 fusion protein to the surface ofthe neuron. In contrast, only the few contaminating neurons in the astro cyte-enriched fraction bound the fusion protein (C). [ 0023] FIG. 4 is a graph that illustrates that soluble CD81 competes with astrocyte- expressed CD81, and blocks neuron-induced astrocyte quiescence. Increasing concentrations of soluble GST-CD81 were added to co-cultures of neurons and astrocytes. The GST-CD81 competed for neurons with the expressed CD81, thereby blocking neuronal CD81 -receptor binding at the astrocyte cell surface. As a result of this competition, astrocytes remained in the cell cycle. 40-50% of neuron-induced inhibition of astrocyte proliferation was achieved with as little as 1 μg/ml of GST-CD81. Maximal inhibition was obtained with 3 μg/ml of soluble protein. The soluble proteins had no observable effects on astrocyte proliferation in the absence of neurons. The specificity of he effect of GST-CD81 was verified by the addition of another, irrelevant GST fusion protein, GST-SCIP, which had no effect on neuron-induced astrocyte quiescence at any concentration tested. Statistical analysis was carried out using the two-tailed students' t-test.

[ 0024] FIG. 5 is a pair of graphs that shows that CD81 is required for neuron-induced astrocyte growth regulation. Mixed cultures of wild-type, CD81+/-, or CD81-/- cerebellar astrocytes and granule cell neurons were established and evaluated for the role of endogenous CD81 in neuron-induced astrocyte responses. Astrocyte proliferation was determined by double labeling for GFAP and BrdU 48 h after the cultures were established, as described below. Astrocyte proliferation in wild-type co-cultures was determined, and arbitrarily set at 1. CD81 haplo-insufficient astrocytes showed a slight loss of neuron responsiveness (20%). However, astrocytes null at the CD81 locus lost all responsiveness to neurons under these conditions, doubling in number within 48 h after explanation. Assays were done in triplicate for each animal tested, and the overall experiment was repeated three times.

[ 0025] FIG. 6 is a northern blot illustrating that CD81 RNA is absent from a range of astrocytic tumor cell lines. A hallmark of cell transformation is a loss of proliferative arrest in response to naturally-occurring cues. To determine if astrocytic tumor cell lines expressed altered levels of CD81, the inventor extracted RNA from a variety of cell lines, and performed northern blot analysis. Not unexpectedly, CD81 message was found in wild-type astrocytes. The CD81 levels increased by approximately two-fold when the cells were co-cultured for 48 h with neuronal membranes, suggesting a positive feed-back mechanism affecting CD81 expression. In stark contrast, none ofthe tumor cell lines tested had any detectable CD81 message, even when the blot was over-exposed (not shown). These astrocytic tumors tested were rat: C6 and 9L; human: A172 and U251MG; and mouse: LN308 and LN18. An 18S probe was used as a loading control for RNA.

[ 0026] FIG. 7 depicts the nucleotide sequence of human CD81.

[ 0027] FIG. 8 depicts the amino acid sequence of human CD81. [ 0028] FIG. 9 is a graph that demonstrates the saturable nature of neuronal protein binding to astrocytes. Cerebellar granule cell neuron proteins (closed diamonds) or NGF- differentiated PC 12 cell membrane proteins (open diamonds) were isolated, as described below, and allowed to bind to mono layers of astrocytes. The CNS neuron proteins saturated binding at ~10 μg/10⁵ astrocytes; in contrast, the PC12 cell membranes did not bind to the astrocytes at all. [ 0029] FIGS. 10A is a silver stained gel and 10B shows micrographs that illustrate column chromatographic purification of proteins active in neuron-induced astrocyte growth arrest. (A) The inventor established a purification scheme in which lipophilic neuronal proteins were isolated (lane 1), then passed over an FPLC anion exchange column. The peak between 240 mM NaCl and 260 mM NaCl was collected (lane 2), dialyzed, and passed over a column of total astrocyte membrane proteins. This column then was eluted with 10 mM EDTA (lane 3), followed by 500 mM NaCl (lane 4). All ofthe antiproliferative activity was contained in lane 4 (see Table 1). The proteins were run on a 6% SDS-PAGE gel, and imaged by silver staining. The arrow points to the major enriched band at ~65 kD. (B) The protein from lane 4 (panels a and b), or BSA (panels c and d), was coupled to a nitrocellulose-coated tissue culture surface, and non-specific binding was quenched with BSA.. Astrocytes were added to a 48-h culture, and the cells then were stained with rabbit anti-GFAP (panels a and c) and mAb anti-statin (panels b and d). These representative fields demonstrate that statin translocation to the nucleus - a marker of cells in G₀ - occurs specifically when the cells are exposed to proteins that were NaCl-eluted from the final column.

[ 0030] FIG. 11 is a graph that illustrates that GM1109 competes for astrocyte-expressed

CD81, and blocks neuron-induced astrocyte quiescence. Upon addition of increasing concentrations of GM1109 to co-cultures of neurons and astrocytes, GM1109 competed for astrocyte-expressed CD81, thereby blocking the binding ofthe neuronal CD81 -receptor to the astrocyte cell surface. As a result ofthe competition, astrocytes remained in the cell cycle. 40- 50%) of neuron-induced inhibition of astrocyte proliferation was achieved with as little as 1 μg/ml of GM1109; maximal inhibition was obtained with 3 μg/ml of soluble protein. The soluble proteins had no observed effect on astrocyte proliferation in the absence of neurons. Compare the open bars corresponding to control conditions with those corresponding to conditions in which protein was added. The specificity ofthe GM1109 effects was verified by the addition of an irrelevant drug, GM1110, which had no effect on neuron-induced astrocyte quiescence at any concentration tested. Statistical analysis was carried out using the two-tailed students' t-test. [ 0031] FIG. 12 shows micrographs demonstrating that GMl 109 binds to neurons, but not to astrocytes. Highly enriched cultures of either neurons or astrocytes from P4 rat cerebellum were established, as described below. The cells, which were plated at equivalent densities, were cooled on ice, to prevent internalization, and incubated with 10 μg/ml of bacterially-expressed GMl 109 for 1 h. The cells were fixed, then stained with a goat anti-GST antibody, followed by an Alexa red conjugated rabbit anti-goat secondary antibody. Panel a shows the absence of background staining (no primary antibody control). Panel b shows the binding of GMl 109 to the surface ofthe neuron. In contrast, only a few contaminating neurons in the astrocyte-enriched fraction bound the fusion protein (panel c).

[ 0032] FIGS. 13 A and 13B are blots demonstrating that GMl 109 specifically binds to neuronal protein extract. Far- Western analysis using soluble GMl 109 showed that this protein specifically binds to a neuronal, detergent-soluble protein of approximately 70 kD. Granule cell neurons (GCN) from either rat or mouse, as well as rat Schwann cells and 3T3 cells, were solublized in Triton X-100. Ten micro grams (10 μg) of each protein were separated on a 12%> SDS-PAGE gel. In addition, 1 μg ofthe probe also was run on the gel, and was then transferred to nitrocellulose. The blot on the left (A) was probed with GMl 109, and the blot on the right (B) was probed with GST alone. Both were probed at 4°C, and thoroughly washed. The GST or GMl 109 was visualized with an anti-GST antibody. GMl 109 reacted solely with a single band in the neuronal fractions, thereby demonstrating the specificity of GMl 109 interaction. The GMl 109-positive control was recognized by the anti-GST secondary antibody in both panels, and served as an internal control ofthe specificity of GMl 109 and the secondary antibody. [ 0033] FIG. 14 is a silver stained gel that shows that GMl 109 pulls down NrSl from neurons. The inventor purified rat cerebellar granule cell neurons, as previously described. After contaminating astrocytes were preferentially adhered out, the cells were mixed with 10 μg of soluble GMl 109. The cells were lysed, and the GMl 109 and associated proteins were pulled down with Sepharose-glutathione beads (lane 1). 3T3 cells were treated identically after the purification step. The pull-down from the 3T3 cells is shown in lane 2. The proteins were separated on a 10% SDS-PAGE gel, which was then silver stained. The heavy bands at -30 kD represent the starting GMl 109, which was purified from bacterial lysate by Sepharose-glutathione chromatography. The single band at 70 kD represents the lone protein species expressed by the neuron that binds to astrocyte-expressed GM9 in trans. [ 0034] FIG. 15 is a western blot that illustrates that GMl 109 induces phosphorylation of

NrSl. Human granule cell neurons were purified from a 22-week old male fetal cerebellum, as described below. Following isolation, the cells were cultured in serum-free medium for 1 h and harvested. GMl 109 then was added for the indicated times and the cells were lysed. The GMl 109 and associated NrSl proteins were pulled down with glutathione-conjugated Sepharose (Pharmacia), separated on a 10%) SDS-PAGE, transblotted to a nitrocellulose membrane, and probed with a polyclonal anti-phosphotyrosine antiserum. The data show that NrSl is phosphorylated, in response to CD81 signaling, in a time-dependent manner - a result consistent with active signaling from the astrocyte to the neuron, via a CD81-NrSl pathway. αY* = phosphotyrosine

[ 0035] FIGS. 16A-16C are micrographs that demonstrate that full astrocyte differentiation depends upon NrSl and other neuronal signals. Astrocytes were grown in the presence or absence of neurons, with or without dbcAMP. (A) Astrocytes in the presence of complete medium supplemented with dbcAMP grew to confluence, with no sign of astro cytic-pro cess outgrowth. (B) When the same cells were cultured with NrSl, they growth-arrested, but failed to form processes. (C) However, when the astrocytes were provided with both NrSl and dbcAMP, they both growth-arrested, and formed complex processes, mimicking astrocyte responses to viable neurons.

[ 0036] FIGS. 17A-17C are micrographs, and FIG. 17D is a western blot showing that the astrocyte mimetic, GMl 109, induces the rapid cleavage of neuronal, transmembrane neuregulin (Nrg), and the migration of ICD_Nrg to the neuronal nucleus. Cerebellar granule cells were acutely dissociated from P3 Long-Evans rats, and purified, as previously described (27). The cells were washed free of serum, then plated in defined medium that was supplemented with 10%> BSA. After the cells were allowed to adhere to the coverslips, either control GST protein (A) or GMl 109 (B) was added to the medium, and the cells were allowed to incubate for 5 min. The cells were fixed, permeabilized, and stained with an antibody that recognizes the intracellular domain of neuregulin (ICD_Nrg). As the figure illustrates, the binding of GMl 109 to the neuron induces nuclear translocation ofthe ICD_Nrg. As a result of GMl 109, the Nrg extracellular domain (ECD) is rapidly shed to the supernatant. (D) Treatment of neurons with GMl 109 caused a 2- fold increase in the amount of soluble Nrg in the supernatant, within 5 min, as compared to neurons treated with an irrelevant peptide. [ 0037] FIG. 18 is a micrograph showing that GMl 109 induces neuronal survival and neuritogenesis. Neonatal rodent cerebellar neurons were purified to >99%> homogeneity, and cultured at low density, in the presence of either a control peptide or GMl 109, for 96 h. The cells were fixed and stained with anti-GFAP and anti-neuron-specific tubulin antibodies. Treatment with the CD 81 /astrocyte mimetic, GMl 109, induced neuronal survival and robust axonal outgrowth (panels B and C, at the right). In contrast, most cells grown in the absence of GMl 109 failed to survive; those that did survive extended meager processes (panel A, at the left). [ 0038] FIGS. 19A and 19B are illustrations depicting a model ofthe molecular basis of neuron-astrocyte interactions. (A) Astrocytes express CD81, which is required for neuron- induced growth arrest. Neurons signal to astrocytes by binding to CD81 via NrSl, resulting in the rapid phosphorylation of NrS 1. Concomitantly with these events, neuronally-expressed, transmembrane Nrg is cleaved, resulting in the secretion of Nrg. Secreted Nrg binds the astrocyte erbB2/4 heterodimers, inducing elevation of intracellular cAMP in the astrocyte. Together with CD81 signaling, these events lead to astrocyte growth arrest and maturation. In addition, the cleavage of neuronal Nrg results in the translocation of ICD_Nrg to the nucleus, as well as neuronal survival and differentiation. (B) GMl 109 acts as an astrocyte mimetic on the neuron. In the absence of added growth factors or astrocytes, treatment of neurons with GMl 109 results in cleavage of Nrg, translocation of ICD_Nrg to the neuronal nucleus, and neuronal survival. [ 0039] FIGS. 20 A is a graph and 20B shows micrographs demonstrating that GM1416 rescues CNS neurons from cell death. (A) Neonatal rodent cerebellar neurons were purified to >99% homogeneity, and cultured at low density in the presence of 10 μg/ml GMl 109 as a positive control, or 1 μg/ml, 3 μg/ml, and 10 μg/ml ofthe mimetics GM1414, GM1415, and GM1416, respectively, or no treatment. The peptides were added fresh daily for 5 days, and then viability was assayed using a commercial live/dead assay from Molecular Probes. As expected, most neurons died in the absence of treatment. However, consistent with data shown in FIG. 19, GMl 109 maintained neuronal viability. At the lowest concentration of GMl 414, neuronal survival was almost at GMl 109 levels; however, at increasing concentrations, neuronal survival was lost. GM1415 had no neuronal-survival-inducing activity. In contrast, GM1416 potently induced neuronal survival, surpassing even that observed for GMl 109. (B) Neurons treated with GM1416 survived, and produced extensive neurites. As a positive control, purified granule cell neurons were treated with 10 μg/ml of GMl 109, inducing survival and neuritogenesis (panel a). Sister cultures were treated with GM1416, a small peptide derivative of GMl 109, which also induced survival and neurite production (panel b). In contrast, neurons cultured in the absence of these molecules promptly died, as determined by inclusion of he EtBr stain (panel c). [ 0040] FIG. 21 shows micrographs establishing that removal of CD81 by homologous recombination results in global astrocytosis. The vast majority of CD81-/- animals die perinatally. However, a small number of animals live for ~2 weeks. Examination ofthe brains of these outlying animals shows a global astrocytosis. Sections were cut through the cerebellum of either wild-type (+/+), or CD81-/- (-/-) animals and stained with an antibody to GFAP or for NrSl expression. There are a small number of GFAP-positive astrocytes in the wild-type cerebellum, and those cells that are present are tightly apposed to the Purkinge (PC) and granule cell neuron soma. In the molecular layer (ML), astrocyte are running perpendicular to the granule cell processes in the +/+ cerebellum. In contrast, the -/- cerebellum shows a large increase in astrocyte cell number, Bergmann-type glia (arrows) remaining in the ML, loss of PCs (asterisks), and a loss ofthe tight contacts between the neurons and astrocytes (arrow heads). [ 0041] FIG. 22 shows micrographs of CD81-/- animals suffering neuronal atrophy and death, as well as heterotopia. Cerebelli of two week old CD81 +/+ and -/- were stained for NrSl expression. In the absence of CD81, Purkinje cells (adjacent to the dotted lines in both panels) undergo severe atrophy and death. In addition, there is heterotopia, in which some granule cell neurons failed to complete their migration along Bergmann glia into the granule cell layer (GCL). This tissue is outlined in dark dashes.

[ 0042] FIG. 23 is a micrograph showing that CD81 expression is lost in astrocytomas in situ. We have stained numerous biopsy and resection specimens from patients with confirmed diagnoses of astrocytoma, all of which have lost CD81 expression in the neoplastic tissue. This representative needle biopsy is from a 34 year old woman who died of her disease. The tissue was stained with an antibody, 2F7, which stained the astrocytes in the margins around the mass, but failed to stain the astrocytic cells that comprise the bulk ofthe tumor. The box in the lower left is an enlargement of CDS 1 positive cells at the border ofthe lesion.

[ 0043] FIG. 24 shows micrographs providing evidence that CD81 is required for initial astro cyte/neuron interactions. In the absence of neurons, acutely dissociated astrocytes undergo rapid extensions and retractions of membrane blebs (a). This behavior lasts from hours to days. When CNS neurons are added to these cells, the astrocytes rapidly stop blebbing upon neuronal contact, when they exit the cell-cycle and extend organized processes (b). Astrocytoma cells (C6 in this case) bleb in a manner similar to astrocytes (c), however, these cells ignore the addition of neurons (d). When the astrocytoma cells are transfected with CD81, they regain responsiveness to neurons (e). (A=astrocyte; Am=astrocytoma; N=neuron).

[ 0044] FIG. 25 shows micrographs providing evidence that re-expression of CD81 rescues astrocytoma neuron responsiveness. C6 astrocytoma cells were stably transfected with a

CD81 cDNA and cloned by limiting dilution. The transfectants (a) or the parent cell line (b) showed no difference in growth or morphology in the absence of neurons. However, when neurons were added to the CD 81 expressing cells (c), they exited the cell-cycle, extended processes and remained quiescent as long as the neurons were present. In contrast, the parental

C6 cells were insensitive to the neuronal anti-proliferative signal, and continued to expand (d).

The arrow heads point to the remaining neurons in panel d.

[ 0045] FIG. 26 shows two-dimensional gel electrophoresis separation of NrSl pulled down by GMl 109. NrSl was pulled down, as described in the body ofthe application. This material was separated in the first dimension using iso-electric focusing, and in the second dimension based on molecular mass. The four predominant spots shown were subjected to mass spectroscopy analysis and Edman degradation analysis. These analyses yielded data consistent with our previous observations that NrSl is a tyrosine kinase. Moreover they suggested that

NrSl has been previously identified as a orphan receptor tyrosine kinase termed EHK-1. EHK-1 is a member ofthe ephrin family of tyrosine kinases.

[ 0046] FIG. 27 shows the published nucleotide sequence of EHK-1. GenBank Accession number XM_046083.

[ 0047] FIG. 28 shows the published amino acid sequence of EHK-1. GenBank Accession number XP_046083.2.

[ 0048] FIG. 29 shows the sequence of a novel exon 5' to the published sequence of

EHK-1.

[ 0049] FIG. 30 shows micrographs establishing that NrSl is expressed by the majority, if not all, cortical, hippocampal, thalamic, motor, and cerebellar neurons. Our biochemical data demonstrated that the CD81 mimetic, GMl 109, bound to and was used to identify NrSl. Based on the ability of GMl 109 to bind to this ligand, we used biotinylated GMl 109 to localize NrSl expression. Binding of this probe to mouse brain shows NrSl is expressed by virtually all cortical, hippocampal and thalamic neurons. Biotinylated GST was used as a control, which failed to bind to the brain sections. [ 0050] FIG. 31 shows micrographs establishing that NrSl is expressed by most retinal neurons. Staining of rat retinae with the same probe used in FIG. 30 revealed NrSl expression by ganglion cell neurons, photo receptors and most cells in the inner nuclear layer (A). Retinae were unstained by the control, biotinylated GST probe (B).

[ 0051] FIG. 32 shows micrographs establishing that NrSl is expressed by enteric neurons.

Staining of mouse intestine with the same probe used in FIG. 30 revealed NrSl expression by the myenteic plexus (mp) and the submucosal plexus (sp), the plexi ofthe gut. In addition, the sympathtic nerves (sn) that innervate the intestinal wall (iw) are also strongly NrSl positive. GMl 109 was visualized with either a colorometric substrate (b and d), or a fluorescinated streptavidan substrate (a and c).

[ 0052] FIG. 33 is an illustration of a model of GMl 109/NrS 1 interactions. GMl 109 is a

CD81 mimetic, and binds to NrSl, as shown herein. Binding of GMl 109 to its ligand induces the rapid phosphorylation of NrSl.

[ 0053] FIG. 34 is a western blot showing that binding of CD81 and CD81 mimetics to

NrSl induces shedding ofthe extracellular domain of Nrg to the medium. One minute following binding of GMl 109 to neurons, the medium was collected. Fifty μl ofthe medium was electrophoresed on a 12% SDS-PAGE gel, then transferred to nitrocellulose. The membrane was probed with an anti-Nrg antibody, specific for the mature, secreted form ofthe growth factor. As shown, there is a rapid accumulation ofthe protein in the medium, while there is a concomitant translocation ofthe intracellular domain ofthe protein to the neuronal nucleus (see above). [ 0054] FIG. 35 shows western blots establishing that GMl 109-induced, neuron-secreted

Nrg induces a rapid phosphorylation of its receptor, ErbB2, on the astrocyte. The medium used in FIG. 34 was added to astrocytes that had been cultured for 48 hours in defined medium. The cells were lysed at the indicated times, separated by 8% SDS-PAGE, and transferred to nitrocellulose. The membrane was probed with a mAb specific for the phosphorylated form of ErbB2, part ofthe Nrg receptor heterodimer. The blot shows full phosphorylation ofthe receptor within 5 minutes ofthe addition ofthe medium.

[ 0055] FIG. 36 shows micrographs demonstrating that GM1416 rescues retinal ganglion cells from death in a pressure induced stroke model. Ten micrograms of GM1416RI, a retro- inverso stabilized form of GM1416, was infused into the posterior chamber ofthe eye at the time of stroke. Animals were sacrificed at the indicated times, and the retinae processed for NrSl expression. The ganglion cell soma are preserved, as are their axons in the inner plexiform layer

(IPL).

[ 0056] FIG. 37 is a graph showing preservation of axon tracks is a surrogate marker for neuronal viability. The thickness ofthe IPL was measured in control and GM1416RI treated retinae. GM1416RI preserves the full thickness ofthe IPL at 72 hours after stroke.

[ 0057] FIG. 38 provides sequences of active fragments of CD81. A - GMl 109; B -

GM1414; C - GM1415; D - GM1416; E - active 15-mer of GM1416.

Detailed Description ofthe Invention

[ 0058] The present invention provides methods for inhibiting proliferation of astrocytes, by contacting astrocytes with an amount of CD 81 effective to inhibit proliferation of astrocytes. Unless otherwise indicated, "CD81" includes any mammalian a CD81 protein (p27), a CD81 analogue, and a CD81 fragment. As used herein, one form of CD81 protein has the amino acid sequence set forth in FIG. 8. The skilled artisan would expect that any mammalian CD81 has at least 80%) amino acid homology to any other mammalian CD81. It would also be expected that any mammalian CD81 would be functional in the methods ofthe invention in any mammal. However, it is preferred that the CD81 utilized in those methods be from the same species, to avoid possible immune rejection problems.

[ 0059] A "CD81 analogue", as defined herein, is a functional variant of a CD81 protein, having CD81 -protein biological activity, that has at least 60%, preferably at least 70%, more preferably at least 80%, even more preferably at least 90%>, still more preferably at least 95%>, and most preferably at least 97%> amino acid sequence homology with the CD81 protein, as well as a fragment ofthe CD81 protein having CD81-protein biological activity, such as GMl 109, GM1416, and the active 15-mer from GM1416 (FIG. 38). The skilled artisan could envision many variants or fragments of CD81 that would be expected to have CD81 -protein biological activity. As used herein, the term "CD81-protein biological activity" refers to protein activity which modulates and inhibits proliferation of astrocytes and astrocytoma cells, as disclosed herein. Additionally, as used herein, a "CD81 derivative" is a chemical substance derive from CD81, either directly or by modification, truncation, or partial substitution. For example, the CD81 derivative for use in the present invention may be the extracellular domain (ECD) of CD81. In addition, the CD81 derivative ofthe present invention may be retro-inverso versions of CD81 or the variants or fragments, in which the primary sequence ofthe peptides is in an inverse orientation and comprise D-amino acids. These modifications are known to stabilize the protein. [ 0060] CD81 and its analogues and derivatives may be produced synthetically or recombinantly, or may be isolated from native cells; however, they are preferably produced synthetically, using conventional techniques and cDNA encoding CD81 (FIG. 7). In one embodiment ofthe present invention, the astrocytes are undifferentiated, i.e., they are not in cell- cycle arrest, and they have not formed complex processes.

[ 0061] The methods ofthe present invention may be used to inhibit proliferation of astrocytes in vitro, or in vivo in a subject. As used herein, the term "inhibit proliferation of astrocytes" means inhibit cell division and growth of astrocytes, and includes limiting the proliferative rate of astrocytes, as disclosed herein. Inhibition ofthe growth and proliferation of astrocytes may be detected by known procedures, including any ofthe methods, molecular procedures, and assays disclosed herein.

[ 0062] In accordance with the methods ofthe present invention, CD81 may be contacted with astrocytes in vitro, or in vivo in a subject, by introducing the CD81 protein into the membranes of astrocytes, or by introducing into the astrocytes a nucleic acid encoding CD81 in a manner permitting expression of CD81 protein. The subject may be any animal, but is preferably a mammal (e.g., humans, domestic animals, and commercial animals). More preferably, the subject is a human. The astrocytes may be contained in neural tissue and other tissue ofthe nervous system ofthe subject, either alone or with other types of neural cells, including, for example,^, neurons and oligodendroglia. Astrocytes may be detected in tissue ofthe subject by standard detection methods readily determined from the known art, examples of which include, without limitation, immuno logical techniques (e.g., immunohisto chemical staining), fluorescence imaging techniques, and microscopic techniques.

[ 0063] CD81 protein may be introduced into the membranes of astrocytes, either in vitro or in vivo in a subject, by known techniques used for the introduction of proteins into cell membranes (e.g., by means of micro-encapsulated preparations, such as liposomes). The amount of CD81 protein to be used is an amount effective to inhibit proliferation of astrocytes, as defined above, and may be readily determined by the skilled artisan.

[ 0064] For introduction of CD81 protein by way of liposome delivery, liposomal vesicles may be prepared by various methods known in the art, and liposome compositions may be prepared using any one of a variety of conventional techniques for liposome preparation known to those skilled in the art. Examples of such methods and techniques include, without limitation, chelate dialysis, extrusion (with or without freeze-thaw), French press, homogenization, micro emulsification, reverse phase evaporation, simple freeze-thaw, solvent dialysis, solvent infusion, solvent vaporization, sonication, and spontaneous formation. Preparation ofthe liposomes may be carried out in a solution, such as an aqueous saline solution, aqueous phosphate buffer solution, or sterile water. Liposome compositions also may be prepared by various processes involving shaking or vortexing. CD81 protein may be incorporated into the layers of a liposome such that its intracellular domain extends outside the liposome, and its extracellular domain extends into the interior ofthe liposome. The liposome containing CD81 then may be fused with an astrocyte, in accordance with known methods of fusion of liposomes to cell membranes, such that the CD81 protein is delivered into the membrane ofthe astrocyte with its intracellular domain extending into the interior ofthe astrocyte, and its extracellular domain extending outside the membrane ofthe astrocyte.

[ 0065] In the methods ofthe present invention, CD81 also may be contacted with astrocytes, either hi vitro or in vivo in a subject, by introducing into a sufficient number of astrocytes ofthe subject a nucleic acid encoding CD81, in a manner permitting expression of CD81. The nucleic acid may be introduced using conventional procedures known in the art, including, without limitation, electroporation, DEAE Dextran transfection, calcium phosphate transfection, monocationic liposome fusion, polycationic liposome fusion, protoplast fusion, creation of an in vivo electrical field, DNA-coated microprojectile bombardment, injection with recombinant replication-defective viruses, homologous recombination, in vivo gene therapy, ex vivo gene therapy, viral vectors, and naked DNA transfer, or any combination thereof. Recombinant viral vectors suitable for gene therapy include, but are not limited to, vectors derived from the genomes of viruses such as retrovirus, HSV, adenovirus, adeno -associated virus, Semiliki Forest virus, cytomegalo virus, and vaccinia virus. The amount of nucleic acid encoding CD81 to be used is an amount that will express CD81 protein in an amount effective to inhibit proliferation of astrocytes, as defined above. These amounts may be readily determined by the skilled artisan.

[ 0066] It is also within the confines ofthe present invention that a nucleic acid encoding

CD81 may be introduced into suitable cells in vitro, using conventional procedures, to achieve expression in the cells of CD81 protein. Cells expressing CD81 protein then may be introduced into a subject to inhibit proliferation of astrocytes in vivo. In such ex vivo gene therapy approaches, the cells are preferably removed from the subject, subjected to DNA techniques to incorporate nucleic acid encoding CD81, and then reintroduced into the subject. [ 0067] The ability of CD81 to modulate astrocyte proliferation renders CD81 particularly useful for treating conditions associated with a defect in astrocyte proliferation. As used herein, "a defect in astrocyte proliferation" includes pathologic proliferation of astrocytes in a particular tissue, as compared with normal proliferation in the same type of tissue. It is believed that, by modulating astrocyte proliferation, CD81 will be useful for the treatment of conditions associated with defects in astrocyte proliferation. It is further believed that CD81 would be effective either alone or in combination with therapeutic agents, such as chemotherapeutic agents or antiviral agents, which are typically used in the treatment of these conditions. [ 0068] Accordingly, the present invention provides methods for treating a condition associated with a defect in astrocyte proliferation in a subject in need of treatment, comprising contacting astrocytes in the subject with an amount of CD81 effective to inhibit proliferation of astrocytes, thereby treating the condition. As described above, the subject may be any animal, but is preferably a mammal (e.g., humans, domestic animals, and commercial animals). More preferably, the subject is a human.

[ 0069] Examples of conditions associated with a defect in astrocyte proliferation include, without limitation, astrocytosis, glial scars, hyperplasia, neoplasia, and neuritic plaques (particularly those commonly found in Alzheimer's disease patients). As used herein, "astrocytosis" refers to the proliferation of astrocytes owing to a destruction of nearby neurons. As further used herein, "hyperplasia" refers to the abnormal multiplication or increase in the number of normal astrocytes, in normal arrangement, within a tissue. In one embodiment ofthe present invention, the condition associated with a defect in astrocyte proliferation is astrocytosis. In another embodiment ofthe present invention, the condition associated with a defect in astrocyte proliferation is a neuritic plaque.

[ 0070] Astrocytosis, glial scars, hyperplasia, neoplasia, neuritic plaques, and other conditions associated with a defect in astrocyte proliferation may be caused by, or associated with, a variety of factors, including, without limitation, neuronal cell death and neural degeneration resulting from neurodegenerative diseases, CNS traumas, and the acquired secondary effects of non-neural dysfunction. Examples of neurodegenerative diseases include, without limitation, Alzheimer's disease, amyotrophic lateral sclerosis (Lou Gehrig's Disease), Binswanger's disease, Huntington's chorea, multiple sclerosis, myasthenia gravis, Parkinson's disease, and Pick's disease. Examples of CNS traumas include, without limitation, blunt trauma, hypoxia, and invasive trauma. Examples of acquired secondary effects of non-neural dysfunction include, without limitation, cerebral palsy, congenital hydro cephalus, muscular dystrophy, stroke, and vascular dementia. [ 0071] In the treatment of a condition associated with a defect in astrocyte proliferation,

CD81 may be contacted with astrocytes by introducing the CD81 protein into the membranes of astrocytes, in accordance with known methods (e.g., liposome delivery), as described above. The amount of CD81 protein to be used is an amount effective to inhibit proliferation of astrocytes, as defined above, and may be readily determined by the skilled artisan. [ 0072] Alternatively, in accordance with known methods, including those described above, CD81 may be contacted with astrocytes to treat a condition associated with a defect in astrocyte proliferation by introducing into the astrocytes a nucleic acid encoding CD81, in a manner permitting expression of CD81 protein. The nucleic acid may be introduced using . conventional procedures known in the art, including, without limitation, electroporation, DEAE Dextran transfection, calcium phosphate transfection, monocationic liposome fusion, polycationic liposome fusion, protoplast fusion, creation of an in vivo electrical field, DNA-coated microprojectile bombardment, injection with recombinant replication-defective viruses, homologous recombination, in vivo gene therapy, ex vivo gene therapy, viral vectors, and naked DNA transfer, or any combination thereof. Recombinant viral vectors suitable for gene therapy include, but are not limited to, vectors derived from the genomes of viruses such as retro virus, HSV, adenovirus, adeno-associated virus, Semiliki Forest virus, cytomegalovirus, and vaccinia virus. The amount of nucleic acid encoding CD81 to be used is an amount that will express CD81 protein in an amount effective to inhibit proliferation of astrocytes, as defined above. These amounts may be readily determined by the skilled artisan.

[ 0073] The present invention is also directed to methods for inhibiting proliferation of astrocytic tumor cells, by contacting astrocytic tumor cells with an amount of CD81 effective to inhibit proliferation of astrocytic tumor cells. As used herein, the term "astrocytic tumor cells" refers to a tumorigenic form of astrocytes (i.e., transformed astrocytes), and includes astrocytoma cells (i.e., cells of all astrocytomas, including, without limitation, Grades I-IV astrocytomas, anaplastic astrocytoma, astroblastoma, astrocytoma fibrillare, astrocytoma protoplasmaticum, gemistocytic astrocytoma, and glioblastoma multiforme). As defined above, "CD81" includes a CD81 protein (p27), a CD81 analogue, and a CD81 derivative. [ 0074] The methods ofthe present invention may be used to inhibit proliferation of astrocytic tumor cells in vitro, or in vivo in a subject. As used herein, the term "inhibit proliferation of astrocytic tumor cells" means inhibit cell division and growth of astrocytic tumor cells, and includes limiting the proliferative rate of astrocytic tumor cells. Inhibition ofthe growth and proliferation of astrocytic tumor cells may be detected by known procedures, including any of the methods, molecular procedures, and assays disclosed herein.

[ 0075] In accordance with the methods ofthe present invention, CD81 may be contacted with astrocytic tumor cells in vitro, or in vivo in a subject, by introducing the CD81 protein into the membranes of astrocytic tumor cells, or by introducing into the astrocytic tumor cells a nucleic acid encoding CD81 in a manner permitting expression of CD81 protein. The subject may be any animal, but is preferably a mammal (e.g., humans, domestic animals, and commercial animals). More preferably, the subject is a human. The astrocytic tumor cells may be found in neural tissue and other tissue ofthe nervous system ofthe subject, either alone or with other types of cells, including, without limitation, neurons and oligodendroglia. Astrocytic tumor cells may be detected in tissue ofthe subject by standard detection methods readily determined from the known art, including, without limitation, immuno logical techniques (e.g., immunohistochemical staining), fluorescence imaging techniques, and microscopic techniques.

[ 0076] CD81 protein may be introduced into the membranes of astrocytic tumor cells, either in vitro or in vivo in a subject, by known techniques used for the introduction of proteins (e.g., liposome delivery), as described above. For liposome delivery, liposomal vesicles and liposome compositions may be prepared using a variety of conventional techniques, including those described above. CD81 protein may be incorporated into the layers of a liposome such that its extracellular domain extends outside the liposome, and its intracellular domain extends into the interior ofthe liposome. The liposome containing CD81 then may be fused with astrocytic tumor cells, in accordance with known methods of fusion of liposomes to cell membranes, such that the CD81 protein is delivered into the membrane ofthe astrocytic tumor cells. The amount of CD81 protein to be used is an amount effective to inhibit proliferation of astrocytic tumor cells, as defined above, and may be readily determined by the skilled artisan. [ 0077] In the method ofthe present invention, CD81 also may be contacted with astrocytic tumor cells, either in vitro or in vivo in a subject, by introducing into a sufficient number of astrocytic tumor cells ofthe subject a nucleic acid encoding CD81, in a manner permitting expression of CD81. The nucleic acid may be introduced using conventional procedures known in the art, including in vivo gene therapy, ex vivo gene therapy, and all other above-described procedures. Recombinant viral vectors suitable for gene therapy include all of the vectors described above. The amount of nucleic acid encoding CD81 to be used is an amount that will express CD81 protein in an amount effective to inhibit proliferation of astrocytic tumor cells, as defined above. These amounts may be readily determined by the skilled artisan. [ 0078] The ability of CDS 1 to modulate astrocyte proliferation, and the absence of CD81 from astrocytic tumor cell lines, together suggest that CD81 may be useful for treating astrocytomas and other conditions associated with proliferation of astrocytic tumor cells. Furthermore, it is believed that CD81 would be effective either alone or in combination with therapeutic agents, such as chemotherapeutic agents or antiviral agents, which are typically used in the treatment of these conditions.

[ 0079] Accordingly, the present invention provides a method for treating a condition associated with proliferation of astrocytic tumor cells in a subject in need of treatment, comprising contacting astrocytic tumor cells in the subject with an amount of CD81 effective to inhibit proliferation of astrocytic tumor cells, thereby treating the condition. As described above, the subject may be any animal, but is preferably a mammal (e.g., humans, domestic animals, and commercial animals). More preferably, the subject is a human.

[ 0080] As used herein, the term "conditions associated with proliferation of astrocytic tumor cells" includes pathologic proliferation of astrocytic tumor cells, such as astrocytoma cells, and other forms of neoplasia. The term "neoplasia", as further used herein, refers to the uncontrolled and progressive multiplication of astrocytic tumor cells under conditions that would not elicit, or would cause cessation of, multiplication of normal astrocytes. Neoplasia results in the formation of a "neoplasm", which is defined herein to mean any new and abnormal growth, particularly a new growth of tissue, in which the growth of cells is uncontrolled and progressive. Neoplasms include benign tumors and malignant tumors (e.g., astrocytomas, such as Grades I-IV astrocytomas, anaplastic astrocytoma, astroblastoma, astrocytoma fibrillare, astrocytoma protoplasmaticum, gemistocytic astrocytoma, and glioblastoma multiforme, and other brain tumors). Malignant neoplasms are distinguished from benign in that the former show a greater degree of anaplasia, or loss of differentiation and orientation of cells, and have the properties of invasion and metastasis. Thus, neoplasia includes "cancer", which herein refers to a proliferation of astrocytic tumor cells having the unique trait of loss of normal controls, resulting in unregulated growth, lack of differentiation, local tissue invasion, and metastasis. In one embodiment ofthe present invention, the condition associated with proliferation of astrocytic tumor cells is an astrocytoma.

[ 0081] In the treatment of a condition associated with proliferation of astrocytic tumor cells, CD81 may be contacted with astrocytic tumor cells by introducing the CD81 protein into the membranes of astrocytic tumor cells, in accordance with known methods (e.g., liposome delivery), as described above. The amount of CD81 protein to be used is an amount effective to inhibit proliferation of astrocytic tumor cells, as defined above, and may be readily determined by the skilled artisan.

[ 0082] Alternatively, in accordance with known methods, including those described above, CD81 may be contacted with astrocytic tumor cells to treat a condition associated with a defect in astrocytic tumor cell proliferation by introducing into the astrocytic tumor cells a nucleic acid encoding CD81, in a manner permitting expression of CD81 protein. The nucleic acid may be introduced using conventional procedures known in the art, including in vivo gene therapy, ex vivo gene therapy, and all above-described procedures. Recombinant viral vectors suitable for gene therapy include all vectors described above. The amount of nucleic acid encoding CD81 to be used is an amount that will express CD81 protein in an amount effective to inhibit proliferation of astrocytic tumor cells, as defined above. These amounts may be readily determined by the skilled artisan.

[ 0083] The present invention further provides methods for inhibiting proliferation of astrocytes, comprising contacting astrocytes with a modulator of CD81 expression, in an amount effective to inhibit proliferation of astrocytes. The modulator may be a protein, polypeptide, peptide, nucleic acid (including DNA or RNA), antibody, Fab fragment, F(ab')₂ fragment, molecule, compound, antibiotic, drug, neuron, or other agent, as defined herein, that induces or upregulates CD81 expression. Examples of modulators include, without limitation, neurons, FK506, and other neuroimmunophilins.

[ 0084] Additional modulators of CD81 may be identified using a simple screening assay based on procedures described below. For example, to screen for candidate modulators of CD81, astrocytic tumor cells may be plated onto microtiter plates, then contacted with a library of drugs. Any resulting expression of CD81 may be detected using nucleic acid hybridization and/or immuno logical techniques known in the art, including an ELISA. Modulators of CD81 will be those drugs which induce or upregulate expression of CD81. In this manner, agents also may be screened for their ability to inhibit proliferation of astrocytes or astrocytic tumor cells using CD81 expression as an indicator that cell division or growth of astrocytes or astrocytic tumor cells is decreasing in rate, or has stopped.

[ 0085] The present invention further provides methods for inhibiting proliferation of astrocytic tumor cells, comprising contacting astrocytic tumor cells with a modulator of CD81 expression, in an amount effective to inhibit proliferation of astrocytic tumor cells. Examples of such modulators of CD81 expression include all of those described above. Additional modulators of CD81 may be screened in accordance with the above-described methods. [ 0086] The present invention also provides methods for treating a condition associated with a defect in astrocyte proliferation in a subject in need of treatment. The methods comprise administering to the subject an amount of CD81 effective to treat the condition associated with a defect in cell proliferation. The subject may be any animal, but is preferably a mammal (e.g., humans, domestic animals, and commercial animals). More preferably, the subject is a human. [ 0087] As described above, examples of conditions associated with a defect in astrocyte proliferation include, without limitation, astrocytosis, glial scars, hyperplasia, neoplasia, and neuritic plaques (particularly those commonly found in Alzheimer's disease patients). Additionally, conditions associated with a defect in astrocyte proliferation may be caused by, or associated with, a variety of factors, including, without limitation, neuronal cell death and neural degeneration resulting from neurodegenerative diseases, CNS traumas, and the acquired secondary effects of non-neural dysfunction. Examples of neurodegenerative diseases, CNS traumas, and acquired secondary effects of non-neural dysfunction include all of those described above. In one embodiment ofthe present invention, the condition associated with a defect in astrocyte proliferation is astrocytosis.

[ 0088] The CD81 ofthe present invention is administered to a subject in need of treatment for a condition associated with a defect in astrocyte proliferation in an amount that is effective to treat the condition associated with a defect in astrocyte proliferation in the subject. As used herein, the phrase "effective to treat the condition associated with a defect in astrocyte proliferation" means effective to ameliorate or minimize the clinical impairment or symptoms of the condition associated with a defect in astrocyte proliferation. For example, where the condition associated with a defect in astrocyte proliferation is astrocytosis, the clinical impairment or symptoms ofthe astrocytosis may be ameliorated or minimized by reducing the mass of astrocytes produced by the astrocytosis, thereby minimizing any potential obstruction of axons which may occur. The amount of CD81 effective to treat a condition associated with a defect in astrocyte proliferation in a subject in need of treatment therefor will vary depending upon the particular factors of each case, including the type of defect in astrocyte proliferation, the stage ofthe defect in astrocyte proliferation, the subject's weight, the severity ofthe subject's condition, and the method of administration. This amount can be readily determined by the skilled artisan. [ 0089] According to the method ofthe present invention, CD81 may be administered to a human or animal subject by known procedures, including, without limitation, oral administration, parenteral administration, transdermal administration, and administration through an osmotic mini- pump. Preferably, the CD81 is administered parenterally, by intracranial, intraspinal, intrathecal, or subcutaneous injection. The CD81 ofthe present invention also may be administered to a subject in accordance with any ofthe above-described methods for effecting in vivo contact between astrocytes and CD81.

[ 0090] For oral administration, the formulation of CD81 may be presented as capsules, tablets, powders, granules, or as a suspension. The formulation may have conventional additives; such as lactose, mannitol, corn starch, or potato starch. The formulation also may be presented with binders, such as crystalline cellulose, cellulose derivatives, acacia, corn starch, or gelatins. Additionally, the formulation may be presented with disintegrators, such as corn starch, potato starch, or sodium carboxymethylcellulose. The formulation also may be presented with dibasic calcium phosphate anhydrous or sodium starch glycolate. Finally, the formulation may be presented with lubricants, such as talc or magnesium stearate.

[ 0091] For parenteral administration (i.e., administration by injection through a route other than the alimentary canal), CD81 may be combined with a sterile aqueous solution that is preferably isotonic with the blood ofthe subject. Such a formulation may be prepared by dissolving a solid active ingredient in water containing physiologically-compatible substances, such as sodium chloride, glycine, and the like, and having a buffered pH compatible with physiological conditions, so as to produce an aqueous solution, then rendering said solution sterile. The formulations may be presented in unit or multi-dose containers, such as sealed ampoules or vials. The formulation may be delivered by any mode of injection, including, without limitation, epifascial, intracapsular, intracranial, intracutaneous, intrathecal, intramuscular, intraorbital, intraperitoneal, intraspinal, intrasternal, intravascular, intravenous, parenchymatous, or subcutaneous.

[ 0092] For transdermal administration, CD81 may be combined with skin penetration enhancers, such as propylene glycol, polyethylene glycol, isopropanol, ethanol, oleic acid, N- methylpyrrolidone, and the like, which increase the permeability ofthe skin to the CD81, and permit the CD81 to penetrate through the skin and into the bloodstream. The CD81/enhancer compositions also may be further combined with a polymeric substance, such as ethylcellulose, hydroxypropyl cellulose, ethylene/vinylacetate, polyvinyl pyrrolidone, and the like, to provide the composition in gel form, which may be dissolved in solvent, such as methylene chloride, evaporated to the desired viscosity, and then applied to backing material to provide a patch. CD81 may be administered transdermally at the site in the subject where neural trauma has occurred, or where the defect in astrocyte proliferation is localized. Alternatively, CD81 may be administered transdermally at a site other than the affected area, in order to achieve systemic administration.

[ 0093] The CD81 ofthe present invention also may be released or delivered from an osmotic mini-pump or other time-release device. The release rate from an elementary osmotic mini-pump may be modulated with a microporous, fast-response gel disposed in the release orifice. An osmotic mini-pump would be useful for controlling release, or targeting delivery, of CD81.

[ 0094] The present invention also provides methods for treating a condition associated with proliferation of astrocytic tumor cells in a subject in need of treatment therefor. The methods comprise administering to the subject an amount of CD81 effective to treat the condition associated with proliferation of astrocytic tumor cells. The subject may be any animal, but is preferably a mammal (e.g., humans, domestic animals, and commercial animals). More preferably, the subject is a human. As described above, examples of conditions associated with proliferation of astrocytic tumor cells include, without limitation, astrocytomas, brain tumors, and other forms of neoplasia. In one embodiment ofthe present invention, the condition associated with proliferation of astrocytic tumor cells is an astrocytoma.

[ 0095] The CD81 ofthe present invention is administered to a subject in need of treatment for a condition associated with proliferation of astrocytic tumor cells in an amount that is effective to treat the condition associated with proliferation of astrocytic tumor cells in the subject. As used herein, the phrase "effective to treat the condition associated with proliferation of astrocytic tumor cells" means effective to ameliorate or minimize the clinical impairment or symptoms ofthe condition associated with proliferation of astrocytic tumor cells.

[ 0096] For example, where the condition associated with proliferation of astrocytic tumor cells is an astrocytoma, the clinical impairment or symptoms ofthe astrocytoma may be ameliorated or minimized by diminishing any pain or discomfort suffered by the subject; by extending the survival ofthe subject beyond that which would otherwise be expected in the absence of such treatment; by inhibiting or preventing the development or spread ofthe neoplasm; or by limiting, suspending, terminating, or otherwise controlling the maturation and proliferation of astrocytic tumor cells in the astrocytoma. The amount of CD81 effective to treat a condition associated with proliferation of astrocytic tumor cells in a subject in need of treatment will vary depending upon the particular factors of each case, including the type of condition associated with proliferation of astrocytic tumor cells, the stage ofthe condition associated with proliferation of astrocytic tumor cells, the subject's weight, the severity ofthe subject's condition, and the method of administration. This amount can be readily determined by the skilled artisan. [ 0097] According to the method ofthe present invention, CD81 may be administered to a human or animal subject by known procedures, including, without limitation, oral administration, parenteral administration, transdermal administration, and administration through an osmotic mini- pump. Preferably, the CD81 is administered parenterally, by intracranial, intraspinal, intrathecal, or subcutaneous injection. The CD81 ofthe present invention also may be administered to a subject in accordance with any ofthe above-described methods for effecting in vivo contact between astrocytic tumor cells and CD81.

[ 0098] For oral administration, the formulation of CD81 may be presented as capsules, tablets, powders, granules, as a suspension, or in any ofthe above-described formulations. For parenteral administration, CD81 may be combined with a sterile aqueous solution which is preferably isotonic with the blood ofthe subject. Such a formulation may be prepared in accordance with the above-described method of preparation. The formulations for parenteral administration may be presented in unit or multi-dose containers, such as sealed ampoules or vials, and may be delivered by any ofthe modes of injection described above. [ 0099] For transdermal administration, CD81 may be combined with skin penetration enhancers, such as those described above. The CDSl/enhancer compositions also may be further combined with a polymeric substance, such as any of those described above, to provide the composition in gel form. CDS 1 may be administered transdermally at the site in the subject where astrocytic tumor cell proliferation has occurred. Alternatively, CD81 may be administered transdermally at a site other than the affected area, in order to achieve systemic administration. Finally, the CD81 ofthe present invention also may be released or delivered from an osmotic mini-pump or other time-release device, as described above. [ 0100] The present invention further provides methods for treating a condition associated with a defect in astrocyte proliferation in a subject in need of treatment. The methods comprise administering to the subject a modulator of CD 81 expression in an amount effective to induce or enhance expression of CD81 and treat a condition associated with a defect in astrocyte proliferation, as defined above, in the subject. Examples of such modulators of CD81 expression include all of those described above. Additional modulators of CD81 may be screened in accordance with the above-described methods. The modulator of CD81 may be administered to a subject in any ofthe formulations, and by any ofthe modes of administration, described above. [ 0101] The present invention also provides methods for treating a condition associated with proliferation of astrocytic tumor cells in a subject in need of treatment. The methods comprise administering to the subject a modulator of CD81 expression in an amount effective to induce or enhance expression of CD81 and treat the condition associated with proliferation of astrocytic tumor cells, as defined above, in the subject. Examples of such modulators of CDS 1 expression include all of those described above. Additional modulators of CD81 may be screened in accordance with the above-described methods. The modulator of CDS 1 may be administered to a subject in any ofthe formulations, and by any ofthe modes of administration, described herein. Moreover, the modulator of CD81 also may be administered along with a chemotherapeutic agent, such as a ricin-conjugated CD81 -binding protein. [ 0102] In view ofthe foregoing, it is predicted that administration of CD81 will provide an effective treatment option for conditions associated with either a defect in astrocyte proliferation or a proliferation of astrocytic tumor cells. The therapies described herein offer real treatment options for inhibiting astrocyte and astrocytoma proliferation, without the massive side- effects and bystander effects that typically accompany the current treatment regimes. The population at risk for these conditions is large, and the needs currently are not being met. [ 0103] The present invention further provides pharmaceutical compositions comprising

CD81 and a pharmaceutically-acceptable carrier, wherein CD81 is present in an amount sufficient or effective to treat a condition associated with a defect in astrocyte proliferation, as defined above, in a subject to whom said pharmaceutical composition is administered. Such a pharmaceutical composition would be useful for administering CD81 to a subject in need of treatment for a condition associated with a defect in astrocyte proliferation, in order to treat said condition. The CD 81 is provided to the subject in an amount that is effective to treat the condition associated with a defect in astrocyte proliferation, as defined above, in the subject. This amount may be readily determined by the skilled artisan. The pharmaceutical composition may be administered to a subject in accordance with any ofthe methods of administration described above.

[ 0104] Formulations ofthe pharmaceutical composition ofthe present invention may be conveniently presented in unit dosage, and may be presented in oral dosage form (e.g., CD81 and a pharmaceutically-acceptable carrier may be combined in an ampule, capsule, pill, powder, or tablet) or in a form suitable for injection. The pharmaceutically-acceptable carrier may be a solid, liquid, or gel. Furthermore, the pharmaceutically-acceptable carrier ofthe present invention must be "acceptable" in the sense of being compatible with the other ingredients ofthe composition, and not deleterious to the recipient thereof. Examples of acceptable pharmaceutical carriers include carboxymethylcellulose, crystalline cellulose, glycerin, gum arabic, lactose, magnesium stearate, methyl cellulose, polypeptides, powders, saline, sodium alginate, starch, sucrose, talc, and water, among others. The carrier selected will depend upon the route of administration, and the form in which CD 81 is introduced.

[ 0105] The formulations ofthe present invention may be prepared by methods well known in the pharmaceutical art. For example, CD81 may be brought into association with a carrier or diluent, as an emulsion, suspension, or solution. Moreover, CD81 may be blended, at need, with another component, to the extent that such blending does not impair the object ofthe present invention. Such other component may be suitably selected in accordance with the purpose of use and type of formulation. Optionally, one or more accessory ingredients (e.g., buffers, colorants, flavoring agents, surface active agents, and the like) also may be added.

[ 0106] The present invention is also directed to pharmaceutical compositions comprising a nucleic acid encoding CD81 and a pharmaceutically-acceptable carrier, wherein the nucleic acid expresses CD81 in an amount sufficient or effective to treat a condition associated with a defect in astrocyte proliferation, as defined above, in a subject to whom said pharmaceutical composition is administered. Such pharmaceutical compositions are useful for administering CD81 to a subject in need of treatment for a condition associated with a defect in astrocyte proliferation, in order to treat said condition in the subject. The nucleic acid is provided to the subject in an amount such » that it expresses CD81 protein in an amount that is effective to treat a condition associated with a defect in astrocyte proliferation, as defined above, in the subject. These amounts may be readily determined by the skilled artisan. Additionally, the pharmaceutical composition may be administered to a subject in accordance with any ofthe above-described methods of administration and introduction of nucleic acids.

[ 0107] Formulations ofthe pharmaceutical composition ofthe present invention may be conveniently presented in unit dosage, and may be presented in a form suitable for administration of nucleic acid (e.g., by injection). The nucleic acid encoding CD81 may be presented in any form well known in the art for introduction of nucleic acids, including, without limitation, naked DNA, plasmid DNA, and vector DNA (including viral vectors, as described above), and may be prepared in accordance with methods well known in the arts of gene therapy and molecular genetics. In addition, the pharmaceutically-acceptable carrier ofthe present invention must be "acceptable" in the sense of being compatible with the other ingredients ofthe composition, and not deleterious to the recipient thereof. Examples of acceptable pharmaceutical carriers include carboxymethylcellulose, crystalline cellulose, glycerin, gum arabic, lactose, magnesium stearate, methyl cellulose, polypeptides, powders, saline, sodium alginate, starch, sucrose, talc, and water, among others. The carrier selected will depend upon the route of administration, and the form in which nucleic acid encoding CD 81 is introduced.

[ 0108] The formulations ofthe pharmaceutical composition ofthe present invention may be prepared by methods well known in the pharmaceutical art. For example, nucleic acid encoding CD81 may be brought into association with a carrier or diluent, as an emulsion, suspension, or solution. Moreover, nucleic acid encoding CD81 may be blended, at need, with another component, to the extent that such blending does not impair the object ofthe present invention. Such other component may be suitably selected in accordance with the purpose of use and type of formulation. Optionally, one or more accessory ingredients (e.g., buffers, colorants, surface active agents, and the like) also may be added.

[ 0109] The present invention further provides pharmaceutical compositions comprising

CD81 and a pharmaceutically-acceptable carrier, wherein CD81 is present in an amount sufficient or effective to treat a condition associated with proliferation of astrocytic tumor cells, as defined above, in a subject to whom said pharmaceutical composition is administered. Such a pharmaceutical composition would be useful for administering CD81 to a subject in need of treatment for a condition associated with proliferation of astrocytic tumor cells, in order to treat said condition in the subject. The CD81 is provided to the subject in an amount that is effective to treat the condition associated with proliferation of astrocytic tumor cells, as defined above, in the subject. This amount may be readily determined by the skilled artisan. The pharmaceutical composition may be administered to a subject in accordance with any ofthe methods of administration, and in any ofthe formulations, described above. The formulations ofthe present invention may be prepared in accordance with methods well known in the pharmaceutical art, including those described above.

[ 0110] The present invention is also directed to pharmaceutical compositions comprising a nucleic acid encoding CD 81 and a pharmaceutically-acceptable carrier, wherein the nucleic acid expresses CD81 in an amount sufficient or effective to treat a condition associated with proliferation of astrocytic tumor cells, as defined above, in a subject to whom said pharmaceutical composition is administered. Such a pharmaceutical composition would be useful for administering CD81 to a subject in need of treatment for a condition associated with proliferation of astrocytic tumor cells, in order to treat said condition in the subject. The nucleic acid is provided to the subject in an amount such that it expresses CD81 protein in an amount that is effective to treat a condition associated with proliferation of astrocytic tumor cell, as defined above, in the subject. These amounts may be readily determined by the skilled artisan. Additionally, the pharmaceutical composition may be administered to a subject in accordance with any ofthe above-described methods of administration and introduction of nucleic acids, and in any ofthe formulations described above. The formulations ofthe pharmaceutical composition ofthe present invention may be prepared in accordance with methods well known in the pharmaceutical art, including those described above.

[ 0111] The present invention further provides methods for determining whether a subject has an astrocytoma. The methods comprise assaying for CD81 expression in a diagnostic sample of cells of astrocytic lineage ofthe subject, wherein no detection of expression of CD81 in cells of astrocytic lineage ofthe subject is diagnostic of an astrocytoma. The subject may be any animal, but is preferably a mammal (e.g., humans, domestic animals, and commercial animals). More preferably, the subject is a human. As used herein, "CD81" includes CD81 protein, cDNA, and mRNA.

[ 0112] As used herein, "no detection of expression of CD81" means that CD81 is not present in astrocytic tumor cells ofthe subject at a detectable level. As further used herein, the term "cells of astrocytic lineage" includes astrocytes and astrocytic tumor cells, as defined above. It is also within the confines ofthe present invention to provide a method for confirming a diagnosis of astrocytoma in a subject, comprising assaying for CD81 expression a diagnostic sample of cells of astrocytic lineage ofthe subject, wherein no detection of expression of CDS 1 in cells of astrocytic lineage ofthe subject is diagnostic of an astrocytoma.

[ 0113] According to the method ofthe present invention, the diagnostic sample of cells of astrocytic lineage ofthe subject may be assayed for CD81 expression in vitro, or in vivo in a subject. In accordance with the present invention, where the assay is performed in vitro, a diagnostic sample of cells of astrocytic lineage, or tissue containing cells of astrocytic lineage, may be removed from the subject using standard procedures, including biopsy and aspiration. Preferably, the diagnostic sample of cells or tissue is removed using multidirectional fine-needle aspiration biopsy (FNAB). This method of removal is preferred, as it is less invasive than a standard biopsy. The diagnostic sample taken from the subject may be, for example, any tissue known to have an astrocytoma, any tissue suspected of having an astrocytoma, or any tissue believed not to have an astrocytoma.

[ 0114] Protein may be isolated and purified from the diagnostic sample ofthe present invention using standard methods known in the art, including, without limitation, extraction from a tissue (e.g., with a detergent that solubilizes the protein) where necessary, followed by affinity purification on a column, chromatography (e.g., FTLC and HPLC), immunoprecipitation (with an antibody to CD81), and precipitation (e.g., with isopropanol and a reagent such as Trizol). Isolation and purification ofthe protein may be followed by electrophoresis (e.g., on an SDS- polyacrylamide gel). Nucleic acid may be isolated from a diagnostic sample using standard techniques known to one of skill in the art.

[ 0115] In accordance with the method of he present invention, an astrocytoma in a subject may be diagnosed by assaying a diagnostic sample ofthe subject for expression of CD81. Because CD81 is generally expressed in cells of astrocytic lineage from healthy, nondiseased subjects (i.e., those who do not have an astrocytoma), no detection of CD81 expression in a diagnostic sample of cells of astrocytic lineage of a subject is diagnostic of an astrocytoma. As used herein, "expression" means the transcription ofthe CD81 gene into at least one mRNA transcript, or the translation of at least one mRNA into a CD81 protein, as defined above. Accordingly, a diagnostic sample maybe assayed for CD81 expression by assaying for CD81 protein (as defined above), cDNA, or mRNA. The appropriate form of CD81 will be apparent based on the particular techniques discussed herein.

[ 0116] In these methods, a diagnostic sample of cells of astrocytic lineage a subject may be assayed for CD81 expression, and CD81 expression may be detected in a diagnostic sample, using assays and detection methods readily determined from the known art, including, without limitation, immunological techniques, hybridization analysis, fluorescence imaging techniques, and/or radiation detection. For example, astrocytes or cells that are removed from the subject using FNAB may be analyzed using immuno cytofluorometry (FACS analysis). In another embodiment ofthe present invention, the diagnostic sample is assayed for expression of CD81 using Northern blot analysis of CD81 mRNA extracted from cells of astrocytic lineage. [ 0117] According to these methods, a diagnostic sample ofthe subject may be assayed for

CD81 expression using an agent reactive with CD81. As used herein, "reactive" means the agent has affinity for, binds to, or is directed against CD81. As further used herein, an "agent" shall include a protein, polypeptide, peptide, nucleic acid (including DNA or RNA), antibody, Fab fragment, F(ab')₂ fragment, molecule, compound, antibiotic, drug, and any combinations thereof. Moreover, an agent reactive with CDSl may be either natural or synthetic. The agent may be in the form of an antibody, a Fab fragment, an F(ab')₂ fragment, a peptide, a polypeptide, a protein, and any combinations thereof. A Fab fragment is a univalent, antigen-binding fragment of an antibody, which is produced by papain digestion. An F(ab')₂ fragment is a divalent antigen- binding fragment of an antibody, which is produced by pepsin digestion. Preferably, the agent is a high-affinity antibody labeled with a detectable marker. Where the agent is an antibody, the absence of expression of CD81 may be detected from binding studies using one or more antibodies immunoreactive with CDSl, along with standard immunological detection techniques, such as Western blotting.

[ 0118] As used herein, the antibody ofthe present invention may be polyclonal or monoclonal, and may be produced by techniques well known to those skilled in the art. Polyclonal antibody, for example, may be produced by immunizing a mouse, rabbit, or rat with purified CD81. Monoclonal antibody then may be produced by removing the spleen from the immunized mouse, and fusing the spleen cells with myeloma cells to form a hybridoma which, when grown in culture, will produce a monoclonal antibody. Monoclonal antibodies that are reactive with CD81 also may be obtained from Pharmingen (San Diego, CA) (e.g., mAb 2F7) and Boehringer (Mannheim, Germany) (e.g., mAbs Eatl and Eat2).

[ 0119] The antibodies used herein may be labeled with a detectable marker. Labeling of the antibody may be accomplished using one ofthe variety of different chemiluminescent and radioactive labels known in the art. The detectable marker ofthe present invention may be, for example, a nonradio active or fluorescent marker, such as biotin, fluorescein (FITC), acridine, cholesterol, or carboxy-X-rhodamine, which can be detected using fluorescence and other imaging techniques readily known in the art. Alternatively, the detectable marker may be a radioactive marker, including, for example, a radioisotope. The radioisotope may be any isotope that emits detectable radiation, such as ³⁵S, ³²P, or ³H. Radioactivity emitted by the radioisotope can be detected by techniques well known in the art. For example, gamma emission from the radioisotope may be detected using gamma imaging techniques, particularly scintigraphic imaging. Preferably, the agent ofthe present invention is a high-affinity antibody labeled with a detectable marker. The antibodies ofthe present invention also may be incorporated into kits that include an appropriate labeling system, buffers, and other necessary reagents for use in a variety of detection and diagnostic applications.

[ 0120] Where the agent ofthe present invention is an antibody reactive with CD81, a diagnostic sample taken from the subject may be purified by passage through an affinity column which contains CD81 antibody as a ligand attached to a solid support such as an insoluble organic polymer in the form of a bead, gel, or plate. The antibody attached to the solid support may be used in the form of a column. Examples of suitable solid supports include, without limitation, agarose, cellulose, dextran, polyacrylamide, polystyrene, sepharose, or other insoluble organic polymers. The CD81 antibody may be further attached to the solid support through a spacer molecule, if desired. Appropriate binding conditions (e.g., temperature, pH, and salt concentration) may be readily determined by the skilled artisan. In a preferred embodiment, the CDSl antibody is attached to a sepharose column, such as Sepharose 4B. [ 0121] Where the agent is an antibody, a diagnostic sample ofthe subject may be assayed for CD81 expression using binding studies that utilize one or more antibodies immunoreactive with CD81, along with standard immunological detection techniques. For example, the CD81 protein eluted from the affinity column may be subjected to an ELISA assay, Western blot analysis, flow cytometry, or any other immunostaining method employing an antigen-antibody interaction. Preferably, the diagnostic sample is assayed for CD81 expression using Western blotting.

[ 0122] Alternatively, a diagnostic sample of cells of astrocytic lineage of a subject may be assayed for CDSl expression using hybridization analysis of nucleic acid extracted from a sample of cells of astrocytic lineage, or tissue containing cells of astrocytic lineage, taken from the subject. According to this method ofthe present invention, the hybridization analysis may be conducted using Northern blot analysis of mRNA. This method also may be conducted by performing a Southern blot analysis of DNA using one or more nucleic acid probes which hybridize to nucleic acid encoding CD81. The nucleic acid probes may be prepared by a variety of techniques known to those skilled in the art, including, without limitation, the following: restriction enzyme digestion of CD 81 nucleic acid; and automated synthesis of oligonucleotides having sequences that correspond to selected portions ofthe nucleotide sequence ofthe CD81 nucleic acid, using commercially-available oligonucleotide synthesizers, such as the Applied Biosystems Model 392 DNA/RNA synthesizer.

[ 0123] The nucleic acid probes used in the present invention may be DNA or RNA, and may vary in length from about 8 nucleotides to the entire length ofthe CD81 nucleic acid. The CD81 nucleic acid used in the probes may be derived from mammalian CD81. The nucleotide sequences for both rat, mouse, and human CDSl are known (19). Using these sequences as probes, the skilled artisan could readily clone corresponding CD81 cDNA from other species. In addition, the nucleic acid probes ofthe present invention may be labeled with one or more detectable markers. Labeling of he nucleic acid probes may be accomplished using one of a number of methods known in the art (e.g., nick translation, end labeling, fill-in end labeling, polynucleotide kinase exchange reaction, random priming, or SP6 polymerase for riboprobe preparation), along with one of a variety of labels (e.g., radioactive labels, such as ³⁵S, ³²P, or ³H, or nonradio active labels, such as biotin, fluorescein (FITC), acridine, cholesterol, or carboxy-X- rhodamine (ROX)). Combinations of two or more nucleic acid probes (or primers), corresponding to different or overlapping regions ofthe CD81 nucleic acid, also may be used to detect expression of CDSl, using, for example, PCR or RT-PCR, and may be included in kits for use in a variety of detection and diagnostic applications.

[ 0124] It is contemplated that the diagnostic sample in the present invention frequently will be assayed for CD81 expression not by the subject, nor by his/her consulting physician, but by a laboratory technician or other clinician. Accordingly, the method ofthe present invention further comprises providing to a subject's consulting physician a report ofthe results obtained upon assaying a diagnostic sample ofthe subject for CD81 expression.

[ 0125] The present invention also provides methods for treating astrocytoma in a subject or patient. The subject may be any animal, but is preferably a mammal (e.g., humans, domestic animals, and commercial animals). More preferably, the subject is a human. The methods comprise the steps of: (a) diagnosing an astrocytoma in the subject or patient by detecting an absence of expression of CDSl in cells of astrocytic lineage ofthe subject or patient; and (b) treating the astrocytoma diagnosed in the subject or patient. The absence of expression of CDSl in cells of astrocytic lineage ofthe subject or patient may be detected by any ofthe methods described above. The astrocytoma diagnosed in the subject or patient may be treated by any method or combination of methods commonly used to treat astrocytoma, including, without limitation, surgery, radiotherapy, chemotherapy, immunotherapy, and systemic therapy. Preferably, however, an astrocytoma which is diagnosed in accordance with the method described herein is treated by administering CD81 to the subject or patient, as described above. [ 0126] It is also within the confines ofthe present invention to use detected levels of

CD81 expression as a clinical or pathologic staging tool, to determine which treatment options may be appropriate. In particular, detection of CD81 expression may be used to determine whether any ofthe treatment methods ofthe present invention is appropriate. Moreover, detected levels of CD81 expression may be used to grade brain tumors, particularly astrocytomas. [ 0127] The present invention further provides methods for assessing the efficacy of . astrocytoma therapy in a subject who has undergone or is undergoing treatment for astrocytoma. The subject may be any animal, but is preferably a mammal (e.g., humans, domestic animals, and commercial animals). More preferably, the subject is a human. The method ofthe present invention comprises assaying for CD81 expression a diagnostic sample of astrocytic tumor cells of the subject, wherein no detection of expression of CD81 in astrocytic tumor cells ofthe subject is indicative of unsuccessful astrocytoma therapy. The diagnostic sample may be any of those described above, and may be assayed for expression of CDSl either in vitro or in vivo in a subject. In addition, the diagnostic sample may be assayed for expression of CD81 using all ofthe various assays and methods of detection described above. This method ofthe present invention provides a means of monitoring the effectiveness of astrocytoma therapy by permitting the periodic assessment of levels of CD81 expression in astrocytic tumor cells ofthe subject. [ 0128] According to these methods, a diagnostic sample of astrocytic tumor cells of a subject may be assayed, and levels of CD81 expression may be assessed, at any time following the initiation of therapy to treat an astrocytoma. For example, levels of CD81 expression maybe assessed while the subject or patient is still undergoing treatment for the astrocytoma. Where expression of CD 81 remains absent from astrocytic tumor cells ofthe subject, a physician may choose to continue with the astrocytoma treatment. Where levels of CD81 expression become detectable in astrocytic tumor cells ofthe subject, and then increase through successive assessments, it may be an indication that the astrocytoma treatment is working, and that treatment doses could be decreased or even ceased. Where levels of CD81 do not noticeably increase through successive assessments, it may be an indication that the astrocytoma treatment is not working, and that treatment doses could be increased. Where CD81 expression is eventually detected in astrocytic tumor cells of a subject or patient at a level expected for normal, non- diseased astrocytes, a physician may conclude that the astrocytoma treatment has been successful, and that such treatment may cease. It is also within the confines of he present invention to assess levels of CD81 expression following completion ofthe subject's or patient's astrocytoma treatment, in order to determine whether the astrocytoma has recurred in the subject or patient. Furthermore, it is within the confines ofthe present invention to use assessed levels of CDSl expression as a clinical or pathologic staging tool, to determine the extent of astrocytoma in the subject or patient, to determine appropriate treatment options, and to provide prognostic information.

[ 0129] The present invention also provides methods for enhancing survival of neurons, comprising contacting the neurons with CD 81 protein or a CD 81 derivative, in an amount effective to enhance survival ofthe neurons. As used herein, "neurons" are any ofthe conducting or nerve cells ofthe nervous system that typically consist of a cell body (perikaryon) that contains the nucleus and surrounding cytoplasm; several short, radiating processes (dendrites); and one long process (the axon), which terminates in twig-like branches (telodendrons), and which may have branches (collaterals) projecting along its course. In the method ofthe present invention, the neurons may be contained in neural tissue and other tissue ofthe nervous system, either alone or with other types of neural cells, including, without limitation, astrocytes and oligodendroglia. Neurons may be detected in tissue by standard detection methods readily determined from the known art, examples of which include, without limitation, immunological techniques (e.g., immunohisto chemical staining), fluorescence imaging techniques, and microscopic techniques. [ 0130] As further used herein, the term "enhancing survival of neurons" means the full or partial protection of neurons from damage, death, degeneration, demyelination, or injury. Enhancement ofthe survival of neurons may be measured or detected by known procedures, including any ofthe methods, molecular procedures, and assays disclosed herein. [ 0131] Additionally, unless otherwise indicated, "CD81", as used herein, includes both a

CD81 protein (p27) and a CDSl analogue, as defined above. Additionally, as used herein, a "CD81 derivative" is a chemical substance derived from CD81, either directly or by modification, truncation, or partial substitution. For example, the CD81 derivative for use in the present invention may be the extracellular domain (ECD) of CD81. In addition, the CD81 derivative of the present invention may be the mimetics, GMl 109 or GM1416, as disclosed herein. Since they are smaller than CD81, the derivatives of CD81 more easily pass through the blood-brain barrier, thereby providing suitable therapeutics for enhancing survival of neurons in vivo. [ 0132] These methods may be used to enhance survival of neurons in vitro, or in vivo in a subject. The CD81 or CD81 derivative may be contacted with neurons in vitro, or in vivo in a subject, by introducing to the neurons the CD81 or CD81 -derivative protein, or by introducing to the neurons a nucleic acid encoding CD81 or the CD81 derivative in a manner permitting expression ofthe CD81 or CD81 -derivative protein. As described above, the subject may be any animal, but is preferably a mammal (e.g., humans, domestic animals, and commercial animals). More preferably, the subject is a human.

[ 0133] CDSl or a CD81 derivative may be contacted with neurons in vitro by adding

CD81 protein or a CD81 -derivative protein directly to the culture medium. Alternatively, CD81 or a CDSl derivative may be contacted with neurons in vivo in a subject by introducing into, or administering to, the subject the CD81 protein or a CD81 -derivative protein. The CDSl protein or the CD81 -derivative protein may be contacted with neurons, either in vitro or in vivo in a subject, by known techniques used for the introduction and administration of proteins and other drugs, including, for example, injection and transfusion. The amount of CD81 protein or CD81- derivative protein to be used is an amount effective to enhance survival of neurons, as defined above, and may be readily determined by the skilled artisan.

[ 0134] It may be desirable to enhance the survival of neurons in vitro when preparing nervous tissue for transplantation, diagnostics, drug screening, and the like. In respect of in vivo treatment, the ability of CD81 or a CD81 derivative to enhance the survival of neurons renders CD81 and its derivatives particularly useful for treating neural degeneration in a subject. As used herein, the term "neural degeneration" means a condition of deterioration ofthe neuron, wherein the neuron changes to a lower or less functionally-active form. It is believed that, by enhancing the survival of neurons, CD81 or a CDSl derivative will be useful for the treatment of neural degeneration. It is further believed that CD81 or a CD81 derivative would be effective either alone or in combination with therapeutic agents, such as chemotherapeutic agents or antiviral agents, which are typically used in the treatment of neural degeneration. [ 0135] Accordingly, the present invention provides methods for treating neural degeneration in a subject in need of treatment, by contacting neurons in the subject with an amount of CD81 or a CD81 derivative effective to enhance survival of neurons, thereby treating the neural degeneration. As described above, the subject may be any animal, but is preferably a mammal (e.g., humans, domestic animals, and commercial animals). More preferably, the subject is a human.

[ 0136] Examples of neural degeneration which may be treated by the method ofthe present invention include, without limitation, primary neurologic conditions (e.g., neurodegenerative diseases), as well as acquired secondary effects of non-neural dysfunction (e.g., neural loss secondary to degenerative, pathologic, or traumatic events). Examples of primary neurologic conditions or neurodegenerative diseases include, without limitation, Alzheimer's disease, amyotrophic lateral sclerosis (Lou Gehrig's Disease), Binswanger's disease, Huntington's chorea, multiple sclerosis, myasthenia gravis, Parkinson's disease, and Pick's disease. Examples of acquired secondary effects of non-neural dysfunction include, without limitation, cerebral palsy, congenital hydro cephalus, muscular dystrophy, stroke, and vascular dementia, as well as neural degeneration resulting from any ofthe following: an injury associated with cerebral hemorrhage, developmental disorders (e.g., a defect ofthe brain, such as congenital hydro cephalus, or a defect ofthe spinal cord, such as spina bifida), diabetic encephalopathy, hypertensive encephalopathy, intracranial aneurysms, ischemia, kidney dysfunction, subarachnoid hemorrhage, trauma to the brain and spinal cord, the treatment of therapeutic agents such as chemotherapy agents and antiviral agents, vascular lesions ofthe brain and spinal cord, and other diseases or conditions prone to result in neural degeneration.

[ 0137] It is also within the confines ofthe present invention that the method described herein may be used to treat neural degeneration that is associated with a demyelinating condition. Examples of demyelinating conditions include, without limitation, acute disseminated encephalomyelitis (ADEM), acute transverse myelitis, acute viral encephalitis, adrenoleukodystrophy (ALD), adrenomyeloneuropathy, ADDS-vacuolar myelopathy, HTLV- associated myelopathy, Leber's hereditary optic atrophy, multiple sclerosis (MS), progressive multifocal leukoencephalopathy (PML), subacute sclerosing panencephalitis, and tropical spastic paraparesis.

[ 0138] In the treatment of neural degeneration in a subject, CD81 or a CD81 derivative may be contacted with neurons in vivo, within a subject who has neural degeneration, by administering, to the subject CD81 or a CD81 derivative, in accordance with known methods of protein administration. The amount of CD81 or CD81 -derivative protein to be used is an amount effective to enhance survival of neurons, as defined above, and may be readily determined by the skilled artisan.

[ 0139] To effect contacting in vivo in a subject, CD 81 -derivative proteins (e.g., the ECD of CD81, or the mimetics, GMl 109 and GM1416) may be more appropriate for administration to the subject, as these smaller proteins can more readily cross the blood-brain barrier. Moreover, when target neurons are localized to a particular portion ofthe body of a subject, it may be desirable to introduce the CD81 or CD81 -derivative protein directly to the tissue by injection or by some other means (e.g., by introducing CD81 or CD81 -derivative protein into the blood or another body fluid).

[ 0140] The CDSl or CD81 derivative may be administered to the subject to treat neural degeneration by any ofthe modes of administration, and in any ofthe formulations, described above. For example, the CD81 or CDSl derivative may be administered to the subject by oral administration, parenteral administration, sublingual administration, transdermal administration, or by osmotic pump. In one embodiment ofthe present invention, the CD81 or CD81 derivative is administered to the subject by oral administration or transdermal administration. [ 0141] As disclosed herein, the novel transmembrane tyrosine kinase NrSl, which is located at the neuronal surface, is rapidly phosphorylated in response to CD81 signaling. This rapid phosphorylation ultimately results in forward signaling to the astrocyte. Additionally, the binding ofthe CD81, or a mimetic thereof, to NrSl on the neuronal surface drives cleavage of neuregulin (Nrg) and the translocation ofthe intracellular domain of neuregulin (ICD_Nrg) to the neuronal nucleus, thereby promoting neuronal survival. Thus, there is also reverse signaling to the neuron. The nucleotide sequence of a NrSl gene (previously identified as EHK1) is provided in FIG. 27; a previously unidentified exon of NrSl that is 5' to the published EHK1 sequence is provided in FIG. 29. The corresponding amino acid sequence is provided in FIG. 28. The skilled artisan would understand that the novel NrS 1 exon would be useful, e.g., as a probe for NrS 1. [ 0142] Accordingly, the present invention is further directed to methods for treating neural degeneration in a subject in need of treatment therefor. The methods comprise activating NrSl in the subject. The subject may be any animal, but is preferably a mammal (e.g., humans, domestic animals, and commercial animals). More preferably, the subject is a human. NrSl in a subject may be activated by targeting NrSl directly. NrSl in a subject also may be activated indirectly, by targeting an enzyme or other endogenous molecule that regulates or modulates the functions or levels of NrSl in the subject. [ 0143] Unless otherwise indicated, "NrSl" includes both an NrSl protein as characterized herein (having the amino acid sequence provided in FIG. 28), and an "NrSl analogue". An "NrSl analogue" is a functional variant of the NrSl protein, having NrS 1-protein biological activity, that has 60%) or greater (preferably, 70% or greater; more preferably 80%> or greater, even more preferably 90% or greater; even more preferably 95% or greater; and most preferably 97% or greater) amino-acid-sequence homology with the NrSl protein, as well as a fragment ofthe NrSl protein having NrS 1 -protein biological activity. The skilled artisan could identify such analogs based on functional studies with similar tyrosine kinases. As further used herein, the term "NrSl- protein biological activity" refers to protein activity which modulates and enhances survival of neurons, as disclosed herein. NrSl may be produced synthetically or recombinantly, or may be isolated from native cells; however, it is preferably produced synthetically, using conventional techniques and cDNA encoding NrSl .

[ 0144] As defined above, the term "neural degeneration" means a condition of deterioration ofthe neuron, wherein the neuron changes to a lower or less functionally-active form. Neural degenerations which may be treated by the method ofthe present invention include any of those described above. For example, the neural degeneration may be a primary neurologic condition (e.g., a neurodegenerative disease), as well as an acquired secondary effect of non- neural dysfunction (e.g., neural loss secondary to degenerative, pathologic, or traumatic events). Examples of primary neurologic conditions, neurodegenerative diseases, and acquired secondary effects of non-neural dysfunction include all of those described above. It is also within the confines ofthe present invention that the method described herein may be used to treat neural degeneration that is associated with a demyelinating condition, including all ofthe above- described demyelinating conditions.

[ 0145] Additionally, as used herein, the term "activating NrSl" means stimulating or inducing the functions of NrSl in the subject, particularly the cleavage of Nrg and the translocation ofthe ICD_Nrg to the neuronal nucleus as a result ofthe binding of CD81, or a derivative or mimetic thereof, to the NrSl receptor at the neuronal surface, thereby promoting neuronal survival. In the method ofthe present invention, NrSl in a subject may be activated, for example, by administering to the subject CD81 protein or a CD81 -derivative protein. Examples of suitable CD81 derivatives include the large ECD of CDSl and the mimetics, GMl 109 and GM1416. [ 0146] The CDSl or CD81 derivative ofthe present invention may be administered to a subject in need of treatment for neural degeneration in an amount which is effective to treat the neural degeneration in the subject. As used herein, the phrase "effective to treat the neural degeneration" means effective to ameliorate or minimize the clinical impairment or symptoms of the neural degeneration. For example, where the neural degeneration is Alzheimer's disease, a neurodegenerative disease, the clinical impairment or symptoms ofthe neural degeneration may be ameliorated or minimized by reducing the amount of neurons lost within the cerebral cortex, hippocampus, subcortical structures, locus caeruleus, and nucleus raphae dorsalis, thereby minimizing or attenuating the progressive loss of cognitive function and the development of senile plaques and neurofibriUary tangles.

[ 0147] The amount of CDSl or CD81 derivative effective to treat neural degeneration in a subject in need of treatment will vary depending upon the particular factors of each case, including the type of neural degeneration, the stage ofthe neural degeneration, the subject's weight, the severity ofthe subject's condition, and the method of administration. This amount can be readily determined by the skilled artisan.

[ 0148] According to these methods, the CD81 protein or CDSl-protein derivative may be administered to a human or animal subject by known procedures, including all ofthe modes of administration described above, and in all ofthe formulations described above. In a preferred embodiment ofthe present invention, the CD81 or CD81 derivative is administered by oral or transdermal administration. The CD81 and CDSl derivative ofthe present invention also may be administered to a subject in accordance with any ofthe above-described methods for effecting in vivo contact between neurons and CD81.

[ 0149] As disclosed herein, the inventor has discovered that full astrocyte differentiation depends upon CD81-NrSl signaling. In view of this discovery, it is expected that NrSl will provide a means for regulating astrocyte growth and proliferation. Accordingly, the present invention further provides methods for inhibiting proliferation of astrocytes. The methods comprise contacting the astrocytes with an amount of NrSl effective to inhibit proliferation ofthe astrocytes. Unless otherwise indicated, "NrSl" includes both an NrSl protein and an NrSl analogue, as defined above.

[ 0150] These methods may be used to inhibit proliferation of astrocytes in vitro, or in vivo in a subject. As used herein, the term "inhibit proliferation of astrocytes", as defined above, means inhibit cell division and growth of astrocytes, and includes limiting the proliferative rate of astrocytes, as disclosed herein. Inhibition ofthe growth and proliferation of astrocytes may be detected by known procedures, including any ofthe methods, molecular procedures, and assays disclosed herein.

[ 0151] In accordance these methods, NrSl may be contacted with astrocytes in vitro, or in vivo in a subject, by introducing to the astrocytes the NrSl protein itself, or by introducing to the astrocytes a nucleic acid encoding NrSl in a manner permitting expression of NrSl protein. The subject may be any animal, but is preferably a mammal (e.g., humans, domestic animals, and commercial animals). More preferably, the subject is a human. The astrocytes may be contained in neural tissue and other tissue ofthe nervous system, either alone or with other types of neural cells, including, without limitation, neurons and oligodendroglia. Astrocytes may be detected in tissue by standard detection methods readily determined from the known art, including, without limitation, immunological techniques (e.g., immunohisto chemical staining), fluorescence imaging techniques, and microscopic techniques.

[ 0152] NrS 1 may be contacted with astrocytes in vitro by adding NrS 1 protein directly to the culture medium. Alternatively, NrSl may be contacted with astrocytes in vivo in a subject by introducing into, or administering to, the subject the NrSl protein. The NrSl protein ofthe present invention may be contacted with astrocytes, either in vitro or in vivo in a subject, by known techniques used for the introduction and administration of proteins and other drugs, including, for example, injection and transfusion. When target astrocytes are localized to a particular portion ofthe body ofthe subject, it may be desirable to introduce the NrSl protein directly to the tissue by injection or by some other means (e.g., by introducing NrSl protein into the blood or another body fluid). The amount of NrSl protein to be used is an amount effective to inhibit proliferation ofthe astrocytes, as defined above, and may be readily determined by the skilled artisan.

[ 0153] It may be desirable to inhibit the proliferation of astrocytes in vitro when preparing nervous tissue for transplantation, diagnostics, drug screening, and the like. In respect of in vivo treatment, the ability of NrSl to inhibit the proliferation of astrocytes renders NrSl particularly useful for treating conditions associated with defects in astrocyte proliferation in a subject. As defined above, the term "defect in astrocyte proliferation" includes pathologic proliferation of astrocytes in a particular tissue, as compared with normal proliferation in the same type of tissue. It is believed that, by inhibiting the proliferation of astrocytes, NrSl will be useful for the treatment of conditions associated with defects in astrocyte proliferation. It is further believed that NrSl would be effective either alone or in combination with therapeutic agents, such as chemotherapeutic agents or antiviral agents, which are typically used in the treatment of conditions associated with defects in astrocyte proliferation.

[ 0154] Accordingly, the present invention provides methods for treating a condition associated with a defect in astrocyte proliferation in a subject in need of treatment. The methods comprise contacting astrocytes in the subject with an amount of NrSl effective to inhibit proliferation of astrocytes, thereby treating the condition associated with a defect in astrocyte proliferation. As described above, the subject may be any animal, but is preferably a mammal

(e.g., humans, domestic animals, and commercial animals). More preferably, the subject is a human.

[ 0155] In the method ofthe present invention, the condition associated with a defect in astrocyte proliferation may be any of those described above, including, without limitation, astrocytosis, glial scars, hyperplasia, neoplasia, and neuritic plaques (particularly those commonly found in Alzheimer's disease patients). Such conditions associated with a defect in astrocyte proliferation may be caused by a variety of factors, including any of those described above. In one embodiment ofthe present invention, the defect in astrocyte proliferation is astrocytosis. In another embodiment ofthe present invention, the defect in astrocyte proliferation is a neuritic plaque.

[ 0156] In the treatment of a condition associated with a defect in astrocyte proliferation in a subject, NrSl may be contacted with astrocytes in vivo, within a subject who has a condition associated with a defect in astrocyte proliferation, by administering NrSl to the subject, in accordance with known methods of protein administration. The amount of NrSlprotein to be used is an amount effective to inhibit proliferation of astrocytes, as defined above, and may be readily determined by the skilled artisan.

[ 0157] The NrSl may be administered to the subject by any ofthe modes of administration, and in any ofthe formulations, described above. For example, the NrSl may be administered to the subject by oral administration, parenteral administration, sublingual administration, transdermal administration, or osmotic pump. In one embodiment ofthe present invention, the NrSl is administered to the subject by oral administration or transdermal administration.

[ 0158] The present invention is also directed to methods for treating a condition associated with a defect in astrocyte proliferation in a subject in need of treatment. These methods comprise activating CD81 in the subject. Unless otherwise indicated, CD81 includes both a CD81 protein and a CD81 analogue, as defined above. As used herein, the term "activating CD81" means stimulating or inducing the functions of CD81 in the subject, particularly the modulation or regulation of astrocyte growth and the induction of astrocyte growth arrest. CD81 in a subject may be activated by targeting CD81 directly. CD81 in a subject also may be activated indirectly, by targeting an enzyme or other endogenous molecule that regulates or modulates the functions or levels of CDS 1 in the subject. The subject may be any animal, but is preferably a mammal (e.g., humans, domestic animals, and commercial animals). More preferably, the subject is a human.

[ 0159] In the method ofthe present invention, CDSl in a subject may be activated, for example, by administering NrSl protein or NrSl derivative to the subject. The NrSl protein or NrSl derivative ofthe present invention is administered to a subject in need of treatment for a condition associated with a defect in astrocyte proliferation in an amount which is effective to treat the condition associated with a defect in astrocyte proliferation in the subject. The condition associated with a defect in astrocyte proliferation may be any of those described above, include, without limitation, astrocytosis, glial scars, hyperplasia, neoplasia, and neuritic plaques (particularly those commonly found in Alzheimer's disease patients). Additionally, the condition associated with a defect in astrocyte proliferation may be caused by a variety of factors, including any of those described above. In one embodiment ofthe present invention, the defect in astrocyte proliferation is astrocytosis. In another embodiment of he present invention, the defect in astrocyte proliferation is a neuritic plaque.

[ 0160] As used herein, the phrase "effective to treat the condition associated with a defect in astrocyte proliferation" means effective to ameliorate or minimize the clinical impairment or symptoms ofthe condition associated with a defect in astrocyte proliferation. For example, where the condition associated with a defect in astrocyte proliferation is astrocytosis, the clinical impairment or symptoms ofthe condition may be ameliorated or minimized by reducing the mass of astrocytes produced by the astrocytosis, thereby minimizing any potential obstruction of axons which may occur.

[ 0161] The amount of NrSl protein or NrSl derivative effective to treat a condition associated with a defect in astrocyte proliferation in a subject in need of treatment will vary depending upon the particular factors of each case, including the type of defect in astrocyte proliferation, the stage ofthe defect in astrocyte proliferation, the subject's weight, the severity of the subject's condition, and the method of administration. This amount can be readily determined by the skilled artisan.

[ 0162] According to the method ofthe present invention, the NrSl protein or NrSl derivative may be administered to a human or animal subject by known procedures, including all ofthe modes of administration described above, and in all formulations described above. In a preferred embodiment ofthe present invention, the NrSl protein or NrSl derivative is administered by oral or transdermal administration. The NrSl protein or NrSl derivative ofthe present invention also may be administered to a subject in accordance with any ofthe above- described methods for effecting in vivo contact between astrocytes and a protein. [ 0163] The present invention is also directed to pharmaceutical compositions comprising

NrSl and a pharmaceutically-acceptable carrier, wherein NrSl is present in an amount sufficient or effective to treat neural degeneration, as defined above, in a subject to whom said pharmaceutical composition is administered. Such pharmaceutical compositions are useful for administering NrSl to a subject in need of treatment for neural degeneration, in order to treat the neural degeneration in the subject. The NrSl is provided to the subject in an amount that is effective to treat the neural degeneration, as defined above, in the subject. This amount may be readily determined by the skilled artisan. The pharmaceutical compositions may be administered to a subject in accordance with any ofthe methods of administration, and in any ofthe formulations, described above.

[ 0164] Formulations ofthe pharmaceutical composition ofthe present invention may be conveniently presented in unit dosage, and may be presented in oral dosage form (e.g., NrSl and a pharmaceutically-acceptable carrier may be combined in an ampule, capsule, pill, powder, or tablet) or in a form suitable for injection. The pharmaceutically-acceptable carrier may be a solid, liquid, or gel. Furthermore, the pharmaceutically-acceptable carrier ofthe present invention must be "acceptable" in the sense of being compatible with the other ingredients ofthe composition, and not deleterious to the recipient thereof. Examples of acceptable pharmaceutical carriers include all of those described above. The carrier selected will depend upon the route of administration, and the form in which NrSl is introduced. In addition, the formulations ofthe present invention may be prepared by methods well known in the pharmaceutical art, including all ofthe preparation methods described above.

[ 0165] The present invention is also directed to pharmaceutical compositions comprising nucleic acid encoding NrSl and a pharmaceutically-acceptable carrier, wherein the nucleic acid expresses NrSl in an amount sufficient or effective to treat neural degeneration, as defined above, in a subject to whom said pharmaceutical composition is administered. Such a pharmaceutical composition would be useful for administering NrSl to a subject in need of treatment for neural degeneration, in order to treat the neural degeneration in the subject. An amount ofthe nucleic acid is provided to the subject such that it expresses NrSl protein in an amount that is effective to treat the neural degeneration, as defined above, in the subject. These amounts may be readily determined by the skilled artisan. Additionally, the pharmaceutical compositions may be administered to a subject in accordance with all ofthe above-described methods for administration and introduction of nucleic acids.

[ 0166] Formulations of these pharmaceutical compositions may be conveniently presented in unit dosage, and maybe presented in a form suitable for administration of nucleic acid (e.g., by injection). The nucleic acid encoding NrSl may be presented in any form well known in the art ■ for the introduction of nucleic acids, including, without limitation, naked DNA, plasmid DNA, and vector DNA (including viral vectors, as described above), and may be prepared in accordance with methods well known in the arts of gene therapy and molecular genetics. In addition, the pharmaceutically-acceptable carrier ofthe present invention must be "acceptable" in the sense of being compatible with the other ingredients ofthe composition, and not deleterious to the recipient thereof. Examples of acceptable pharmaceutical carriers include all of those described above. The carrier selected will depend upon the route of administration and the form in which nucleic acid encoding NrSl is introduced. In addition, the formulations ofthe pharmaceutical composition ofthe present invention may be prepared by methods well known in the pharmaceutical art, including all ofthe preparation methods described above. [ 0167] Some preferred embodiments ofthe invention are described in the following

Examples. Other embodiments within the scope ofthe claims herein will be apparent to one skilled in the art from consideration ofthe specification or practice ofthe invention as disclosed herein. It is intended that the specification, together with the examples, be considered exemplary only, with the scope and spirit ofthe invention being indicated by the claims which follow the Examples.

Example 1. Molecular characteristics of neuron-astrocyte interactions. Introduction [ 0168] The establishment and maintenance ofthe appropriate number and type of constituent cells in the central nervous system (CNS) ofthe mammal is a daunting problem. Not only do the correct numbers of cells end up in the correct locations during early development; in the absence of trauma or disease, the total numbers and the ratios of cell types remain relatively constant during the longest developmental phase - adulthood - when homeostasis is established. This observation implies much about homeostasis in the CNS. For example, unlike the vast majority of cells in the body, neurons are incapable of dividing and are not replaced over a lifetime. This indicates that there are unique mechanisms which provide continuous support for these cells. In the case of astrocytes, however, the situation is more complex, as these cells are able to re-enter the cell cycle at virtually any point in their history, and do so in response to trauma and disease (15). In spite of this proliferative ability, the number of astrocytes remains largely unchanged throughout life (23, 24, 25). Moreover, it has been demonstrated that astrocytes can be maintained out ofthe cell cycle while they are in direct contact with the neuronal surface (11, 12, 29).

[ 0169] It is important to determine how mitotic quiescence is established and maintained in astrocytes, as there are major sequelae, both positive and negative, that result from astrocyte proliferation in the adult mammal. These sequelae include gliotic responses to injury, the secondary blocking of axonal migration and neuronal remodeling, collateral damage to otherwise intact tissue resulting from astrocytotic spillover, and possible spontaneous mutations and/or viral integration in astrocytes transiting the cell cycle. All of these can lead eventually to astrocytoma. [ 0170] In the case of CNS injury, resulting astrocytosis is thought to be a major contributor to the formation of a glial scar, which in turn may play an important role in blocking regenerating axons (6). While the isolation of damaged tissue is likely to be an important aspect in re-establishing the blood-brain barrier, this behavior implies that such an isolated area is permanently removed from the neural tissue available for regeneration. Further, because astrocytes do proliferate throughout life, albeit at very low levels (7), they are vulnerable to errors in DNA replication, as well as viral integration. This makes these cells susceptible to transformation over the lifetime ofthe mammal. Indeed, in those people who have been diagnosed with a brain tumor, the majority of tumors will be of astrocyte lineage. In view ofthe foregoing, it is clear that an understanding ofthe basic biology of neuron-glia interaction may provide insight into the means by which astrocyte growth control is both initially achieved and subsequently maintained throughout life. Such knowledge likely will yield further insight into methods for re-establishing ordered growth in transformed cells.

[ 0171] Earlier work has demonstrated that astrocytes become growth-arrested upon contact with the neuronal cell surface, and that this activity is neuron-specific in the CNS (28). It also has been shown that the antiproliferative activity of neurons can be isolated from the ability of neurons to drive astrocyte-process outgrowth, by treating astrocytes with either fixed neurons or a neuronal-membrane-enriched protein fraction. In the presence of these preparations, astrocytes exit the cell cycle, but fail to extend astrocytic processes (28) or to upregulate the glutamate transporter, GLT-1 (unpublished observation).

[ 0172] While the cell biology of neuron-astrocyte interactions has been amply described, the molecular correlates of this cell biology have been largely underexplored. In order to begin to define the nature of these interactions, the inventor has undertaken a series of differential gene screens to compare expression patterns in purified astrocyte cultures with those in astrocytes that are co-cultured with neurons. To avoid the problem of contaminating astrocyte RNA with RNA from the "effector" neurons, the inventor took advantage of his earlier observation that neuronal cell membranes are sufficient to drive astrocytes into quiescence (29). [ 0173] From that observation, the inventor identified a number of genes in which expression was upregulated by the neuron-stimulated astrocyte. The majority of genes that were identified in this assay system were known, and had well-documented expression patterns in non- neural tissues. Some ofthe genes in the screen were known to be expressed in the nervous system, but little was known of their biological significance. Among this latter class was a tetraspanin, CDSl, also known as Target of Anti-Pro liferative Antibody or TAP A. The expression of CDSl had been shown in astrocytes, and was known to be upregulated following neural trauma (8). However, the function of CDSl in trauma, and in homeostasis, was previously undefined.

[ 0174] Using a combination of antibody perturbation, biochemical competition, and gene- knockout studies, the inventor has shown that CD81 is a critical modulator of astrocyte growth control. See, e.g., FIG. 21. This observation is crucial, as astrocytosis results from numerous neural traumas, and the resulting glial scar is believed to present a major barrier to productive neural regeneration. In addition, astrocytomas are the predominant single form of brain cancer, with a prevalence on the order of 17,000 cases per year. All astrocytoma cells tested herein failed to express CDSl message or protein, raising the possibility that CD81 plays an important role in astro cyte-tumor formation and/or metastasis.

[ 0175] In addition, the inventor describes herein the identification ofthe neuronal ligand for CD81, which has been named NrSl. The inventor also discloses a series of events that ensue from stimulation of neurons with a small mimetic of CDSl, which has been named GMl 109.

Materials and Methods

[ 0176] Animals. Pregnant Sprague-Dawley rats and C57BL/6 mice were obtained from

Charles Rivers Laboratories. CDSl heterozygous mice were backcrossed greater than 10 generations into the C57BL/6 background. The generation of these mice has been previously described (17). Heterozygous animals were crossed, and offspring were born with the expected

Mendelian frequency. However, an attenuated postnatal viability was observed in the CD81-/- animals. The genotypes ofthe progeny of these crosses was determined exactly as described (17).

Notably, in earlier backcrosses, there was normal Mendelian distribution and normal survival of embryos, but death common in the perinatal period in the homozygous null animals.

[ 0177] Tissue Culture - Primary Neural Cells. Primary cerebellar neurons and astrocytes were prepared as described (18). In brief, cerebella were dissected from rat or mouse pups at postnatal day 4 or 5, the meninges were stripped, and the remaining tissue was washed in Ca 2+ / Mg 2+ free PBS (CMF-PBS). The tissue then was trypsinized, triturated through decreasing caliber needles in the presence of DNase, and pelleted. The cells were resuspended in CMF-PBS, and the single cell suspension was overlaid and separated on a Percoll step gradient (30/60%>;

Amersham Pharmacia), all as described (18). Following extensive washing to remove residual

Percoll, the neuron- and astrocyte-enriched fractions were further enriched: differential adhesion removed contaminating astrocytes from the neuron preparation, and treatment with anti-Thyl and complement-mediated cytolysis removed neurons and fibroblasts from the astrocyte-enriched fraction. All cells were cultured in D , consisting of DMEM (Gibco) supplemented with 10%> heat-inactivated fetal bovine serum (FBS; Gemini-Bio-Products, Inc.), 10%) heat-inactivated horse serum (Gemini-Bio-Products, Inc.), 1% non-essential amino acids (Gibco), penicillin-streptomycin

(Gibco; 20 U/ml), Fungizone (Gibco; 0.25 μg/ml), and glucose at 0.6%> final concentration. The astrocyte cell suspensions were seeded at 2 x 10 cells/well in 24-well plates (Costar) or 5 x 10 cells/well in 8-well Lab-Tek tissue culture chambers (Nalge Nunc), which had been treated with 50 μg/ml poly-L-lysine (Sigma). Effector neurons were added at a ratio of 2 neurons per astrocyte.

[ 0178] Astrocytoma cell lines. Several astroglial cell lines (rat C6 and 9L, human A172 and U251MG, and mouse LN308 and LN18) were grown in 100-mm tissue culture dishes (Falcon Labware) in D .

[ 0179] Antibodies. Rabbit anti-cow glial fibrillary acidic protein (GFAP) antibodies, as well as TRITC-conjugated swine anti-rabbit antibodies, were obtained from DAKO A/S (Copenhagen, Denmark). Mouse monoclonal antibodies (mAbs) to bromodeoxyuridine (BrdU) and conjugated to FITC were obtained from Boehringer (Mannheim, Germany). Hamster mAb 2F7 against CD 81 (1) and FITC-conjugated mouse anti-hamster mAbs were obtained from Pharmingen (San Diego, CA). Hamster mAbs Eatl and Eat2 react with distinct epitopes in CDSl, and have been recently described (18). TuJl recognizes a neuron-specific βlll subunit of tubulin. Alexa red conjugated goat anti-mouse secondary antibody was purchased from Molecular Probes, and mouse mAb anti-GST was purchased from Sigma. Biotinylated goat anti- mouse antibody and the Nectastain ABC kit were purchased from Vector Labs. [ 0180] Fusion Proteins. Full-length SCIP and the region ofthe mouse CD81 encoding the large extracellular loop (LEL) were both cloned into pGEX expression vectors (5), to generate GST fusion proteins. Clones were sequenced, and DH5o E. coli were transformed with the respective clones and IPTG induced. The resulting lysate was enriched on glutathione-agarose beads, and the protein concentration was assessed by BCA assay (Pierce). The integrity ofthe material was determined by gel electrophoresis and immunoblotting with antigen-specific antibodies and/or anti-GST antibodies, all by standard techniques. [ 0181] Northern Blot Analysis. Total RNA was extracted from cultured cells, as described by Chomczynski and Sacchi (2). 20 μg of RNA from each sample were electrophoretically separated on denaturing agarose gels, and transferred to nylon membranes (Micron Separations Inc.). The membranes were probed overnight at 42°C with random primed, [ PJdCTP-labeled mouse CD81 cDNA, washed sequentially (three times for 15 min at 65°C in 2x SSC, 0.1% SDS; twice for 10 min at 65°C in 0.2x SSC, 0.1% SDS; and once for 5 min at room temperature in 2 x SSC), dried, and exposed to X-ray film.

[ 0182] Immunoblotting. Cultured cerebellar astrocytes and astrocytes co-cultured with neurons C6 astroglioma were washed twice in ice-cold PBS, scraped from the dishes, pelleted, and resuspended in hypotonic disruption buffer (10 mM HEPES (pH 7.9), 10 mM NaCl, 0.1 mM EGTA, 0.1 mM EDTA, 0.5 mM phenylmethylsulfonylfluoride, 0.5 mg/ml leupeptin, 0.7 mg/ml pepstatin A, and 1 mg/ml aprotinin). The samples were incubated on ice for 15 min, after which NP-40 was added to a final concentration of l%o. The detergent-soluble and -insoluble portions were separated by centrifugation. Protein concentrations were determined for the membrane- containing detergent-soluble fraction using a BCA assay (Pierce). 50 μg of protein were separated on a 10%> SDS-polyacrylamide gel, and the proteins then were transferred onto nitrocellulose using a semidry blotter. The efficiency of transfer was determined by amido black staining. The membrane was blocked in Buffer A, containing 5%o milk and 1% Triton-X 100 in Tris buffered saline. The membrane was then probed with goat anti-CD81 antiserum, followed by peroxidase-conjugated donkey anti-goat secondary antibody. The reaction product was visualized by ECL.

[ 0183] Immunofluorescence. Forty-eight hours after plating, cell cultures were washed in

PBS and fixed in 4%> paraformaldehyde in PBS at 4°C for 30 min. Nonspecific binding was blocked by incubation in 10%> FBS/PBS for 30 min at room temperature. The blocking solution was removed, and primary antibodies (2F7, Eatl, or Eat2 diluted in PBS) were added at 37 C for 1 h. The cultures were rinsed, incubated in mouse FITC-conjugated anti-hamster antibodies at room temperature for 30 min, washed, and mounted (Pro Long Antifade Kit, Molecular Probes). The same types of cultures were also double-labeled for GFAP and BrdU incorporation with FITC-conjugated anti-BrdU antibodies (Sigma) (see below). The cultures were washed in PBS, and fixed in 4% paraformaldehyde/PBS at 4°C for 30 min, washed in PBS, and permeabilized in 0.5%) Triton X-100/PBS at room temperature for 10 min. Nonspecific binding was blocked by incubation in 10% FBS/PBS at room temperature for 30 min, and GFAP was visualized with a TRITC-conjugated secondary antibody.

[ 0184] In vitro proliferation assays. Cultures were established, either under control conditions, or in the presence of fusion protein or increasing concentrations of one ofthe mAbs as described herein. Twenty-four hours later, 10 μM BrdU (Sigma) was added, and the cultures were continued for an additional 24 h. Subsequently, the cells were fixed, and astrocytes were identified by GFAP staining, as described herein. To visualize BrdU incorporation, the chromatin was denatured in 2 M HC1 for 30 min, washed extensively in PBS, and incubated with FITC-anti- BrdU antibodies at room temperature for 1 h. Following incubation, the cells were washed extensively, stained with b/s-benzamide (Sigma) to determine total cell numbers, washed again, and mounted in Anti-Fade. The level of astrocyte proliferation was determined by dividing the number of BrdU-positive astrocytes by the total number of GFAP -positive cells per microscopic area. All proliferation assays were repeated at least three times; 30 microscopic areas were examined from each experimental sample (600-700 cells/sample, for a minimum total of -2000 cells per experimental point). Statistical analysis ofthe data was performed using two-tailed, students' t-test.

Results

[ 0185] CD81 is expressed on the cell surface ofthe astrocyte. The inventor originally identified CDSl expression in astrocytes co -cultured with neuronal membranes by using a differential screening approach. To determine if CD81 protein was indeed expressed by astrocytes, the inventor established cultures of astrocytes or the astrocytoma cell line, C6, as previously described (28). Protein from 48-hour cultures was isolated, separated on an SDS- PAGE gel, and blotted with a polyclonal antibody against CD81. CD81 is constitutively expressed by cultured astrocytes, whereas the C6 glioma cells are CD81 -negative (FIG. 1 A). The addition of neurons to these cultures increased CD81 expression by 50-70% in astrocytes, but had no effect on the C6 cells (some data not shown). To localize CD81 expression on the astrocytes, an anti-CD81 monoclonal antibody (mAb), 2F7 (1), was used to stain rat astrocytes cultured alone, as well as co-cultures of astrocytes and neurons. Figure IB shows the punctate staining pattern of CD81 on the surface of astrocytes. When co-cultured with neurons, astrocytes extend complex processes that serve as guidance pathways for neuronal migration, and matrices for neuronal adhesion and differentiation (11, 29). Staining of neuron-astrocyte co-cultures showed a punctate pattern of CD81 expression, both on the astrocyte cell soma and along the processes (FIG. 1C, arrows). In contrast, neurons in these cultures failed to stain with the 2F7 antibody. [ 0186] Anti-CD81 mAbs recognize unique, non-overlapping, extracellular epitopes.

CDSl has four transmembrane domains, resulting in two loops in the extracellular domain. One is the small extracellular loop (SEL), and the other is the large extracellular loop (LEL). When the 2F7 antibody was used on live, non-permeabilized astrocytes, results indicated that it recognizes an extracellular epitope (FIGS. 1 A and IB). Recent work has shown that the 2F7 mAb recognizes a conformationally-dependent epitope that requires the presence of both the SEL and LEL of CD81. While Eat2 has a higher affinity for CD81 than does 2F7, it also requires both loops for antigen binding. In contrast, Eatl recognizes an epitope within the LEL (18). In an effort to determine if any of these mAbs was able to block neuron-astrocyte interactions, the inventor established co-cultures in the presence of these reagents.

[ 0187] Neuron-astrocyte interactions are blocked in vitro by mAb Eatl. In order to determine the potential efficacy of Eatl and Eat2 in blocking neuron-induced astrocyte differentiation and cell-cycle exit, the inventor established co-cultures of cerebellar granule cells and astrocytes. The cultures were allowed to grow for 48 h, with BrdU added in the last 24 h of culture. As can be seen in FIG. 2A, there is a loss of neuron-induced astrocyte growth arrest, which is dependent on the concentration ofthe Eatl antibody. Notably, the neurons in this culture were viable, and adhered to the astrocytes and extended neurites (see below). Previous work has shown that cerebellar granule cells are exquisitely dependent upon astrocytes and astrocyte-derived factors for survival (13). The inventor's observation that the Eatl mAb blocked neuron-dependent astrocyte proliferative arrest, but not trophic support, suggests that all normal neuron-astrocyte interactions are not lost under these conditions.

[ 0188] Eat2 had no apparent effect on neuron-astrocyte interactions: neuron-astrocyte co- cultures established in the presence of Eat2 were indistinguishable from control co-cultures. Under both control and Eat2 conditions, astrocytes withdrew from the cell cycle, and extended complex processes, when challenged with neurons (FIGS. 2D and 2E). In contrast, the addition of mAb 2F7 enhanced neuron-induced astrocyte proliferative arrest (FIG. 2B), suggesting that the addition of this antibody to the co-culture system augmented neuron-dependent astrocyte growth arrest. Taken with the Eatl data, these observations show that alterations in CD81 bioavailability and/or conformation have a profound effect on modulating astrocytic proliferative responses to neurons.

[ 0189] It has been previously shown that astrocyte proliferation and process formation in response to neuronal contact are separable events. Contact with neuronal membranes is sufficient for astrocyte growth control, but viable neurons are required for both cell-cycle exit and process formation (29, 30). While the inventor saw no differences in astrocytic process outgrowth in neuron-astrocyte co-cultures under control conditions or in the presence of either mAb 2F7 or Eat2, there was a remarkable difference between those cultures and co-cultures established in the presence of Eatl. Examples of these stark differences can be seen in FIGS. 2C and 2D, in which neuron-astrocyte co-cultures were set up in the presence of Eatl and Eat2, respectively. [ 0190] In the presence of Eatl (FIG. 2C), the astrocytes failed to extend typical processes in response to neurons, and remained in the cell-cycle. The astrocytes depicted in FIG. 2C were imaged with anti-GFAP antiserum. These have the appearance of a cluster of daughter cells that arose in situ. In contrast, in the presence of Eat2 (FIG. 2D), the GFAP-expressing astrocytic processes are long, and of a complexity that is indistinguishable from astrocytic responses seen in co-cultures of astrocytes and viable, wild-type granule cell neurons (FIG. 2E). In addition, the integrity of neuritic processes is uncompromised in the presence ofthe anti-CD81 mAbs Eatl and 2F7 (FIGS. 2F, 2G, and 2H). These data demonstrate that the effects ofthe anti-CD81 mAbs occur at the level ofthe astrocyte, and are not the result of a reduction in viability, or attenuation of axonogenic capabilities, of he neuron. Moreover, the data show that the treated astrocytes are able to support neuronal survival and axonal outgrowth, further highlighting the specificity ofthe role of CD81 in one aspect of astrocyte interactions with cognate neurons. [ 0191] Soluble GST-CD81 fusion protein binds to the neuronal cell surface and competes for astrocyte-expressed CD81. The antibody blocking studies were suggestive of a significant role of CD81 in mediating neuron-astrocyte interactions. However, as with any antibody blocking experiments, there is always a concern about steric inhibition. Therefore, to further extend these observations, the inventor used a soluble mouse GST-CD81 large extracellular loop (GST- CDS 1 (LEL)) fusion protein in an effort to compete for neuronal binding to the astrocyte. To determine if the fusion protein was able to bind to neurons, astrocytes, or both, the inventor isolated and purified the respective cell types, as described (28). The viable cells were incubated with 10 μg/ml of GST-CD81(LEL) on ice for 1 h. Thereafter, the cells were fixed, then stained with an anti-GST antibody to avoid staining endogenous CD81. The GST-CD81(LEL) fusion protein adhered to the neuronal fraction, but not the astrocytic fraction (FIG. 3). The few stained cells seen in the astrocyte-enriched fraction were probably neurons, based on the size ofthe cell somata, and the shape ofthe cells. Astrocytes in culture become flat, and bipolar or tripolar, unlike these cells.

[ 0192] CD81 is required for normal neuronal function and neuronal/astrocyte interactions.

In cerebelli of CD81 -/- mice, neuronal atrophy and abnormal histology is evident (FIG. 22). Moreover, CD81 is required for initial astro cyte/neuron interactions (FIG. 24). [ 0193] The observation that the CD81(LEL) protein adhered to the surface ofthe neuron suggests the existence of a CD81 receptor on these cells. This putative receptor may be potentially involved in normal neuron-astrocyte interactions. To determine if blocking such a receptor would block the ability ofthe neuron to bind to CDSl and to differentiate the astrocyte via CD81, the inventor added increasing concentrations of either GST-CDSl(LEL) protein or an irrelevant GST fusion protein, GST-SCIP. The soluble CD81 blocked normal neuron-induced astrocyte proliferative arrest in a dose-dependent manner, while the GST-SCIP had no effect (FIG. 4). Neither fusion protein had any visible effects on neuronal survival or differentiation. Based on the neuronal binding patterns, as well as the blocking, these data suggest that the soluble CD81 is competing for receptors on the neuron, thereby blocking normal neuron-induced, CD81- mediated proliferative arrest (FIG. 4).

[ 0194] Astrocyte cell-cycle withdrawal is CD81 -dependent. In addition to its expression on the astrocyte, CD81 is also expressed by numerous cell types, including lymphocytes. In the immune system, CDSl has been shown to play a vital role, as evidenced by the impaired immunity observed in CD81 -deficient mice (16, 17, 20, 27). Heterozygotic CD81 mice were backcrossed 10 generations onto a C57BL/6 background, to establish the CD81 deletion in a pure C57BL/6 genotype. Mixed neuron-astrocyte cultures were established from animals immediately after birth. These CD81-/- animals were harvested in early postnatal life because they have severely decreased viability beyond the first hours of birth. At the time of harvesting, the additional neural tissue was taken for simultaneous genotyping. The cultures were established, and allowed to grow for 48 h. BrdU was added in the final 24 h of culture. The cells then were fixed and stained, and astrocyte proliferation was determined by BrdU and GFAP double labeling. The proliferation data was tabulated before the genotype ofthe respective cultures was unblinded. The extent of astrocyte proliferation in the wild-type co-culture was set at 1. With respect to this level of proliferation, CD81+/- animals showed a 20%> increase in astrocyte BrdU incorporation, whereas the CD81-/- astrocytes showed a doubling of astrocyte proliferation (FIG. 5).

[ 0195] CD81 is absent in a variety of astrocytic tumor cell lines. Tumorigenesis is a multistep phenomenon which contributes to a loss of growth control. To determine if CD81 might play a role in either astrocytic tumor progression or metastasis, the inventor assayed a number of astrocytoma cell lines for CDSl mRNA expression. Consistent with the immunofluorescence data, astrocytes expressed CD81 mRNA when isolated and cultured as a purified cell population. Upon co-culture with neuronal membranes, astrocyte expression of CD81 was upregulated, suggesting a positive feedback mechanism in maintaining astrocytes out ofthe cell cycle. In contrast, none ofthe astrocytoma cell lines assayed had detectable levels of CD81 message after 3 days of exposure (FIG. 6). Lack of CDSl in astrocytomas can also be demonstrated by immunohisto chemistry (FIG. 23). Furthermore, astrocytoma cells in which CD81 is re-expressed revert to normal appearance and function (FIG. 25). [ 0196] There are likely to be a number of layers of neuron-regulated growth control, as no gross astrocytosis was observed in the CD81-/- animals when they were in the C57BL/6 background. However, when CD81-/- mice are bred into a BALB/c background, there is massive astrocytosis (8, and personal communication). Thus, there may be additional genetic components that interact with CD81 to regulate astrocyte proliferation in vivo. Nevertheless, the present data suggest that CD81 may play a role in astrocyte tumor progression.

[ 0197] NrSl binds CD81. The specificity of neuronal-membrane binding to astrocytes is demonstrated in FIG. 9, which shows the saturable binding of ³⁵S-labeled granule cell neuron protein to mono layers of astrocytes (closed diamonds). In contrast, PC 12 cell membrane protein, prepared in the same manner, failed to demonstrate any significant binding to astrocytes. Neither protein fraction showed any saturable binding to mono layers of 3T3 cells (data not shown). [ 0198] The saturable nature of neuronal-protein binding to astrocytes, as well as the ability of neuronal protein to induce astrocyte growth arrest, led the inventor to devise a protein purification strategy to isolate the active agent. The inventor's strategy included the capture of membrane-associated proteins using Triton-X 114 (FIG. 10A, lane 1) or FPLC chromatography on an anion-exchange column (FIG. 10A, lane 2), followed by affinity chromatography using astrocyte membrane affinity matrix and eluting with either EDTA (FIG. 10 A, lane 3) or NaCl (FIG. lOA, lane 4). The predominant species in the NaCl-eluted fraction was a diffuse band at 70 kD (see below). The material was assayed at every step. A 10,000-fold enrichment in activity was obtained in the salt-eluted fraction. In contrast, there was a lack of antiproliferative activity in the EDTA-eluted fraction. This suggests that the neuron-expressed antve proteins bind independently of divalent cations.

[ 0199] The inventor also developed an assay to determine activity ofthe various protein preparations. In this assay system, proteins of interest can be bound to a clear matrix; in turn, cells can be plated onto the protein-matrix substratum and assayed for a variety of activities, including gene expression. With this assay, the inventor was able to remove bound proteins from the enrichment procedure described above, and assay them for astrocyte proliferation. In brief, following adsorption ofthe test proteins to the matrix, purified mouse astrocytes were added for 48 h, and the number of astrocytes was calculated.

[ 0200] In addition, the inventor assayed for the subcellular distribution ofthe protein, statin - a 57-kD protein that has been shown to translocate from the cytosol to the nucleus in cells that are in G₀ (39, 40, 41). FIG. 10B demonstrates that astrocytes, when plated on the NaCl- eluted fraction, mobilize statin to the nucleus from the cytosol. This activity is quantified in Table 1. After identifying the astrocyte receptor for this activity, the inventor was able to positively identify the neuronal ligand, which has been named NrSl. The cloning and characterization ofthe NrSl receptor on astrocytes (i.e., CD81) and the full identification of NrSl, are described herein.

Table 1. Neurons induce astrocytes into G₀.

Control EDTA Eluate NaCl Eluate

Cells per fields 59 ± 9.6 53.1 ± 8 9.4 ± 2.3

Inhibition of proliferation 0% 10% 100%

Statin^1" nuclei per field 0% 4% 100%

[ 0201] GMl 109 binds to the neuronal cell surface, competing for astrocyte-expressed

CD81. Using antibodies, the inventor previously showed that the disulfide-bonded side-pocket of the large extracellular domain (ECD) of CD81 is required for the transduction of neuron-induced astrocyte growth arrest (34). To further probe this putative binding pocket, the inventor generated GMl 109, a mimetic based on modeling predictions. GMl 109 is a peptide compound designed to fit the neuronal binding site of CD81. The inventor engineered the peptide as a fusion protein with the glutathione-binding domain of glutathione s-transferase. The fusion protein was used for the purpose of affinity purification, and as a means of following peptide binding and function.

[ 0202] The inventor's modeling predicted that GMl 109 would be able to compete with

NrSl for binding to CD81. To test this prediction, the inventor added increasing concentrations of either GMl 109 protein or an irrelevant peptide, GMl 110, to co-cultures of neurons and astrocytes. Consistent with the inventor's model, GMl 109 blocked neuron-induced astrocyte proliferative arrest in a dose-dependent manner, while GMl 110 had no effect (FIG. 11). Neither compound had any observable effect on neuronal survival or differentiation (data not shown), thereby suggesting that the effects are at the level of signaling, not neuronal survival. [ 0203] There are two potential interpretations of these data: (1) GMl 109 interacts with

(an) unknown protein(s) in cis on the astrocyte cell surface, preventing the protein(s) from participating in NrSl signaling through CD81; or (2) GMl 109 binds in trans to NrSl on the surface ofthe neuron, thereby antagonizing signaling. To study these possibilities, the inventor isolated and purified the two cell types, as previously described (27). In brief, cells were acutely isolated from neonatal rodent cerebellum, and separated into neuron and astrocyte fractions by density centrifugation and differential adhesion. Identical numbers of viable neurons or astrocytes were incubated with 10 μg/ml of GMl 109 on ice for 1 h. The cells were fixed, and then stained with an anti-GST antibody to avoid staining endogenous CD81. As shown in FIG. 12, GMl 109 binds exclusively to the neurons, confirming that this molecule acts in trans and binds the neuronal cell surface. [ 0204] GMl 109 binds to a single, neuron-specific protein. As described above, it appeared that at least one neuronal membrane protein is required for the induction of astrocyte growth arrest. In light of this observation, the inventor reasoned that GMl 109 might be used as a hook to capture and identify the neuronal protein of interest, because it is able to bind the neuronal cell surface. In brief, the inventor purified granule cell neurons from early postnatal mouse and rat, and generated detergent-soluble extracts from these cells and from rat Schwann cells and 3T3 fibroblasts. The proteins were separated by SDS-PAGE, transferred to nitrocellulose, and probed with GMl 109, followed by goat anti-GST. Under these stringent conditions, GMl 109 protein interacted specifically with a ~70-kD neuronal protein, which was absent from the control cells (FIG. 13 A). To date, with the exception of central neurons, no other cell type is known to express this protein.

[ 0205] The CD81 ligand is a transmembrane tyrosine kinase. Based on the cell-surface binding of GM1109 (see FIG. 12), the inventor used GM1109 to first bind to, and then pull down, its cognate ligand. Briefly, the inventor bound GMl 109 to purified granule cell neurons on ice for 10 min. These cells then were lysed in the presence of protease inhibitors, and the GMl 109- ligand complexes were purified on glutathione/agarose beads. Following extensive washing, the protein complexes were eluted from the beads, and separated by SDS-PAGE. Control cells (3T3 cells) were run in parallel. The gel then was silver-stained to reveal the isolated proteins (FIG. 14). The only protein associated with GMl 109 was approximately 70 kD, and was not expressed by the control cells.

[ 0206] Using the same procedure, the inventor isolated the p70 band, and subjected it to

Edman degradation and mass-spectrometry analysis. Based on these analyses, GMl 109 apparently binds to a protein tyrosine kinase, referred to herein as NrSl. It is not surprising that, at the structural level, NrSl has a high level of homology with a number of other tyrosine kinases in the enzyme domain. However, the NH₂ terminus, which is modeled as the extracellular domain ofNrSl, is unique.

[ 0207] Many tyrosine kinases are themselves targets of phosphorylation. In order to test if NrSl is itself phosphorylated in response to GMl 109 binding, the inventor conducted a time course of binding of GMl 109 to freshly-isolated human neurons, which were purified from late- term abortus. Following isolation and purification, the cells were mixed with GMl 109 for 0, 2, or 10 min. Thereafter, the cells were lysed, and the GMl 109-NrSl complexes were purified. Following separation by SDS-PAGE, the material was transblotted to nitrocellulose, and probed with an anti-phosphotyrosine antibody. Within 2 min, NrSl was fully tyrosine-phosphorylated; by 10 min, the signal was quenched, suggesting that binding of NrSl to the CD 81 agonist initiates a neuronal-signaling cascade (FIG. 15).

[ 0208] Full astrocyte differentiation depends upon bi-directional signaling. The inventor previously demonstrated that contact with neuronal membranes is necessary for astrocyte growth arrest, but is not sufficient to drive astro cyte-pro cess formation or the upregulation of glutamate transporters (28, and unpublished data). In contrast, viable neurons effect both astrocyte quiescence and full differentiation. The requirement for viable neurons to initiate and drive full astrocytic differentiation suggests that, following CDSl signaling, neuronal responses include changes in gene expression that, in turn, effect the astrocytic responses. The inventor's observation that CDSl induces a rapid phosphorylation of NrSl, and is therefore likely to initiate a signaling cascade in the neuron, is consistent with this hypothesis.

[ 0209] To begin to investigate this system, the inventor established cultures of astrocytes with neuronal membranes, plus or minus dibutyryl-cAMP (dbcAMP), as a mimetic of post-viable neuron signaling events in the astrocyte. This system has been used by others to model astrocytic responses to intact neurons (37, 38, 44). dbcAMP had no effect on the morphology of astrocytes grown in the absence of neurons (FIG. 16A). The inventor previously has demonstrated that dbcAMP has no effect on the astrocyte proliferative index (42). The addition of neuronal membranes to cultured astrocytes arrested astrocyte proliferation, but failed to induce process outgrowth (FIG. 16B). However, the addition of neuronal membranes plus dbcAMP allowed for both growth arrest and process formation (FIG. 16C).

[ 0210] Previous work has shown that astrocyte-derived soluble factors support neuronal survival (33). Here the inventor has begun to define a potential mechanism by which astrocytes signal to neurons, via CD81, thereby setting off a cascade, through phosphorylated NrSl, in which the neuron signals back to the astrocyte.

[ 0211] Binding of a CD81 agonist, GM1109. to NrSl induces neuronal survival and paracrine signaling to astrocytes. It has been appreciated for decades that neurons and glia are interdependent for survival and differentiation; however, the molecular mechanisms underlying these events are only partially understood. One such mechanism that has been partially elucidated is the neuron-induced differentiation of glia by neuregulin (Nrg) (43). Nrg on CNS neurons has been shown to influence astrocyte differentiation (36) by signaling via Nrg cognate receptors on astrocyte erbB2 and erbB4 (31), resulting in increases in astrocyte cAMP levels (32). These data fit the inventor's observations that full astrocyte differentiation is dependent upon both CD81- NrSl signaling and elevation of astrocyte cAMP levels.

[ 0212] To explore whether CDSl -NrSl interactions mediate downstream Nrg signaling in neuron-astrocyte interactions, the inventor purified granule cell neurons and treated the GMl 109, as a mimetic for astrocyte stimulation. Such treatment induced profound and rapid changes in the neuron, as described below, and suggests that GMl 109 might have powerful properties that would be useful in the treatment of neurodegenerative conditions.

[ 0213] Granule cell neurons express a transmembrane form of neuregulin. which is cleaved upon CD 81 signaling. Granule cell neurons were purified to >99%> homogeneity, as described above. The cells were cultured in defined medium, with no added growth factors, and then stained with an antibody that is specific for the intracellular domain of neuregulin (ICD_Nrg) (35). In FIG. 17 A, the staining pattern shows that the ICD_Nrg was associated with the cell membrane in these acutely-dissociated neurons. However, when these cells were treated for as little as 5 min, all ofthe ICD_Nrg was lost from the cell surface. Moreover, the ICD_Nrg migrated to the neuronal nucleus (FIG. 17B), where it is believed to mediate transcriptional changes that allow for neuronal survival and neurite outgrowth (see below). In addition, treatment with GMl 109 induced a 2-fold increase in Nrg concentration in the supernatant of cultured granule cell neurons (FIG. 17D). [ 0214] Taken together, these data suggest that shortly after the astrocyte mimetic,

GMl 109, binds to neurons, there is a cleavage ofthe transmembrane form of Nrg, the extracellular astro cyte-signaling domain is released, and the ICD_Nrg, which has a nuclear localization signal, moves to the neuronal nucleus. Such a biology would explain the mechanism by which neurons, via NrSl, induce astrocyte growth arrest, and then induce full maturation via Nrg signaling. The inventor has termed the reactions in which neurons signal to astrocytes as "forward signaling", and the resulting intracellular neuronal signaling events as "reverse signaling". [ 0215] GMl 109-induced reverse signaling rescues neurons and induces process outgrowth. Neuronal survival has been shown to be dependent upon astrocyte-derived signals (33). However, the precise manner in which astrocytes signal to neurons has not been well documented. In order to determine if the CDSl mimetic, GMl 109, is sufficient to initiate a cascade of neuronal survival and axonogenesis, the inventor cultured purified granule cell neurons at low density (10⁵ cell/well in a 24-well plate), and added either GMl 109 or an irrelevant control peptide. The cells were assessed daily for survival. After 96 h, the cells were fixed and double stained for both neurons and astrocytes. Not unexpectedly, the cells were >99% neurons, with a few, isolated GFAP -positive cells. Neurons cultured in the presence ofthe astro cyte-signaling mimetic, GMl 109, displayed tremendous survival advantages over cells cultured under control conditions, with few pyknotic cells, and numerous process-bearing cells. In contrast, under control conditions, few cells survived, and those that did survive had limited neurite outgrowth. [ 0216] At the conclusion ofthe experiment, the cells were imaged by indirect immunofluorescence. An example is seen in FIG. 19. Cells cultured in the presence of GMl 109 had large, complex neurites (FIGS. 19B and 19C). In contrast, in control cultures, there were few viable cells, and only a few short neurites on those cells that survived (FIG. 19A). Live/dead staining, 24 h after explanation, revealed a greater-than-15-fold increase in viability of neurons that had been cultured with GMl 109, as compared to control-treated cultures. [ 0217] Fifteen amino acids of GMl 109 rescue CNS neurons from cell death. In an effort to further define the domain(s) of GMl 109 that signal(s) to neurons, and induce(s) or enhance(s) neuronal survival, the inventor took advantage of data previously showing that a monoclonal antibody, Eatl, blocks neuron-induced astrocyte growth arrest (34). Epitope-mapping ofthe Eatl antigen defined a region in the large ECD of CD81, thereby suggesting that the function of peptides corresponding to this region might be sufficient. To test this possibility, the inventor synthesized three overlapping 20-mers. The C-terminal 5 amino acids of each peptide overlapped with the N-terminal 5 amino acids of each succeeding peptide. The inventor generated three such peptides, corresponding to a total of 60 amino acids ofthe large ECD of CD81. The peptides were purified by HPLC, yielding >99%> purity. A dose curve of each peptide was added to cultures of purified granule cell neurons, as was GMl 109 (as a positive control). The cells were maintained in Sato's defined medium, that was supplemented with 0.5%> fetal calf serum, for a total of 5 days (120 h). Thereafter, viability was determined with a commercial live/dead stain (Molecular Probes). This assay system is advantageous because esterases in viable cells cleave the calcein dye. The cleavage product then fills the cells, revealing the extent of neurite outgrowth, and permitting the easy identification of neurons.

[ 0218] As shown in FIG. 20 A, in the absence of any treatment, virtually all ofthe granule cell neurons died, while the positive control, GMl 109, rescued the vast majority of cells, in a manner consistent with the data shown in FIG. 19. Ofthe peptides tested, GM1415 had no activity in this assay. In contrast, GM1416 had potent activity in rescuing neurons from cell death, supporting the possibility that GM1416 might be a novel therapeutic for salvaging damaged neurons and enhancing the survival of neurons. Discussion

[ 0219] The establishment ofthe proper ratio of cell types within the mature CNS is not fully understood. While most neuronal populations are unable to re-enter the cell cycle after cellular differentiation, the same does not hold true for astrocytes. These cells are able to proliferate at any point in the life ofthe mammal, and do so under a variety of pathological conditions. However, in homeostasis, the number of astrocytes is remarkably conserved and maintained at a steady state (23, 24, 25). Previous work has shown that neuronal cells are a potent effector of astrocytic proliferative arrest and terminal differentiation. Moreover, these same mechanisms are likely to keep the astrocyte out ofthe cell cycle throughout life. While numerous candidate molecules have been proposed to be mediators of this activity, including NCAM (9), atrial natriuritic protein (21), astrotactin (4, 22), and endothelin 1 (26), none has withstood rigorous analysis.

[ 0220] Much ofthe analysis of neuron-astrocyte interactions has been modeled in vitro.

When challenged with neurons under culture conditions, astrocytes withdraw from the cell cycle, and extend complex, GFAP-rich processes (11, 12, 29, 30). Herein, the inventor reports his recent findings that CD81 is a critical modulator of neuronal-mediated astrocyte differentiation and proliferative arrest. This conclusion is based on three separate, independent lines of experimental evidence. Antibody blocking, antigen competition, and genetic approaches all converge to suggest that CD81 is a critical part of this biology. Moreover, astrocytic tumor cell lines which were tested are CDSl deficient, further suggesting that CD81 is likely to play an important role in normal neuron-astrocyte biology. Importantly, the in vitro findings phenocopy in vivo events in some genetic backgrounds but not others, implying that CD81 either acts as a modifier of, or is modified by, additional genes.

[ 0221] Function-blocking antibodies are valuable tools for testing critical molecular interactions. The inventor has shown herein that Eatl, which binds to a discrete epitope located in the LEL of CD81, is able to ablate astrocytic responsiveness to co-culture with neurons, i.e., the astrocytes remain in the cell cycle, and fail to fully differentiate. In these studies, neurons still were able to adhere to the astrocytic cell surface, where they settled and extended prototypical, complex processes. The survival and differentiation ofthe neuronal cells suggest that there are multiple layers for neuron-astrocyte interactions, and that the astrocytes in these cultures were able to maintain the health ofthe neurons, even without full differentiation ofthe astrocytes. The requirement for astrocytes, or astrocyte-derived support, for granule cell survival and differentiation is well known (13). Therefore, these data suggest that CD81 activity in neuron- astrocyte interactions is specific to neuron-induced astrocyte differentiation. [ 0222] Conformational changes, induced by binding ofthe various anti-CD 81 mAbs to their cognate antigens, result in distinct astrocytic responses to neurons. While Eat2, which binds avidly to CD81, has no effect on function, the Eatl antibody blocks interactions between CD81 and a heretofore unidentified partner. The idea of molecular cross-talk between CD81 and an unknown partner is supported by the observation that 2F7 increases the sensitivity of astrocytes to neuronal antiproliferative signaling, suggesting that conformational changes in CDSl may have profound effects on its activity. This idea is further supported by evidence that shows that the 2F7 mAb is able to block thymocyte maturation (1). There has been at least one other tetraspanin, the Drosophila late bloomer gene, known to have a role in recognition between neural elements in development. Flies mutant in the late bloomer locus fail to make proper neuromuscular synapses in a timely fashion, suggesting a role in recognition of cellular elements in the fly nervous system (14). The tetraspanins, of which CD81 is a member, are thought to be molecular facilitators, bringing together partners within the plane ofthe membrane (19). A definitive answer to the molecular mechanism of CD 81 -mediated signaling between the astrocyte and the neuron awaits the identification of an astrocyte-expressed CDSl binding partner. [ 0223] In general, studies using function-blocking antibodies are inherently limiting, because there are issues of non-specific steric inhibition for which it is difficult to control. The inventor has addressed this potential problem by competing for CD81 binding using soluble, GST- CDS 1 (LEL) fusion proteins. In this assay, the GST-CD81(LEL) fusion proteins bind to the cell surface of neurons, but not astrocytes. The binding of neurons over astrocytes, and the failure of an irrelevant fusion protein to either bind or block function, suggests a specificity of binding, and raises the likelihood of a neuron-expressed CD81 receptor. More importantly, by competing with astrocyte-expressed CD81 for the putative neuronal CDSl receptors, these soluble GST- CDS 1 (LEL) proteins block neuron-induced astrocytic responses. These observations provide direct evidence that CD81 plays an important role in establishing neuron-induced astrocyte activity.

[ 0224] The final confirmation that CD81 plays a vital role in neuron-astrocyte biology was provided by establishing cultures from CD81 heterozygous and homozygous null mice. Using genetics to reduce or ablate CD81 expression, the inventor demonstrated a strict requirement for CD81 in neuron-induced astrocytic responses. The CD81 mice used in these studies were thoroughly backcrossed onto a C57BL/6 background. The +/- mice did not develop spontaneous astrocytic tumors, nor did they show signs of astrocytic hyperplasia, astrocytosis, or any detectable neurological abnormality. However, when the CD81 deletion was backcrossed onto the BALB/c background, a profound astrocytic hyperplasia resulted (8, and personal communication). This is a critical observation, as it demonstrates that there are modifiers of CD81 activity, which depend upon a genetic background to have an observable phenotype. It is notable that the association of a given gene product needs to be considered in the light ofthe surrounding genome, and not en vacuo.

[ 0225] The implications ofthe current study extend beyond questions of regulating astrocyte cell number in homeostasis and injury. The inventor has examined a number of astrocytic tumor cell lines, all of which have severely attenuated levels of CDSl expression. While there is no evidence to suggest that CD81 is a classical tumor-suppressor gene, the absence of CD81 in these astrocytoma cell lines, taken together with CD81 function in normal neuron- astrocyte biology, raises the possibility that CD81 may be part of a tumor-suppressor cascade. The data presented here raise the possibility that mechanisms aimed at re-expressing CDSl in astrocytic tumors in situ may be of significant benefit for patients suffering from astrocytomas. Such an approach would be intended to limit the proliferative rate ofthe tumor cells in situ, thereby changing the otherwise lethal disease to a chronic ailment, and eliminating the neurologic damage of more conventional therapies. The invasive nature of glial tumors, as well as the neurological sequelae of extensive resection, make this type of approach appealing. Further studies, intended to elucidate the transcriptional regulation of CD81 in astrocytes, will provide insights into potential pharmacotherapeutics.

[ 0226] The data presented here clearly show the importance of CD81 in normal, neuron- induced astrocyte proliferative regulation. This observation reveals, to some extent, the mechanism(s) underlying the way in which the ratio of neurons and astrocytes is established and maintained in the adult CNS. Further delineation ofthe molecular mechanisms that control the dynamic interactions between these cell types is critical to the development of a more complete understanding ofthe means by which the mature nervous system achieves and maintains numerical homeostasis, and the way in which this balance may be restored when the nervous system moves out of equilibrium. [ 0227] The inventor also has shown herein that astrocytes express a receptor, CD81, which is absolutely required for neuron-induced growth arrest. The CD81 ligand has been shown herein to be a novel transmembrane tyrosine kinase, NrSl, that is itself rapidly phosphorylated in response to CDSl signaling. Following this rapid phosphorylation, a transmembrane form of Nrg is cleaved (i.e., the extracellular domain of Nrg, which binds erbB2 and erbB4 on the astrocyte), resulting in elevations of intracellular cAMP levels and astrocyte maturation. Accordingly, there is forward signaling to the astrocyte. The binding ofthe astro cyte/CD 81 mimetic, GMl 109, to the neuronal surface drives cleavage of Nrg and the translocation ofthe ICD_Nrg to the neuronal nucleus, thereby promoting neuronal survival. Accordingly, there is also reverse signaling to the neuron. This bi-directional signaling is depicted in FIG. 19.

[ 0228] Taken together, the inventor's observations suggest three distinct therapeutic possibilities: (1) use ofthe CD81 mimetic, GMl 109, or derivatives thereof, in the direct treatment of neural degeneration, including neurodegenerative disease; (2) use of agonists of NrSl signaling in the treatment of conditions associated with a defect in astrocyte proliferation, including astrocytosis; (3) and re-expression of CD81 in astrocytoma cells in order to re-establish growth control.

Example 2. Characterization of NrS 1.

Introduction

[ 0229] Unlike the vast majority of cells in the body, neurons are incapable of dividing or being replaced over a lifetime, suggesting unique mechanism(s) which provide continuous support and cellular repair. In contrast, astrocytes, the cells which provide much ofthe trophic support and provide the physical scaffolding in which the neurons reside, are able to re-enter the cell cycle at virtually any point in their history, and do so in response to trauma and disease (15, 28, 49).

However, despite this proliferative ability, the number of astrocytes remains largely unchanged throughout life (23, 24, 25), thereby posing a conundrum as to the mechanisms(s) that regulate of their growth.

[ 0230] The significance of how this mitotic quiescence is established and maintained is enormous, as there are major sequelae, both positive and negative, that result from astrocyte proliferation in the adult. These include gliotic responses to injury, the secondary block of axonal migration and neuronal remodeling, collateral damage to otherwise intact tissue from astrocytotic spillover, and the possibility of spontaneous mutations and/or viral integration in astrocytes as they transit the cell cycle, all of which can lead to astrocytoma. We have recently described results in which we have discovered the identity ofthe astrocyte receptor required for neuronal signal, termed CD 81. We then turned our attention to the identification ofthe neuronal ligand for CD81, which we have termed NrSl. We have utilized three different approaches to identifying this protein, including protein pull-down followed by mass spectrometry analysis, expression cloning and cDNA cloning. The results from these studies all converged on the identification of a gene previously identified as EHK1, a member ofthe eph class of receptor tyrosine kinases. [ 0231] Eph-related receptors are membrane-bound protein tyrosine kinases that interact with membrane-bound ligands on adjacent cells. Thus, Ephs transduce signals in a manner that is dependent on cell-to-cell contact, and which allows for forward and reverse signaling between the two cell-types after binding. Ehkl was identified in 1993 (46), and since its identification, only two additional papers have appeared in the literature. One of these, by Taylor and colleagues (48), demonstrated expression of Ehkl to be overlapping and virtually identical to the expression pattern we demonstrate for NrSl. Importantly, these authors argue that Ehkl/NrSl expression increases upon neural cell-cell contact, and mediates the aggregation of these cells. This is the basis by which we identified the NrSl/Ehk-1 binding partner, CD81 (34, 45). The only other published manuscript describes the extensive splice variations of Ehkl expressed in the nervous system (47), which we also found, and have expand upon, as described below. Finally, and perhaps most strikingly, the only cell-type other than CNS neurons that have been shown to inhibit astrocyte proliferation are glioma cells (28), which also express Ehkl (48). Taken together, these data support to identification of Ehkl as NrS 1. Below we describe the isolation of Ehk-1 from human brain and from a human fetal brain cDNA library.

Methods and Results

Approaches to identification of NrSl

[ 0232] Protein-protein interactions. We have taken advantage ofthe specific interaction between NrSl and GMl 109 to pull down NrSl -GMl 109 complexes from a whole brain membrane protein preparation. In brief, solubilized membrane preparations from human fetal brain were incubated with GMl 109, the complexes were purified by binding to glutathione/agarose beads, and eluted with free glutathione. The complexes were analyzed by two dimensional gel electrophoresis (iso electric focusing in the first dimension and size fractionation by SDS-PAGE in the second dimension) (FIG. 26). In this analysis a total of 4 spots were consistently obtained which were isolated and analyzed by mass-spectrometry. All the bands had ions that identified Ehk-1. The skilled artisan would understand that apparent molecular mass observed in these experiments could be the result of degradation or expression of splice variants.

[ 0233] Expression cloning. Parallel experiments screened a human fetal brain lambda phage cDNA expression library with GMl 109. A total of 16 clones were isolated by positive binding to GMl 109. The identity ofthe clones was confirmed after excision reactions and sequencing ofthe cDNAs.

[ 0234] cDNA cloning by homology. To avoid the avoid the possibility of false positives that could have arisen by expression screening with the bait, we took advantage ofthe fact that

NrSl is a protein tyrosine kinase and developed an strategy to identify specifically protein tyrosine kinases among the positive clones. We used degenerate oligonucleotide primers corresponding to the amino acid motifs IHRDL and DNWSFG conserved among many tyrosine kinases in regions

VI and IX, respectively, ofthe catalytic domain (50). These primers were used in polymerase chain reactions (PCR) performed on DΝA templates from the 16 positive clones that resulted from the protein library screening. Eight clones generated the expected 200bp PCR fragment upon amplification and they were sequenced to identify the respective tyrosine kinase proteins. A novel exon 5' to the published sequence of EHKl was also identified by these methods (FIG. 29).

[ 0235] The three independent approaches lead to the identification ofthe receptor tyrosine kinase, Ehkl as ΝrSl.

[ 0236] Distribution of ΝrSl Ehkl and protection of retinal ganglion cells and axons in a retinal stroke model. Histological studies established that ΝrSl/Ehkl is widely distributed in cortical, subcortical, retinal, and enteric neurons. (FIGS. 30-32). The presence of ΝrSl in retina allowed the use of an occular stroke model to determine the ability ofthe active CD81 fragment

GM1416 to protect against stroke. Using this model, the inventor determined that GM1416 did protect retinal ganglion when stroke was induced (FIG. 36). Additionally, axons in the inner plexiform layer (IPL) ofthe retina were protected in the retinal stroke model. This further indicates that CDSl or its active fragments are useful in the treatment of stroke.

References

[ 0237] 1. Boismenu et al, A role for CD81 in early T cell development. Science,

271:198-200, 1996.

[ 0238] 2. Chomczynski and Sacchi, Single-step method of RΝA isolation by acid guanidiniumthiocyanate-phenol-chloroform extraction. Analytical Biochemistry, 162:156-59, 1987. [ 0239] 3. Easter et al, Initial tract formation in the mouse brain. Journal of

Neuroscience_^ 13:285-99, 1993. [ 0240] 4. Edmondson et al, Astrotactin: a novel neuronal cell surface antigen that mediates neuron-astroglial interactions in cerebellar micro cultures. Journal of Cell Biology,

106:505-17, 1988.

[ 0241] 5. Flint et al, Characterization of hepatitis C virus E2 glycoprotein interaction with a putative cellular receptor, CD81. Journal of Virology, 73:6235-44, 1999.

[ 0242] 6. Frisen et al, Rapid, widespread, and longlasting induction of nestin contributes to the generation of glial scar tissue after CNS injury. Journal of Cell Biology,

31:453-64, 1995.

[ 0243] 7. Fujita, S., Quantitative analysis of cell proliferation and differentiation in the cortex ofthe postnatal mouse cerebellum. Journal of Cell Biology, 32:277-87, 1967.

[ 0244] 8. Geisert et al, Astrocyte growth, reactivity, and the target ofthe antiproliferative antibody, TAP A. J. Neurosci., 16:5478-87, 1996.

[ 0245] 9. Gimenez y Ribotta et al, Oxysterol (7 beta-hydroxycholesteryl-3-oleate) promotes serotonergic reinnervation in the lesioned rat spinal cord by reducing glial reaction.

Journal of Neuroscience Research, 41:79-95, 1995.

[ 0246] 10. Hatten, M.E., Riding the glial monorail: a common mechanism for glial- guided neuronal migration in different regions ofthe developing mammalian brain. Trends in

Neuroscience, 13:179-84, 1990.

[ 0247] 11. Hatten and Liem, Astroglial cells provide a template for the positioning of developing cerebellar neurons in vitro. Journal of Cell Biology, 90:622-30, 1981.

[ 0248] 12. Hatten et al, Neuron-astroglial interactions in vitro and their implications for repair of CNS injury. Central Nervous System Trauma, 1:15-27, 1984.

[ 0249] 13. Hatten and Shelanski, Mouse cerebellar granule neurons arrest the proliferation of human and rodent astrocytoma cells in vitro. Journal of Neuroscience, 8:1447-53, 1988.

[ 0250] 14. Kopczynski et al, A neural tetraspanin, encoded by late bloomer, that facilitates synapse formation, Science, 271:1867-70, 1996.

[ 0251] 15. Latov et al, Fibrillary astrocytes proliferate in response to brain injury: a study combining immunoperoxidase technique for glial fibrillary acidic protein and radioautography of tritiated thymidine. Developmental Biology, 72:381-84, 1979.

[ 0252] 16. Maecker et al, CDSl on B cells promotes interleukin 4 secretion and antibody production during T helper type 2 immune responses. Proc. Natl Acad. Sci. USA,

95:2458-62, 1998.

[ 0253] 17. Maecker and Levy, Normal lymphocyte development but delayed humoral immune response in CDSl -null mice. J. Exp. Med., 185:1505-10, 1997.

[ 0254] 18. Maecker et al, Differential expression of murine CD81 highlighted by new anti-mouse CD81 monoclonal antibodies. Hybridoma, 19:15-22, 2000.

[ 0255] 19. Maecker et al, The tetraspanin superfamily: molecular facilitators. FASEB

J., 11:428-42, 1997.

[ 0256] 20. Miyazaki et al, Normal development but differentially altered proliferative responses of lymphocytes in mice lacking CD81. EMBO J, 16:4217-25, 1997.

[ 0257] 21. Pedram et al, Astrocyte progression from Gl to S phase ofthe cell cycle depends upon multiple protein interaction. J. Biol. Chem., 273:13966-72, 1998.

[ 0258] 22. Stitt and Hatten, Antibodies that recognize astrotactin block granule neuron binding to astroglia. Neuron, 5:639-49, 1990.

[ 0259] 23. Sturrock, R.R., Histogenesis ofthe anterior limb ofthe anterior commissure ofthe mouse brain. 3. An electron microscopic study of gliogenesis. Journal of

Anatomy, 117:37-53, 1974. [ 0260] 24. Sturrock, R.R., Histogenesis ofthe anterior limb ofthe anterior commissure ofthe mouse brain. III. A quantitative study of changes in the glial population with ag . Journal of Anatomy, 117:17-25, 1974.

[ 0261] 25. Sturrock, R.R., Histogenesis ofthe anterior limb ofthe anterior commissure ofthe mouse brain. II. A quantitative study of pre- and postnatal mitosis. Journal of Anatomy, 117:27-35, 1974.

[ 0262] 25. Teixeira et al, Differential regulation of cyclin Dl and D3 expression in the control of astrocyte proliferation induced by endothelin-1. Journal ofNeurochemistry, 74:1034- 40, 2000.

[ 0263] 26. Tsitsikov et al, Impaired CD19 expression and signaling, enhanced antibody response to type II T independent antigen and reduction of B-1 cells in CD81- deficient mice. Proc. Natl Acad. Sci. USA, 94:10844-49, 1997.

[ 0264] 27. Weinstein, D.E., The Culture of Primary Astrocytes. In Current Protocols in Neuroscience, E. Crawley, C. Gerfen, R. McKay, M. Rogawski, D. Sibley, and P. Skolnick, eds. (New York: Wiley and Son), 1997.

[ 0265] 28. Weinstein et al, C17, a retrovirally immortalized neuronal cell line, inhibits the proliferation of astrocytes and astrocytoma cells by a contact-mediated mechanism. Glia, 3:130-39, 1990.

[ 0266] 29. Weinstein et al, Suppression by antisense mRNA demonstrates a requirement for the glial fibrillary acidic protein in the formation of stable astrocytic processes in response to neurons. Journal of Cell Biology, 112:205-13, 1991.

[ 0267] 30. Beers and Berkow, eds., 77?e Merck Manual of Diagnosis and Therapy,

17^th ed. (Whitehouse Station, NJ: Merck Research Laboratories, 1999) 1395-1398, 1442-48. [ 0268] 31. Francis et al, Neuregulins and ErbB receptors in cultured neonatal astrocytes. J. Neurosci. Res., 57:487-94, 1999.

[ 0269] 32. Gautron et al, Genetic and epigenetic control ofthe Na-G ion channel expression in glia. Glia, 33:230-40, 2001.

[ 0270] 33. Hatten et al , In vitro neurite extension by granule neurons is dependent upon astroglial-derived fibroblast growth factor. Developmental Biology, 125:280-89, 1988. [ 0271] 34. Kelic et al, CD81 regulates neuron-induced astrocyte cell-cycle exit. Mol.

Cell Neurosci., 17:551-60, 2001.

[ 0272] 35. Loeb et al, Expression patterns of transmembrane and released forms of neuregulin during spinal cord and neuromuscular synapse development. Development, 126:781- 791, 1999.

[ 0273] 36. Pinkas-Kramarski et al, Brain neurons and glial cells express Neu differentiation factor/heregulin: a survival factor for astrocytes. Proc. Natl Acad. Sci. U.S.A., 91:9387-91, 1994.

[ 0274] 37. Schlag et al, Regulation ofthe glial Na⁺-dependent glutamate transporters by cyclic AMP analogs and neurons. Mol. Pharmacol, 53:355-69, 1998. [ 0275] 38. Swanson et al. , Neuronal regulation of glutamate transporter subtype expression in astrocytes. Journal of Neuroscience, 17:932-40, 1997. [ 0276] 39. Tsanaclis et al, Statin immuno localization in human brain tumors.

Detection of noncycling cells using a novel marker of cell quiescence. Cancer, 68:786-92, 1991. [ 0277] 40. Wang, A 57,000-mol-wt protein uniquely present in nonproliferating cells and senescent human fibroblasts. J. Cell Biol, 100:545-51, 1985.

[ 0278] 41. Wang, Rapid disappearance of statin, a nonproliferating and senescent cell- specific protein, upon reentering the process of cell cycling. J Cell Biol, 101:1695-1701, 1985. [ 0279] 42. Weinstein, D.E., Molecular Mechanisms of Astro glial Cell Responses to

Neurons (New York: New York University, 1991).

[ 0280] 43. Wu et al. , The POU gene Brn-5 is induced by neuregulin and is restricted to myelinating Schwann cells. Mol. Cell Neurosci., 7:683-95, 2001.

[ 0281] 44. Zelenaia et al, Epidermal growth factor receptor agonists increase expression of glutamate transporter GLT-1 in astrocytes through pathways dependent on phosphatidylinositol 3- kinase and transcription factor NF-kappaB. Mol. Pharmacol, 57:67-78, 2000.

[ 0282] 45. Kelic et al, Early Changes in Astrocyte Gene Expression Following

Neuronal Contact. J Neurosci., 23:1690, 1997.

[ 0283] 46. Maisonpierre et al, Ehk-1 and Ehk-2: two novel members ofthe Eph receptor-like tyrosine kinase family with distinctive structures and neuronal expression. Oncogene,

8:3277-3288, 1993.

[ 0284] 47. Miescher et al, Extensive splice variation and localization of he EHK-1 receptor tyrosine kinase in adult human brain and glial tumors. Brain Res Mol Brain Res, 46: 17-

24, 1997.

[ 0285] 48. Taylor et al, Expression and developmental regulation of Ehk-1, a neuronal Elk-like receptor tyrosine kinase in brain. Neuroscience, 63 : 163-178, 1994.

[ 0286] 49. Tsuchiya et al, Parkinson's disease mimicking senile dementia ofthe

Alzheimer type: a clinicopathological study of four autopsy cases. Neuropathology, 22:77-84,

2002.

[ 0287] 50. Hanks et al, Science, 241 :42-52, 1988.

[ 0288] Other embodiments within the scope ofthe claims herein will be apparent to one skilled in the art from consideration ofthe specification or practice ofthe invention as disclosed herein. It is intended that the specification be considered exemplary only, with the scope and spirit ofthe invention being indicated by the following claims.

[ 0289] In view ofthe above, it will be seen that the several advantages ofthe invention are achieved and other advantages attained.

[ 0290] As various changes could be made in the above methods and compositions without departing from the scope ofthe invention, it is intended that all matter contained in the above description and shown in the accompanying drawings shall be interpreted as illustrative and not in a limiting sense.

[ 0291] All references cited in this specification are hereby incorporated by reference. The discussion ofthe references herein is intended merely to summarize the assertions made by the authors and no admission is made that any reference constitutes prior art. Applicants reserve the right to challenge the accuracy and pertinence ofthe cited references.

Claims

What is claimed is:

1. A method for enhancing survival of neurons, comprising contacting the neurons with CD81 protein, a CD81 derivative or stabilized variants thereof, in an amount effective to enhance survival ofthe neurons.

2. The method of claim 1, wherein the neurons are contacted with a CD81 derivative, the CDSl derivative comprising the amino acid sequence of FIG. 38E.

3. The method of claim 2, wherein the CD81 derivative comprises GMl 109, GM1416 or a stabilized variant thereof.

4. The method of claim 1, wherein the contacting is effected in vitro.

5. The method of claim 1, wherein the contacting is effected in vivo in a mammal.

6. The method of claim 5, wherein the contacting is effected in vivo in the mammal by administering to the mammal the CD81 protein, the CDSl derivative or stabilized variants thereof.

7. The method of claim 6, wherein the CDSl protein or CD81 derivative is administered to the mammal by oral administration, parenteral administration, sublingual administration, transdermal administration, or osmotic pump.

t

8. The method of claim 5, wherein the CD81 or CD81 derivative is administered to the mammal by administering a nucleic acid encoding the CD81 or CDSl derivative, in a manner permitting expression ofthe CD81, CD81 derivative or stabilized variants thereof.

9. The method of claim 5, wherein the mammal is a human.

10. The method of any one of claims 5-9, wherein the mammal has neural degeneration.

11. The method of claim 10, wherein the neural degeneration is a primary neurologic condition or an acquired secondary effect of a non-neural dysfunction.

12. The method of claim 11, wherein the neural degeneration is a primary neurologic condition selected from the group consisting of Alzheimer's disease, amyotrophic lateral sclerosis, Binswanger's disease, Huntington's chorea, multiple sclerosis, myasthenia gravis, Parkinson's disease, and Pick's disease.

13. The method of claim 11, wherein the neural degeneration is an acquired secondary effect of a non-neural dysfunction selected from the group consisting of cerebral palsy, congenital hydro cephalus, muscular dystrophy, stroke, and vascular dementia.

14. The method of claim 10, wherein the neural degeneration is associated with a demyelinating condition.

15. The method of claim 14, wherein the demyelinating condition is selected from the group consisting of acute disseminated encephalomyelitis (ADEM), acute transverse myelitis, acute viral encephalitis, adrenoleuko dystrophy (ALD), adrenomyeloneuropathy, AIDS-vacuolar myelopathy, HTLN-associated myelopathy, Leber's hereditary optic atrophy, multiple sclerosis, progressive multifocal leukoencephalopathy (PML), subacute sclerosing panencephalitis, and tropical spastic paraparesis.

16. A method for treating neural degeneration in a mammal in need of treatment, comprising activating ΝrS 1 in the mammal.

17. The method of claim 16, wherein ΝrS 1 is activated in the mammal by administering to the mammal CD81 protein, a CD81 derivative or a stabilized variant thereof.

18. The method of claim 17, wherein the CD81 protein, CD81 derivative or stabilized variant thereof is administered to the mammal in an amount effective to treat the neural degeneration in the subject.

19. The method of claim 17, wherein the mammal is administered a CD81 derivative, the CD81 derivative comprising the amino acid sequence of FIG. 38E.

20. The method of claim 19, wherein the CD81 derivative comprises GMl 109, GM1416 or a stabilized variant thereof.

21. The method of claim 17, wherein the CD81 protein or CD81 derivative is administered to the subject by oral administration, parenteral administration, sublingual administration, transdermal administration, or osmotic pump.

22. The method of claim 16, wherein the CD81 or CD81 derivative is administered to the mammal by administering a nucleic acid encoding the CD81 or CD81 derivative, in a manner permitting expression ofthe CDSl, CD81 derivative, or stabilized variant thereof.

23. The method of claim 16, wherein the neural degeneration is a primary neurologic condition or an acquired secondary effect of a non-neural dysfunction.

24. The method of claim 23, wherein the neural degeneration is a primary neurologic condition selected from the group consisting of Alzheimer's disease, amyotrophic lateral sclerosis (Lou Gehrig's Disease), Binswanger's disease, Huntington's chorea, multiple sclerosis, myasthenia gravis, Parkinson's disease, and Pick's disease.

25. The method of claim 23, wherein the neural degeneration is an acquired secondary effect of a non-neural dysfunction selected from the group consisting of cerebral palsy, congenital hydro cephalus, muscular dystrophy, stroke, and vascular dementia.

26. The method of claim 16, wherein the neural degeneration is associated with a demyelinating condition.

27. The method of claim 26, wherein the demyelinating condition is selected from the group consisting of acute disseminated encephalomyelitis (ADEM), acute transverse myelitis, acute viral encephalitis, adrenoleukodystrophy (ALD), adrenomyeloneuropathy, AIDS-vacuolar myelopathy, HTLN-associated myelopathy, Leber's hereditary optic atrophy, multiple sclerosis, progressive multifocal leukoencephalopathy (PML), subacute sclerosing panencephalitis, and tropical spastic paraparesis.

28. A method for inhibiting proliferation of astrocytes, comprising contacting the astrocytes with an amount of NrSl protein or NrSl derivative effective to inhibit proliferation of the astrocytes.

29. The method of claim 28, wherein the NrSl protein or NrSl derivative comprises the amino acid sequence of FIG. 28, or a biologically active fragment thereof.

30. The method of claim 28, wherein the contacting is effected in vitro.

31. The method of claim 28, wherein the contacting is effected in vivo in a mammal.

32. The method of claim 31, wherein the contacting is effected in vivo in the mammal by administering an NrSl protein an NrSl derivative or a stabilized variant thereof to the mammal.

33. The method of claim 32, wherein the NrSl protein or NrSl derivative is administered to the mammal by oral administration, parenteral administration, sublingual administration, transdermal administration, or osmotic pump.

34. The method of claim 31, wherein the contacting is effected by administering a nucleic acid encoding the NrSl or NrSl derivative, in a manner permitting expression ofthe NrSl.

35. The method of claim 31 , wherein the mammal is a human.

36. The method of claim 35, wherein the human has a condition associated with a defect in astrocyte proliferation.

37. The method of claim 36, wherein the condition associated with a defect in astrocyte proliferation is astrocytosis.

38. A method for treating a condition associated with a defect in astrocyte proliferation in a subject in need of treatment, the method comprising activating CD81 in the subject.

39. The method of claim 38, wherein CD81 is activated in the subject by administering to the subject NrSl protein, a NrSl derivative, or a stabilized variant thereof.

40. The method of claim 39, wherein the NrSl protein, NrSl derivative or stabilized variant thereof is administered to the subject in an amount effective to treat the condition associated with a defect in astrocyte proliferation in the subject.

41. The method of claim 38, wherein the condition associated with a defect in astrocyte proliferation is astrocytosis.

42. The method of claim 39, wherein the NrS 1 protein or NrSl derivative is administered to the subject by oral administration, parenteral administration, sublingual administration, transdermal administration, or osmotic pump.

43. A method for determining whether a mammal has an astrocytoma, comprising assaying for CDSl expression in a diagnostic sample of cells of astrocytic lineage ofthe mammal, wherein no detection of expression of CD81 in cells of astrocytic lineage ofthe subject is diagnostic of an astrocytoma.

44. The method of claim 43, wherein the mammal is a human.

45. An isolated nucleic acid of less than 10,000 nucleotides, comprising the nucleotide sequence of FIG. 29.

46. The isolated nucleic acid of claim 45, wherein the nucleic acid is less than 5000 nucleotides.

47. The isolated nucleic acid of claim 45, wherein the nucleic acid is less than 1000 nucleotides.

48. The isolated nucleic acid of claim 45, wherein the nucleic acid is less than 200 nucleotides.