WO2002006341A1 - A trophic factor capable of producing a neurosalutary effect in a subject - Google Patents

Abstract

The invention provides compounds, present in Schwann cell conditioned medium compositions and methods useful in treating neurological conditions.

Description

A TROPHIC FACTOR CAPABLE OF PRODUCING A NEUROSALUTARY

EFFECT IN A SUBJECT

Related Applications

This application is a continuation-in-part application of Serial No. 09/616,287, filed July 14, 2000, pending, which is a continuation-in-part application of Serial No. 08/916,642, filed August 22, 1997, abandoned, which is a continuation application of Serial No. 08/296,661, filed August 26, 1994, issued as U.S. Patent No.: 5,898,066, the entire contents of each of which are incorporated herein by reference.

Government Support

The United States government has certain rights in this invention by virtue of National Institutes of Health Grant No. RO1 EY 05690 to Larry Benowitz.

Summary of the Invention

The invention provides an isolated neurotrophic factor of the type that is present in medium in which Schwann cells have been cultured, that stimulates axonal outgrowth of naive goldfish retinal ganglion cells, that stimulates axonal outgrowth of embryonic rat spinal cord neurons, and that passes through a centrifugal filter with a 1 KDa cut-off. In an embodiment, the invention further provides a neurotrophic factor that fails to bind to a Cl 8 reversed-phase HPLC column, that forms a compound that elutes from a reverse-phase HPLC column at 23 minutes, after being chemically derivatized with AQC, and that has an elution time of 6 minutes on a GlO-Sepharose size-exclusive column. In an embodiment, the neurotrophic factor is AF-1, or an analog or derivative or AF-1.

The neurotrophic factor can be mammalian, human, bovine, rat, or can be derived from other types of vertebrates or invertebrates.

The invention provides methods for treating neurological conditions that include administering to a subject a therapeutically effective amount of a neurotrophic factor or composition comprising AF-1, thereby producing a neurosalutary effect in the subject. A "neurosalutary effect" means a response or result favorable to the health or function of a neuron, of a part of the nervous system, or of the nervous system generally.

In one aspect, the neurotrophic factor, such as AF-1, is administered to a subject in accordance with the present invention such that the factor is brought into contact with neurons of the central nervous system of the subject. For example, AF-1 may be administered into the cerebrospinal fluid of the subject into the intrathecal space by introducing the factor into a cerebral ventricle, the lumbar area, or the cisterna magna. In such circumstances, the AF-1 can be administered locally to cortical neurons or retinal ganglion cells to produce a neurosalutary effect.

In certain embodiments, the pharmaceutically acceptable formulation provides sustained delivery, providing effective amounts of the neurotrophic factor to a subject for at least one week, or in other embodiments, at least one month, after the pharmaceutically acceptable formulation is initially administered to the subject. Approaches for achieving sustained delivery of a formulation of the invention include the use of a slow release polymeric capsule, a bioerodible matrix, or an infusion pump that disperses the factor or other therapeutic compound of the invention. The infusion pump may be implanted subcutaneously, intracranially, or in other locations as would be medically desirable. In certain embodiments, the therapeutic factors or compositions of the invention would be dispensed by the infusion pump via a catheter either into the cerebrospinal fluid, or to a site where local delivery was desired, such as a site of neuronal injury or a site of neurodegenerative changes.

In another aspect, the present invention features a method that includes administering to a subject a therapeutically effective amount of a neurotrophic factor in combination with a therapeutically effective amount of a macrophage-derived factor and/or a cAMP modulator, thereby producing a neurosalutary effect in the subject.

Pharmaceutical compositions that include an AF-1 neurotrophic factor and a pharmaceutically acceptable carrier may be packed with instructions for use of the pharmaceutical composition for producing a neurosalutary effect in a subject. In one embodiment, the pharmaceutical composition may further include a cAMP modulator and/or a macrophage-derived factor, such as oncomodulin or TGF-β.

Other features and advantages of the invention will be apparent from the following detailed description, and from the claims.

Brief Description of the Drawings

Figure la is a histogram of the quantitation of neurite outgrowth: axon length distribution was measured five days after nerve cells were cultured with conditioned medium at the indicated concentrations of 0%, 5%, 10% and 15%. Number of cells extending processes l-five'cell diameter in length (light shading); number extending processes greater than five cell diameters in length (dark shading). Values represent averages from 4 wells for each conditioned medium concentration; error bars show + standard error of the mean (SEM). Figure lb is a graph with dose-response curves of 2 separate experiments showing neurite outgrowth (% cells with axons greater than five cell diameters) in response to increasing concentrations of conditioned medium (0%, 5%, 10%, and 15%). Data represent the percentage of cells with processes greater than five cell diameter in length, a cut-off point selected based upon the histogram data in Figure la. In both experiments, maximal outgrowth is attached in response to conditioned medium at a 10% concentration (i.e., total protein concentration of about 10 μg/ml). Error bars are not shown if less than 1%.

Figure lc is a graph of cell survival as a function of conditioned medium concentration in two independent experiments (5,6-CFDA labeled cells counted in 14 successive microscope fields, averaged for 4 wells, normalized by L-15 control values).

Figures 2a and 2b are graphs of neurite-promoting activity versus size fractions of conditioned medium. Figure 2a is conditioned medium obtained from previously intact optic nerves (day O) or from optic nerves that had been injured 3 or 7 days previously, separated into high and low molecular weight fractions by ultrafiltration (3,000 Da cut-off). In all cases, both low (light shading) and high (dark shading) molecular weight fractions yielded high levels of neurite-promoting activity. Figure 2b compares control with conditioned medium and conditioned medium separated with a molecular weight cut-off of 1,000 Da. Figure 2c is a chromatogram of the high molecular weight fraction of conditioned medium separated by size-exclusion high performance liquid chromatography (optical density, O.D., read at 280 nm). The high molecular weight fraction was concentrated 70-fold and then separated into 1 ml fractions (numbered bars). Arrows indicate the retention times of the molecular weight calibration standards (BSA, bovine serum albumin; Ova, ovalbumin; Cyto C, cytochrome C). Figure 2d is a graph of neurite outgrowth in response to column fractions assayed at concentrations of 25% (calculated on the basis of the starting material). Fractions were first bioassayed in pairs; if any activity was seen, they were retested individually, and otherwise they were retested in pairs. Only fractions 12 and 13 contained significant neurite-promoting activity.

Figure 3 is a graph of neurite outgrowth showing that the low molecular weight factor, AF-1, can be isolated using a two-phase solvent extraction system. The negative control is culture medium alone. The positive control, lane 2, consists of the low molecular weight (MW less than 3,000 Da) fractions of the molecules secreted by the glial sheath cells into culture medium that induce high levels of regenerative changes in the test cell cultures. When this material is mixed with an organic solvent at pH 7.5, isobutanol, little activity remains in the aqueous phase (pH 7.5 Aq). When the organic phase is then mixed with a low pH buffer (pH2 Aq), the biologically active molecule goes into the aqueous phase and nothing remains in the organic phase, (pH2 org). Figure 4 is a graph of neurite outgrowth showing that when the partially purified extract containing the low molecular weight trophic factor, AF-1, is separated by reversed faze HPLC, the active component appears in particular column fractions (FC, FD, FE). As in Figure 3, the negative control (L-15) is the tissue culture medium alone; the positive control is the unfractionated low molecular weight component of the molecules secreted by optic nerve glia (conditioned medium less than 3,000 Da, 10%) concentration); FA-FI indicate column fractions from the high performance liquid chromatography separation.

Figure 5 A depicts the elution profile obtained after the separation of a bovine vitreous fluid sample on a normal-phase LC-NH₂ column. Figure 5B is a graph depicting the results from the analysis of the elution fractions (from the normal-phase LC-NH₂ column) using an axonal growth assay.

Detailed Description of the Invention

I. Discovery of the Molecular Signals that Initiate Nerve Regeneration.

The capacity of lower vertebrates to re-grow an injured optic nerve has been the subject of numerous studies aimed at understanding CNS development and plasticity. To characterize the endogenous factors that induce retinal ganglion cells to regenerate their axons, a dissociated model of the goldfish retina cultured in serum-free, defined media has been developed. Under these conditions, retinal ganglion cells extend lengthy, axon-like processes in response to two soluble factors that derive from cells of the goldfish optic nerve. One of these factors, referred to herein as axogenesis factor 1 (AF-1), is a small, heat-stable, protease-sensitive molecule that passes through a 1 kDa cut-off filter; the second factor, referred to herein as axogenesis factor 2 (AF-2), is a heat-labile protein with an estimated size of 8-15 kDa.

These studies were conducted as follows. Three days after crushing the optic nerve behind the orbit, optic nerves and tracts were removed, cut into 0.5 to 1 mm pieces, and incubated in HEPES-buffered LI 5 medium for 3 to 4 hours. This conditioned medium was filtered and subjected to ion-exchange, HPLC reversed-phase and size-exclusion chromatography to purify factors that induce neurite outgrowth from dissociated retinal neurons in culture. Cultures were obtained by dissecting retinas from normal goldfish, treating with papain for 45 minutes, triturating gently, and sedimenting out large pieces of tissue. Outgrowth was scored blind in 3 to 6 wells per condition based on the fraction of large, viable cells extending neurites greater than or equal to five cell diameter after 5 days in culture, with viability assessed using 5(6)-carboxyfluorescein diacetate. A several-fold increase in neurite outgrowth was induced by a trypsin-sensitive, heat-stable, basic protein, M_r = 10 to 15 kDa. A second, distinct peak of neurite-promoting activity has a M_r of less than 1 kDa. These studies indicated that the goldfish optic nerve secretes multiple trophic factors that may make distinct contributions to axonal outgrowth.

As described herein, under baseline conditions, cells remained viable for at least a week but showed little outgrowth, as assessed using the vital dye 5,6-carboxyfluorescein diacetate (5,6-CFDA). Addition of conditioned medium (conditioned medium) containing molecules secreted by the support cells of the optic nerve (AF-1 and/or AF-2) induced up to 25% of neurons to extend processes greater than or equal to five cell diameters in length after five to six days. In some instances, this growth exceeded 300μm. To verify that this outgrowth was from retinal ganglion cells (RGCs) per se, the lipophilic dye 4-Di-10 ASP was applied to the optic tectum 5 to 7 days before dissecting retinas. After six days in culture, cells that were retrograde labeled with 4-Di-10 ASP showed twice as much neurite outgrowth as the overall population, indicating that conditioned medium acts upon RGCs selectively. The effect of conditioned medium was shown not to be secondary to enhanced viability, since neither the percentage of 4-Di-10 ASP-labeled cells in the total population nor overall cell survival was affected by the presence of conditioned medium.

AF-1 does not seem to coincide with any of the molecules identified previously in conditioned medium from the goldfish optic nerve. Mizrachi et al. (1986) described a 10 kDa protein in optic nerve conditioned medium that binds to DEAE at neutral pH and that is adsorbed onto polylysine substrate. This protein enhanced neurite outgrowth in retinal explants that had begun to regenerate their axons in vivo, but did not induce outgrowth from unprimed retinas. Other components of conditioned medium that also differ from the ones described here include apolipoprotein A, a 28 kDa protein that binds to heparin sulfate proteoglycans and that may contribute to lipid transport (Harel et al. 1989); a 60-65 kDa plasminogen activator that may be involved in the proteolysis of the extracellular matrix, thereby allowing growing axons to advance (Salles et al. 1990); a 28 kDa protein resembling interleukin-2 (IL-2; Eitan et al. 1992); a transglutaminase that may contribute to the dimerization of IL-2, rendering it toxic to oligodendrocytes (Eitan & Schwartz, 1993); platelet-derived growth factor (Eitan et al. 1992); an acidic 26 kDa protein that binds to polylysine substrate and induces embryonic mammalian neurons to extend long, unbranched axons (Caday et al. (1989) Mol. Brain Res. 5:45-50); and laminin, a 106 kDa glycoprotein that is a major constituent of the extracellular matrix (Hopkins et al. 1985; Battisti et al. 1992; Reichardt and Tomaselli (1991) Ann. Rev. Neurosci. 14:531-70); Giulian et al. (1986a) J. Cell Biol. 102:803-811; Giulian and Young (1986b) J. Cell Biol. 102:812-820; have described polypeptides of 3, 6, 9 and 125 kDa that are secreted by the tectum after optic nerve injury and that contribute to the proliferation of particular macroglial populations of the nerve. Finally, a group of glycoproteins with molecular weights greater than or equal 37 kDa (ependymins or X-GPs) that are secreted by cells of the choroid plexus (Thormodsson et al. (1992) Exp. Neural. 118:275-283) and the subependymal layer (Shashoua (1985) Cell. Mol. Neurobiol. 5:183-207), have been shown to promote axonal outgrowth in primed explants (Schmidt et al. 1991).

II. Characterization of Trophic Factors

Size-separation studies revealed that conditioned medium contains an active component, AF-1, that passes through a 1 kDa cut-off filter. It is heat resistant but sensitive to proteinase K digestion. AF-1 is considerably more concentrated in conditioned medium than in optic nerve homogenates, suggesting that it is actively secreted.

AF-1 induces vigorous neurite outgrowth from RGCs regardless of whether the regenerative response had been initiated in vivo by a priming lesion. Moreover, ganglion cells primed to grow by a conditioning lesion show essentially no outgrowth in the absence of AF-1. Thus, under the experimental conditions used herein, AF-1 is required to induce and maintain axonal regeneration.

The goldfish optic nerve consists of several cell types, including oligodendroglia, astrocytes, macrophages, microglia, and epithelial cells (Battisti et al, 1992). The trophic factors could be secreted from any of these or, alternatively, they might only be released from the cytoplasm of cells injured by nerve crush or by dissection in culture. To address this issue, the concentration of AF-1 in conditioned media and in cytosol fractions prepared from optic nerve homogenates was compared. AF-1 was found to be present in significantly higher concentrations in conditioned medium than in optic nerve cytosol, suggesting that it is actively secreted. Further analysis indicated that media containing factors secreted from dissociated goldfish optic nerve glia contains appreciable levels of a trophic factor of less than 3,000 Da, and lower levels of one greater than 3,000 Da. These findings indicated that it is the glial cells of the optic nerve that are the source of AF-1, and not damaged axons or blood. The latter source is also rendered unlikely by the absence of neurite-promoting activity in media conditioned by a variety of other tissues.

We have also found the vitreous fluid of newborn calves to be a rich source of AF-1. Characterization of AF-1

AF-1 was initially characterized by determination of the presence of trophic activity after passage of material through molecular weight filters or sieves of specific molecular weight. Several methods were used to determine the size of the active factor. Conditioned medium was first separated by centrifugal ultrafiltration using filters with molecular weight cut-offs of 10, 100, and 1000 kDa. Filtrates and retentates were tested in the bioassay. Next, conditioned medium was passed through a 6,000 Da desalting column and fractions were monitored by absorbance at 280 nm (for protein) and by measuring conductivity (for low-molecular weight fractions containing salts). Fractions containing high and low molecular weight constituents were evaluated by bioassay and were both found to be active. Fractions greater than 6,000 Da were pooled, concentrated 10- to 100-fold using a filter with a 3,000 Da cut-off, then separated by high performance liquid chromatography (HPLC). The low molecular weight material, less than 6,000 Da, was characterized further by being passed through a filter with a 1,000 Da molecular weight cut-off.

Anion-exchange chromatography of the trophic factor, for example, on diethylaminoethyl cellulose columns, was then carried out. The column was initially washed with 25 mM HEPES, and then stepwise eluted with 0.1, 0.2, 0.5 and 1.0 M NaCl in 25 mM HEPES.

The purification of AF-1 based on molecular weight and anion exchange chromatography, as described herein, yields samples with sufficient purity for structural and molecular weight analysis by mass spectrometry.

Having identified the structure of AF-1, various means of producing purified AF- 1 compound are available. The compound may be isolated from tissue or cell culture sources, or may be produced by synthetic means. These methods are known to those skilled in the art. These methods can be used to synthesize either AF-1 as characterized herein, or additionally compounds having AF-1 as a component of a larger chemical structure. Such compounds may be readily synthesize to produce compounds that will have desired characteristics of stability, pharmacokinetics, or other characteristics pertinent to medical uses. The compounds can be screened for activity as described above and in the following sections.

III. Use in Diagnostic and Screening Assays

Evolutionary Conservation and Isolation of Other Trophic Factors.

Research over the past ten years or so has clearly shown that the molecular elements that underlie the development and functioning of the nervous system are phylogenetically ancient and highly conserved throughout vertebrate evolution. Molecules such as transcription factors controlling programs of gene expression, known trophic factors, cell recognition molecules, transmitters and their receptors have remained remarkably unaltered over the last several hundred million years, and it is therefore predictable that homologous equivalents of AF-1 exists in the nervous system of higher vertebrates, including humans, and can be identified based on the analogous structures and sequences of the fish, rat, and bovine trophic factors initially described herein. Moreover, although these molecules were studied in the visual system, the retina and optic nerve develop ontogenetically as extensions of the midbrain and are essentially identical to other portions of the central nervous system. It is therefore expected that AF-1 will act upon other populations of neurons besides the retinal ganglion cells, including both central nervous structures such as the spinal cord and cerebral cortex, as well as peripheral nervous system components such as motor and sensory neurons. Injury to neurons in the cerebral cortex is a major pathogenic factor in stroke, while failure of spinal cord neurons to regenerate damaged axons is critical in limiting recovery after spinal cord injury. Confirmation of the efficacy of Af-1 in other model systems can be established using a primary culture system of dissociated neurons from the rat spinal cord, for example, as described by Banker and Goslin, eds. Culturing Nerve Cells (MIT Press, 1991).

Further experimental evidence of efficacy in producing neurosalutatory effects can be obtained using an in vivo model to investigate the effects of AF-1, for example, a mammalian spinal cord injury model. The spinal cord is transected through the dorsal columns, after which AF-1 is delivered either through the use of a minipump or by embedding in slow-release capsules, as described in more detail below.

Diagnostic and Screening Applications

AF-1 compositions can be used not only in the treatment of patients as described below, but in screening assays for the identification of drugs that modulate the activity of AF-1 and in screening of patient samples for the presence of a functional neurotrophic factor. Screening may be accomplished using antibodies, typically labeled with a fluorescent, radiolabeled, or enzymatic label, or by isolation of target cells and screening for binding activity, using methods known to those skilled in the art.

Production of Antibodies to AF-1

Animals such as mice may be immunized by administration of an amount of immunogen effective to produce an immune response. It may be advantageous to generate antibodies to a antigen of a different species of origin than the species in which the antibodies are to be tested or utilized. The methods involved are known to those skilled in the art. Monoclonal antibody technology can be used to obtain mAbs immunoreactive with AF-1. These may be useful in the purification of the neurotrophic factor. Methods for making monoclonal antibodies are now routine for those skilled in the art. See, for example, Galfre and Milstein (1981) Meth. Enzymol. 73:3-46, incorporated herein by reference. HAT-selected clones are injected into mice to produce large quantities of nxAb in ascites, that can be purified using protein A column chromatography (BioRad, Hercules, CA). MAbs are selected on the basis of their (a) specificity for a particular antigen, (b) high binding affinity, (c) isotype, and (d) stability. MAbs can be screened or tested for specificity using any of a variety of standard techniques, including Western blotting or enzyme-linked immunosorbent assay (ELISA) (Koren et al. (1986) Biochim. Biophys. Acta 876:91-100).

Screening for Drugs that Modulate Neurotrophic Factor Function, Expression or Activity

The neurotrophic factor is useful as a target for identifying compounds that modulate its function, expression or activity. Accordingly, the present invention provides a method or "screening assay" for identifying modulators that have a stimulatory or inhibitory effect on the function, in vivo concentration or activity of a neurotrophic factor or of a neurotrophic factor substrate. Potential modulators may be peptides, peptidomimetics, small molecules, carbohydrates, or other types of molecules.

In one embodiment, the invention provides assays for screening candidate or test compounds that are reactive with, bind to or modulate the activity of a neurotrophic factor of the invention. The test compounds of the present invention can be obtained using any of the numerous approaches in combinatorial library methods known in the art, including: biological libraries; spatially addressable parallel. solid phase or solution phase libraries; synthetic library methods requiring deconvolution; the 'one-bead one- compound' library method; and synthetic library methods using affinity chromatography selection. The biological library approach is limited to peptide libraries, while the other four approaches are applicable to peptide, non-peptide oligomer or small molecule libraries of compounds (Lam, K.S. (1997) Anticancer Drug Des. 12:145).

Examples of methods for the synthesis of molecular libraries can be found in the art, for example in: DeWitt et al. (1993) Proc. Natl. Acad. Sci. U.S.A. 90:6909; Erb et al. (1994) Proc. Natl. Acad. Sci. USA 91 :11422; Zuckermann et al. (1994). J. Med. Chem. 37:2678; Cho et al. (1993) Science 261:1303; Carrell et al. (1994) Angew. Chem. Int. Ed. Engl. 33:2059; Carell et al. (1994) Angew. Chem. Int. Ed. Engl. 33:2061; and in Gallop et al. (1994) J. Med. Chem. 37:1233. Libraries of compounds may be presented in solution (e.g., Houghten (1992) Biotechniques 13:412-421), or on beads (Lam (1991) Nature 354:82-84), chips (Fodor (1993) Nature 364:555-556), bacteria (Ladner USP 5,223,409), spores (Ladner USP '409), plasmids (Cull et al. (1992) Proc. Natl. Acad. Sci. USA 89:1865-1869) or on phage (Scott and Smith (1990) Science 249:386-390); (Devlin (1990) Science 249:404- 406); (Cwirla et al. (1990) Proc. Natl. Acad. Sci. 87:6378-6382); (Felici (1991) J. Mol. Biol. 222:301-310); (Ladner supra.).

This invention further pertains to novel agents identified by the above-described screening assays. Accordingly, it is within the scope of this invention to further use an agent identified as described herein in an appropriate animal model. For example, an agent identified as described herein (e.g., a neurotrophic factor modulating agent) can be used in an animal model to determine the efficacy, toxicity, or side effects of treatment with such an agent. Alternatively, an agent identified as described herein can be used in an animal model to determine the mechanism of action of such an agent. Furthermore, this invention pertains to uses of novel agents identified by the above-described screening assays for treatments as described herein.

The ability of AF-1 or a neurotrophic factor modulator to produce a neurosalutary effect in a subject may be assessed using any of a variety of known procedures and assays. For example, the ability of AF-1 or a neurotrophic factor modulator to re-establish neural connectivity and/or function after an injury, such as a CNS injury, may be determined histologically (either by slicing neuronal tissue and looking at neuronal branching, or by showing cytoplasmic transport of dyes). Alternatively, such ability may be assessed by monitoring the ability of AF-1 to fully or partially restore the electroretinogram after damage to the neural retina or optic nerve, or to fully or partially restore a pupillary response to light in the damaged eye.

Other tests that may be used to determine the ability of a neurotrophic factor to produce a neurosalutary effect in a subject include standard tests of neurological function in human subjects or in animal models of spinal injury (such as standard reflex testing, urologic tests, urodynamic testing, tests for deep and superficial pain appreciation, proprioceptive placing of the hind limbs, ambulation, and evoked potential testing). In addition, nerve impulse conduction can be measured in a subject, such as by measuring conduct action potentials, as an indication of the production of a neurosalutary effect.

Animal models suitable for use in the assays of the present invention include the rat model of partial transection (described in Weidner et al. (2001) Proc. Natl. Acad. Sci. USA 98:3513-3518). This animal model tests how well a compound can enhance the survival and sprouting of the intact remaining fragment of an almost fully-transected cord. Accordingly, after administration of the neurotrophic factor, e.g., AF-1 or AF-2, (alone or in combination with a macrophage-derived factor and/or a cAMP modulator) or the neurotrophic factor modulator, these animals may be evaluated for recovery of a certain function, such as how well the rats may manipulate food pellets with their forearms (to which the relevant cord had been cut 97%).

Another animal model suitable for use in the assays of the present invention includes the rat model of stroke (described in Kawamata et al. (1997) Proc. Natl. Acad. Sci. USA 94(15): 8179-8184). This paper describes in detail various tests that may be used to assess sensorimotor function in the limbs as well as vestibulomotor function after an injury. Administration to these animals of the compounds of the invention can be used to assess whether a given compound, route of administration, or dosage provides a neurosalutary effect, such as increasing the level of function, or increasing the rate of regaining function or the degree of retention of function in the test animals.

Standard neurological evaluations used to assess progress in human patients after a stroke may also be used to evaluate the ability of a neurotrophic factor, e.g., AF-1 or AF-2, (alone or in combination with a macrophage-derived factor and/or a cAMP modulator) or a neurotrophic factor modulator to produce a neurosalutary effect in a subject. Such standard neurological evaluations are routine in the medical arts, and are described in, for example, "Guide to Clinical Neurobiology" Edited by Mohr and Gautier (Churchill Livingston Inc. 1995).

For assessing function of the peripheral nervous system, standard tests include electromyography, nerve conduction velocity measurements, evoked potentials assessment and upper/lower extremity soma to-sensory evoked potentials. Electromyography records the electrical activity in muscles, and is used to assess the function of the nerves and muscles. The electrode is inserted into a muscle to record its electrical activity. It records activity during the insertion, while the muscle is at rest, and while the muscle contracts. The nerve conduction velocity test evaluates the health of the peripheral nerve by recording how fast an electrical impulse travels through it. A peripheral nerve transmits information between the spinal cord and the muscles. A number of nervous system diseases may reduce the speed of this impulse. Electrodes placed on the skin detect and record the electrical signal after the impulse travels along the nerve. A second stimulating electrode is sends a small electrical charge along the nerve; the time between the stimulation and response will be recorded to determine how quickly and thoroughly the impulse is sent.

Standard tests for function of the cranial nerves, as known to those skilled in the neurological medical art, include facial nerve conduction studies; orbicularis oculi reflex studies (blink reflex studies); trigeminal-facial nerve reflex evaluation as used in focal facial nerve lesions, Bell's palsy, trigeminal neuralgia and atypical facial pain; evoked potentials assessment; visual, brainstem and auditory evoked potential measurements; thermo-diagnostic small fiber testing; and computer-assisted qualitative sensory testing. IV. Treatment of Neurological Disorders

The neurotrophic AF-1 of the invention may be used to produce a neurosalutary effect in a subject. Typically, such methods include administering to a subject a therapeutically effective amount of a neurotrophic factor, e.g., an AF-1 factor, thereby producing a neurosalutary effect in a subject. Producing a neurosalutary effect in a subject may result or aid in the treatment of a neurological disorder in a subject.

As used herein, a "neurosalutary effect" means a response or result favorable to the health or function of a neuron, of a part of the nervous system, or of the nervous system generally. Examples of such effects include improvements in the ability of a neuron or portion of the nervous system to resist insult, to regenerate, to maintain desirable function, to grow or to survive. The phrase "producing a neurosalutary effect" includes producing or effecting such a response or improvement in function or resilience within a component of the nervous system. For example, examples of producing a neurosalutary effect would include stimulating axonal outgrowth after injury to a neuron; rendering a neuron resistant to apoptosis; rendering a neuron resistant to a toxic compound such as D -amyloid, ammonia, or other neurotoxins; reversing age-related neuronal atrophy or loss of function; or reversing age-related loss of cholinergic innervation.

The neurotrophic factor of the invention may be administered alone or in combination with a cAMP modulator and/or a macrophage-derived factor, such as oncomodulin or TGF-β.

The term "cAMP modulator" includes any compound that has the ability to modulate the amount, production, concentration, activity or stability of cAMP in a cell, or to modulate the pharmacological activity of cellular cAMP. cAMP modulators may act at the level of adenylate cyclase, upstream of adenylate cyclase, or downstream of adenylate cyclase, such as at the level of cAMP itself, in the signaling pathway that leads to the production of cAMP. Cyclic AMP modulators may act inside the cell, for example at the level of a G-protein such as Gi, Go, Gq, Gs and Gt, or outside the cell, such as at the level of an extra-cellular receptor such as a G-protein coupled receptor. Cyclic AMP modulators include activators of adenylate cyclase such as forskolin; non- hydrolyzable analogues of cAMP including 8-bromo-cAMP, 8-chloro-cAMP, or dibutyryl cAMP (db-cAMP); isoprotenol; vasoactive intestinal peptide; calcium ionophores; membrane depolarization; macrophage-derived factors that stimulate cAMP; agents that stimulate macrophage activation such as zymosan or IFN-D; phosphodiesterase inhibitors such as pentoxifylline and theophylline; specific phosphodiesterase IV (PDE IV) inhibitors; and beta 2-adrenoreceptor agonists such as salbutamol. The term cAMP modulator also includes compounds that inhibit cAMP production, function, activity or stability, such as phosphodiesterases, such as cyclic nucleotide phosphodiesterase 3B. cAMP modulators that inhibit cAMP production, function, activity or stability are known in the art and are described in, for example, Nano et al. (2000) Pflugers Arch 439(5): 547-54, the contents of which are incorporated herein by reference.

"Phosphodiesterase IV inhibitor" refers to an agent that inhibits the activity of the enzyme phosphodiesterase IV. Examples of phosphodiesterase IV inhibitors are known in the art and include 4-arylpyrrolidinones, such as rolipram, nitraquazone, denbufylline, tibenelast, CP-80633 and quinazolinediones such as CP-77059.

"Beta-2 adrenoreceptor agonist" refers to an agent that stimulates the beta-2 adrenergic receptor. Examples of beta-2 adrenoreceptor agonists are known in the art and include salmeterol, fenoterol and isoproterenol.

The term "administering" to a subject includes dispensing, delivering or applying an active compound in a pharmaceutical formulation to a subject by any suitable route for delivery of the active compound to the desired location in the subject, including delivery by either the parenteral or oral route, intramuscular injection, subcutaneous/intradermal injection, intravenous injection, buccal administration, transdermal delivery and administration by the rectal, colonic, vaginal, intranasal or respiratory tract route.

As used herein, the language "contacting" is intended to include both in vivo or in vitro methods of bringing a compound of the invention into proximity with a neuron such that the compound can exert a neurosalutary effect on the neuron.

As used herein, the term "effective amount" includes an amount effective, at dosages and for periods of time necessary, to achieve the desired result, such as sufficient to produce a neurosalutary effect in a subject. An effective amount of an active compound as defined herein may vary according to factors such as the disease state, age, and weight of the subject, and the ability of the active compound to elicit a desired response in the subject. Dosage regimens may be adjusted to provide the optimum therapeutic response. An effective amount is also one in which any toxic or detrimental effects of the active compound are outweighed by the therapeutically beneficial effects.

A therapeutically effective amount or dosage of an active may range from about 0.001 to 30 mg kg body weight, with other ranges of the invention including about 0.01 to 25 mg/kg body weight, about 0.1 to 20 mg/kg body weight, about 1 to 10 mg/kg, 2 to 9 mg/kg, 3 to 8 mg kg, 4 to 7 mg/kg, and 5 to 6 mg kg body weight. The skilled artisan will appreciate that certain factors may influence the dosage required to effectively treat a subject, including but not limited to the severity of the disease or disorder, previous treatments, the general health and/or age of the subject, and other diseases present. Moreover, treatment of a subject with a therapeutically effective amount of an active compound can include a single treatment or a series of treatments. In one example, a subject is treated with an active compound in the range of between about 0.1 to 20 mg/kg body weight, one time per week for between about 1 to 10 weeks, alternatively between 2 to 8 weeks, between about 3 to 7 weeks, or for about 4, 5, or 6 weeks. It will also be appreciated that the effective dosage of an active compound used for treatment may increase or decrease over the course of a particular treatment.

The term "subject" is intended to include animals. In particular embodiments, the subject is a mammal, a human or nonhuman primate, a dog, a cat, a horse, a cow or a rodent.

"Neurological disorder" is intended to include a disease, disorder, or condition that directly or indirectly affects the normal functioning or anatomy of a subject's nervous system. Elements of the nervous system subject to disorders that may be effectively treated with the compounds and methods of the invention include the central, peripheral, somatic, autonomic, sympathetic and parasympathetic components of the nervous system, neurosensory tissues within the eye, ear, nose, mouth or other organs, as well as glial tissues associated with neuronal cells and structures. Neurological disorders may be caused by an injury to a neuron, such as a mechanical injury or an injury due to a toxic compound, by the abnormal growth or development of a neuron, or by the misregulation (such as downregulation or upregulation) of an activity of a neuron. Neurological disorders can detrimentally affect nervous system functions such as the sensory function (the ability to sense changes within the body and the outside environment); the integrative function (the ability to interpret the changes); and the motor function (the ability to respond to the interpretation by initiating an action such as a muscular contraction or glandular secretion). Examples of neurological disorders include traumatic or toxic injuries to peripheral or cranial nerves, spinal cord or to the brain, cranial nerves, traumatic brain injury, stroke, cerebral aneurism, and spinal cord injury. Other neurological disorders include cognitive and neurodegenerative disorders such as Alzheimer's disease, dementias related to Alzheimer's disease (such as Pick's disease), Parkinson's and other Lewy diffuse body diseases, senile dementia, Huntington's disease, Gilles de la Tourette's syndrome, multiple sclerosis, amyotrophic lateral sclerosis, hereditary motor and sensory neuropathy (Charcot-Marie-Tooth disease), diabetic neuropathy, progressive supranuclear palsy, epilepsy, and Jakob- Creutzfieldt disease. Autonomic function disorders include hypertension and sleep disorders. Also to be treated with compounds and methods of the invention are neuropsychiatric disorders such as depression, schizophrenia, schizoaffective disorder, Korsakoff s psychosis, mania, anxiety disorders, or phobic disorders, learning or memory disorders (such as amnesia and age-related memory loss), attention deficit disorder, dysthymic disorder, major depressive disorder, mania, obsessive-compulsive disorder, psychoactive substance use disorders, anxiety, phobias, panic disorder, bipolar affective disorder, psychogenic pain syndromes, and eating disorders. Other examples of neurological disorders include injuries to the nervous system due to an infectious disease (such as meningitis, high fevers of various etiologies, HIV, syphilis, or post- polio syndrome) and injuries to the nervous system due to electricity (including contact with electricity or lightning, and complications from electro-convulsive psychiatric therapy). The developing brain is a target for neurotoxicity in the developing central nervous system through many stages of pregnancy as well as during infancy and early childhood, and the methods of the invention may be utilized in preventing or treating neurological deficits in embryos or fetuses in utero, in premature infants, or in children with need of such treatment, including those with neurological birth defects. Further neurological disorders include, for example, those listed in Harrison's Principles of Internal Medicine (Braunwald et al. McGraw-Hill, 2001) and in the American Psychiatric Association's Diagnostic and Statistical Manual of Mental Disorders DSM- IV (American Psychiatric Press, 2000) both incorporated herein by reference in their entirety.

The term "stroke" is art recognized and is intended to include sudden diminution or loss of consciousness, sensation, and voluntary motion caused by rupture or obstruction (for example, by a blood clot) of an artery of the brain.

"Traumatic brain injury" is art recognized and is intended to include the condition in which, a traumatic blow to the head causes damage to the brain or connecting spinal cord, often without penetrating the skull. Usually, the initial trauma can result in expanding hematoma, subarachnoid hemorrhage, cerebral edema, raised intracranial pressure, and cerebral hypoxia, that can, in turn, lead to severe secondary events due to low cerebral blood flow.

Pharmaceutical Compositions

Pharmaceutical compositions and packaged formulations comprising a neurotrophic factor of the invention and a pharmaceutically acceptable carrier are also provided by the invention. These pharmaceutical compositions may also include a macrophage-derived factor and/or a cAMP modulator.

In a method of the invention, the neurotrophic factor can be administered in a pharmaceutically acceptable formulation. Such pharmaceutically acceptable formulation may include AF-1 as well as one or more pharmaceutically acceptable carriers and/or excipients. As used herein, "pharmaceutically acceptable carrier" includes any and all solvents, dispersion media, coatings, antibacterial and anti fungal agents, isotonic and absorption delaying agents, and the like that are physiologically compatible. For example, the carrier can be suitable for injection into the cerebrospinal fluid. Excipients include pharmaceutically acceptable stabilizers and disintegrants. The present invention pertains to any pharmaceutically acceptable formulations, including synthetic or natural polymers in the form of macromolecular complexes, nanocapsules, microspheres, or beads, and lipid-based formulations including oil-in-water emulsions, micelles, mixed micelles, synthetic membrane vesicles, and resealed erythrocytes.

In one embodiment, the pharmaceutically acceptable formulations comprise a polymeric matrix. The terms "polymer" or "polymeric" are art-recognized and include a structural framework comprised of repeating monomer units that is capable of delivering a macrophage-derived factor such that treatment of a targeted condition, such as a neurological disorder, occurs. The terms also include co-polymers and homopolymers such as synthetic or naturally occurring. Linear polymers, branched polymers, and cross-linked polymers are also meant to be included.

The active therapeutic compound can be encapsulated in one or more pharmaceutically acceptable polymers, to form a microcapsule, microsphere, or microparticle, terms used herein interchangeably. In another embodiment, the pharmaceutically acceptable formulations comprise lipid-based formulations. Prior to introduction, the formulations can be sterilized, by any of the numerous available techniques of the art, such as with gamma radiation or electron beam sterilization.

Administration of the Pharmaceutically Acceptable Formulation

The pharmaceutically acceptable formulations of the invention are administered such that the active compound comes into contact with a subject's nervous system to produce a neurosalutary effect. Both local and systemic administration of the formulations are contemplated by the invention. Desirable features of local administration include achieving effective local concentrations of the active compound as well as avoiding adverse side effects from systemic administration of the active compound. In one embodiment, the active compound is administered by introduction into the cerebrospinal fluid of the subject. In certain aspects of the invention, the active compound is introduced into a cerebral ventricle, the lumbar area, or the cisterna magna. In another aspect, the active compound is introduced locally, such as into the site of nerve or cord injury, into a site of pain or neural degeneration, or intraocularly to contact neuroretinal cells.

The pharmaceutically acceptable formulations can be suspended in aqueous vehicles and introduced through conventional hypodermic needles or using infusion pumps.

In another embodiment of the invention, the active compound formulation is administered into a subject intrathecally. As used herein, the term "intrathecal administration" is intended to include delivering an active compound formulation directly into the cerebrospinal fluid of a subject, by techniques including lateral cerebroventricular injection through a burrhole or cisternal or lumbar puncture or the like (described in Lazorthes et al. Advances in Drug Delivery Systems and Applications in Neurosurgery, 143-192 and Omaya et al. Cancer Drug Delivery, 1 : 169-179, the contents of which are incorporated herein by reference). The term "lumbar region" is intended to include the area between the third and fourth lumbar (lower back) vertebrae. The term "cisterna magna" is intended to include the area where the skull ends and the spinal cord begins at the back of the head. The term "cerebral ventricle" is intended to include the cavities in the brain that are continuous with the central canal of the spinal cord. Administration of an active compound to any of the above mentioned sites can be achieved by direct injection of the active compound formulation or by the use of infusion pumps. Implantable or external pumps and catheter may be used.

For injection, the active compound formulation of the invention can be formulated in liquid solutions, preferably in physiologically compatible buffers such as Hank's solution or Ringer's solution. In addition, the active compound formulation may be formulated in solid form and re-dissolved or suspended immediately prior to use. Lyophilized forms are also included. The injection can be, for example, in the form of a bolus injection or continuous infusion (such as using infusion pumps) of the active compound formulation.

The active compound formulation may be administered by injection or through a surgically inserted shunt into a subject's cerebral ventricle, such as the lateral, third or fourth ventricle, or into the cisterna magna or the lumbar area of a subject. An additional means of administration to intracranial tissue involves application of compounds of the invention to the olfactory epithelium, with subsequent transmission to the olfactory bulb and transport to more proximal portions of the brain. Such administration can be by nebulized or aerosolized preparations.

In another embodiment of the invention, the active compound formulation is administered to a subject at a site of neurological injury.

Regardless of site or method of administration, it is initiated preferably within 100 hours of when an injury occurs (such as within 6, 12, or 24 hours of the time of the injury). In one embodiment, the active compound formulation described herein is administered to the subject in the period from the time of a neural system injury or appearance of neurological symptoms up to about 100 hours after the injury has occurred.

AF-1 can be used in conjunction with other compounds used to treat neurological disorders or conditions. For example, AF-1 may be used in conjunction with a PARS inhibitor (see, for example, published patent application WOO 142219 to Inotek), an interferon, human transforming growth factor (TGF) beta-2-cyclophosphamide glatiramer acetate (also known as Cop- 1 or copolymer- 1), azathloprine, corticosterolds such as methylprednisolone given intravenously or oral prednisone, immunoblobulins or one or more immunosuppressant treatments. The associated immunosuppressant treatment can be a chemical or physical treatment. For example, immunosuppressant agents or procedures can include total lymphoid irradiation, cladribine. mitoxantrone. sulfasalazine. cyclosporine. acyclovir, and oral bovine myelin. AF-1 may be administered in conjunction with antibodies that block growth-inhibiting proteins on the surface of CNS oligodendrocytes, as well as with agents that prevent free radical formation, such as LaZaroid, a 21-aminosteriod, or free radical scavengers such as phenylbutylnitrone and derivatives.

Duration and Levels of administration

In a preferred embodiment of the method of the invention, the active compound is administered to a subject for an extended period of time to produce a neurosalutary effect. Sustained contact with the active compound can be achieved by, for example, repeated administration of the active compound over a period of time, such as one week, several weeks, one month or longer.

More preferably, the pharmaceutically acceptable formulation used to administer the active compound provides sustained delivery, such as "slow release" of the active compound to a subject. For example, the formulation may deliver the active compound for at least one, two, three, or four weeks after the pharmaceutically acceptable formulation is administered to the subject. Preferably, a subject to be treated in accordance with the present invention is treated with the active compound for at least 30 days (either by repeated administration or by use of a sustained delivery system, or both).

As used herein, the term "sustained delivery" is intended to include continual delivery of the active compound in vivo over a period of time following administration, preferably at least several days, a week, several weeks, one month or longer. Sustained delivery of the active compound can be demonstrated by, for example, the continued therapeutic effect of the active compound over time (such as sustained delivery of the macrophage-derived factor can be demonstrated by continued production of a neurosalutary effect in a subject). Alternatively, sustained delivery of the active compound may be demonstrated by detecting the presence of the active compound in vivo over time.

Preferred approaches for sustained delivery include use of a polymeric capsule, a minipump to deliver the formulation, bioerodible implant, or implanted transgenic autologous cells (as described in U.S. Patent No. 6,214,622). Implantable infusion pump systems (such as Infusaid; see such as Zierski, J. et al. (1988) Acta Neurochem. Suppl. 43:94-99; Kanoff, R.B. (1994) J. Am. Osteopath. Assoc. 94:487-493) and osmotic pumps (sold by Alza Corporation) are available in the art. Another mode of administration is via an implantable, externally programmable infusion pump. Suitable infusion pump systems and reservoir systems are also described in U.S. Patent No. 5, 368,562 by Blomquist and U.S. Patent No. 4,731,058 by Doan, developed by Pharmacia Deltec Inc.

It is to be noted that dosage values may vary with the severity of the condition to be alleviated. It is to be further understood that for any particular subject, specific dosage regimens should be adjusted over time according to the individual need and the professional judgment of the person administering or supervising the administration of the active compound and that dosage ranges set forth herein are exemplary only and are not intended to limit the scope or practice of the claimed invention.

The invention, in another embodiment, provides a pharmaceutical composition consisting essentially of an AF-1 compositions and a pharmaceutically acceptable carrier, as well as methods of use thereof to effect neurosalutatory changes by contacting neurons with the composition. By the term "consisting essentially of is meant that the pharmaceutical composition does not contain any other modulators of neuronal growth such as, for example, nerve growth factor (NGF). In one embodiment, the pharmaceutical composition of the invention can be provided as a packaged formulation. The packaged formulation may include a pharmaceutical composition of the invention in a container and printed instructions for administration of the composition for producing a neurosalutary effect in a subject having a neurological disorder. Use of a macrophage- derived factor in the manufacture of a medicament for effecting a neurosalutatory change of neurons is also encompassed by the invention.

In vitro Treatment of CNS Neurons

Neurons derived from the central or peripheral nervous system can be contacted with a neurotrophic factor such as AF-1 (alone or in combination with a macrophage- derived factor and/or a cAMP modulator) in vitro to modulate axonal outgrowth. Accordingly, neurons can be isolated from a subject and grown in vitro, using techniques well known in the art, and then treated in accordance with the present invention to effect neurosalutatory changes, such as maintaining normal cellular architecture or chemical composition, promoting cellular survival, or enhancing neurocellular regeneration as evidenced by increasing GAP-43 expression or stimulating axonal outgrowth. Briefly, a neuronal culture can be obtained by allowing neurons to migrate out of fragments of neural tissue adhering to a suitable substrate (such as a culture dish), or by producing a suspension of neurons by disaggregating the tissue using mechanical, enzymatic or other means. For example, the enzymes trypsin, collagenase, elastase, hyaluronidase, DNase, pronase, dispase, or various combinations thereof can be used. Methods for isolating neuronal tissue and the disaggregation of tissue to obtain isolated cells are described in Freshney, Culture of Animal Cells: A Manual of Basic Technique, Third Ed. 1994, the contents of which are incorporated herein by reference.

Such cells can be subsequently contacted with an AF-1 neurotrophic factor in amounts and for a duration of time as described above. Once neurosalutatory changes been achieved in the neurons, these cells can be re-administered to the subject, such as by implantation.

The present invention will be further understood by reference to the following non-limiting examples. The contents of all patents, published patent applications, scientific articles and other references cited herein are hereby incorporated by reference.

EXAMPLES

Example 1: Development of model to identify endogenous neurotrophic factors.

An experimental model for identification of endogenous factors that induce regeneration of the optic nerve was developed. The goal was to establish a cell culture system enriched in retinal ganglion cells maintained at low cell densities to allow for •objective quantitation of neurite outgrowth and to minimize indirect effects mediated through other cell types. To identify the factors that initiate neurosalutatory effects, retinas were obtained from 'naive' animals and were cultured in the absence of serum

The results demonstrate that dissociated retinal ganglion cells survive well in the defined, serum-free conditions established here, but show little axonal outgrowth unless exposed to one of two factors that are secreted by the glial cells of the goldfish optic nerve. These factors are (a) a protease-sensitive, heat-resistant molecule less than one kilodalton in size, namely AF-1, and (b) a heat-and protease-sensitive molecule of 8 to 15 kDa, termed AF-2. Retrograde labeling experiments demonstrate that the retinal ganglion cells are the principal targets of these factors, reinforcing the likelihood that these molecules play a major role in initiating optic nerve regeneration in vivo.

Methods. Conditioned Media.

Comet goldfish (3 to 4 inches in length, Mt. Parnell Fisheries, Ft. Loudon, PA) were used to prepare both conditioned medium (conditioned medium) and dissociated retinal cultures. Animals were anesthetized by chilling to 4°C and then sacrificed by cervical transection. Optic nerves and tracts were dissected free of bone and connective tissue in two stages: a gross dissection under 2X magnification to yield optic nerves and tracts freed from the eyes and optic tecta but with some connective tissue and bone still attached, then a second stage carried out under 12X magnification using a table-top dissecting microscope (Wild). Following procedures described by Schwartz et al. (1985) and modified by Finkelstein et al. (1987), 6 optic nerves were placed in 3 ml HEPES- buffered Liebovitz L-15 medium (Gitco/BRL, Gaithersburg, MD) and cut into 1 to 2 mm segments. These were incubated for 3 to 4 hrs at 37°C in a 5% CO2 environment, then filter-sterilized with a 0.2 μm pore low protein-binding syringe filter (Acaroids, Gilman Sciences, Ann Arbor, MI). Conditioned medium was usually aliquoted and stored at -80°C immediately after preparation, though in some cases it was stored at 4°C for one to five days before being fractionated or used in bioassays. Protein determinations (Bradford kit, BSA standard; BioRad, Richmond, CA) carried out on several batches of conditioned medium showed a protein concentration of about 100 μ g/ml.

In cases where optic nerve surgery was carried out prior to dissection, animals were anesthetized in 0.5 mg/ml 3-aminobenzoic acid ethylester (Sigma Chemical Co., St. Louis, MO) and placed in a Plexiglass™ holder that fixed the position of the head and delivered a constant flow of aerated tank water to the gills. Two incisions were made in the superior rim of the orbit 3 mm apart, the bone flap was retracted, and orbital soft tissue and adventitia were dissected away to expose the optic nerves. Nerves were crushed bilaterally 1 to 2 mm behind the eyes using curved 4' jeweler's forceps. Orbital bleeding or transections of the nerves were considered grounds for eliminating animals from the study.

Dissociated Retinal Cultures.

Cultures were prepared using a modification of techniques described by Landreth & Agranoff (1976, 1979) and Dowling et al. (1985) Brain Res. 3 60: 331-338. Goldfish were dark-adapted in covered tanks for at least 30 min before being sacrificed. Eyes were removed rapidly and washed in sterile L-15, 70%) ethanol, and L-15 in quick succession. Lens, cornea, and iris were removed using iris scissors. The retina was teased from the sclera and pigment epithelium using microdissection scissors and jeweler's forceps under 25X magnification. Four retinas were placed in 5 ml of sterile digestion solution inside a laminar flow hood, in which the remainder of the culture preparation was carried out. To prepare the digestion solution, 100 units of papain (Worthington) plus 2.5 mg L-cysteine (Sigma) were added to 5 ml HEPES buffered L-15 brought to pH 7.4 with NaOH, then filter sterilized. After 4s min, the digestion solution was replaced with 5 ml sterile L-15 and the tissue was gently triturated 5 times to break the retina into small pieces. The solution was again replaced with 5 ml sterile L-15 and the tissue was triturated vigorously 5 times, separating retinas into fine fragments and removing photoreceptor cells. This step was repeated in fresh L-15 to create a single cell suspension. Successive trituration steps enriched the concentration of ganglion cells by removing most of the photoreceptor cells and mesenchymal tissue.

Cells were plated in 24-well tissue culture dishes (Costar, Cambridge, MA) coated with poly-L-lysine (MW greater 300,000, Sigma). Each well first received 200 pi of 2x medium E, that was developed based upon publications of Bottenstein (1983) In: Current Methods in Cellular Neurobiology, Vol. IV: Model Systems. J.L. Barker and J.F. McKelvy, eds., 107-130 (John Wiley & Sons, New York); Dichter Brain. Res. 149:279-293 (1978); Walicke et al. J. Neurosci. 6:114-121 (1986); and Aizenman & deVellis Brain Res. 406:32-42 (1987)). At final concentration, Medium E contains 20 nM hydrocortisone, 1 mM kainurinate, 100 μM putrescine, 20 nM progesterone, 30 nM selenium, 0.3 nM 3, 3'5-triiodo-L-thyronine, 50 μg/ml transferrin, 150 U/ml catalase, 60 U/ml superoxide dismutase, 1%> bovine serum albumin (Type V), 10 μg/ml gentamicin, 5 μg/ml insulin, and 15 mM HEPES (all reagents from Sigma). Medium E was titrated to pH 7.4 and filter-sterilized prior to being added to culture plates. To facilitate preparation and to help ensure reproducibility, the first six constituents were prepared together and stored at a 25X concentration in 0.5 ml aliquots at -20°C. After adding Medium E, each well received 50 pi of cell suspension, then the experimental or control sample brought up to 150 pi with L-15. Except where noted otherwise, experimental samples were set up in a blinded, randomized fashion by another member of the lab so that the investigator was unaware of the conditions present in any well. Within a given experiment, each experimental condition was represented in 4 to 8 wells; every experiment also included at least 4 wells of a positive control (of previously validated conditioned medium at a 5 to 15% concentration) and at least 4 wells of an L-15 and Medium E alone as a negative control. Plates were incubated for 5 to 6 days in a dark humidified tank at room temperature before being evaluated. Most experiments were repeated with material from 2 to 5 separate preparations. Data are presented as the mean + standard error for the 4 to 8 replicates. Where noted, some results are normalized by subtracting the growth in the negative controls and then dividing by the net growth in the positive controls.

Neurite Outgrowth Assay.

Neurite outgrowth was quantified after 5 or 6 days. Culture medium was replaced by 0.1 mg/ml 5,6-carboxyfluorescein diacetate (CFDA: Sigma) in phosphate-buffered saline (PBS) and incubated at room temperature for 10 min. CFDA, a vital dye, is taken up and metabolized by living cells to yield a fluorescent product that is distributed throughout the entire cell, allowing us to assess both cell viability and neurite outgrowth. Cultures were examined at lOOx magnification under fluorescent illumination (Nikon AF-BS inverted microscope) using a green barrier filter. The total number of viable cells in fourteen consecutive microscope frames (i.e., single well radii) was recorded starting at the top of the well. Cells matching the morphological criteria for retinal ganglion cells (RGCs), as established in retrograde labeling experiments (i.e., size and number of processes), were scored according to the length of their neurites, i.e., cells with neurites extending one to five cell diameters in length, five to ten cell diameters, 10 to 20 cell diameter, and greater than 20 cell diameter. In most instances, however, the last three bins were collapsed to give a single measure of neurite outgrowth, i.e., ([number of cells with neurites greater than five cell diameters] + [total number of viable cells] x 100).

Identification of Retinal Ganglion Cells.

Fish were anesthetized and a series of scalpel incisions were made within a region of the skull defined by the bone sutures above the optic tectum. The bone flap was retracted and crystals of the lipophilic dye, 4-(4-didecylaminostyryl)-N- methylpyridinium iodide (4-di 10 ASP: Molecular Probes, Inc., Portland, OR) were placed directly on the optic tecta. The bone flap was replaced and sealed with Aron Alpha (Ted Pella, Inc.). After allowing five to nine days for the dye to be transported back to the ganglion cells, retinas were dissected and cultured in the presence of either 10%) conditioned medium or control media alone, as described above. After six days in culture, neurite outgrowth was quantified under fluorescent microscopy for cells that were retrograde labeled with 4-di- 10 ASP. In addition to providing information about neurite outgrowth in ganglion cells per se, these studies helped establish criteria that were used to identify RGCs in the standard heterogeneous cultures.

Example 2: Determination of source of tropic factors.

Source of Trophic factor.

To investigate whether the trophic factor are actively secreted or just released from cells of the optic nerve that are damaged during the dissection, the activity of the high- and low-molecular weight fractions of conditioned medium and optic nerve cytosol were compared. Cytosol fractions were prepared by homogenizing 10 optic nerves in 25 mM HEPES, pH 7.4 or L-15. The high-speed supernatant of this extract was matched for protein concentration to whole conditioned medium using the BioRad™ protein assay. The optic nerve cytosol and conditioned medium were then separated into high and low molecular weight fractions with a Centriprep-3™ filter. Fractions were screened in the bioassay. The factors were also examined to determine if they were secreted selectively by the optic nerve by comparing standard conditioned medium with media conditioned with factors secreted by other goldfish tissues. The optic nerves required to prepare 3 ml of conditioned medium were weighed prior to mincing. Equal masses of tissue from goldfish skeletal muscle, liver, and gill were used to prepare conditioned media as described above.

To evaluate the effects of molecules previously found to affect growth in retinal explant cultures, taurine, at concentrations of 10-9 to 10-3 M (Sigma), retinoic acid (10-9 to 104 M: Sigma), and NGF (β-subunit, 100 nM: Collaborative Research, Bedford, MA) were tested in the bioassay. Additional experiments were carried out to examine whether the response of RGCs to conditioned medium depended on the density of plating in culture. In addition to the standard cell density used throughout the studies, the cells were also plated at 1/3, 1/9 and 1/27 of this density. Finally, experiments were carried out to compare the axonal outgrowth of 'primed' retinal ganglion cells that had begun to regenerate their axons in vivo, with that of 'native' retinas dissected from previously intact fish. Retinas were primed by allowing the regenerative process to proceed in vivo for 10 days prior to dissociating and plating, a period previously shown to maximally enhance axonal outgrowth in retinal explants (Landreth & Agranoff, 1976), and in nerves crushed a second time and allowed to regenerate in vivo (McQuarrie and Grafetein (1981) Brain Research, 216:253-264). Comparisons made between conditions are based on 2-tailed t-tests throughout.

Results.

Dissociated retinal cells respond to factors derived from the optic nerve.

The response of dissociated retinal cells to factors secreted by the optic nerve was determined using the dye 5,6-CFDA. Under baseline conditions, cells remain viable but show little outgrowth. With the addition of conditioned medium containing factors secreted by the optic nerve, cells 10-17 μm across extend one or two long processes of a uniformly thin caliber that sometime terminate in a prominent growth cone. Larger, polygonal cells are not counted in quantifying neurite outgrowth.

Six days after plating with L-15 and medium E alone, retinal cells remained viable but showed little outgrowth. Staining with 5,6-CFDA revealed a density of about 70 cells/mm2. Counting 14 microscope fields (i.e., 1 well radius) allowed a sampling of 200-300 cells/well. Addition of media conditioned by factors secreted by the optic nerve induced cells to extend long neurites that resemble axons. Dose-response Characteristics

The response of retinal cells to increasing concentrations of conditioned medium is shown in Figure 1. Figure la is a histogram of axon length distribution 5 days after being cultured with conditioned medium at the indicated concentrations. Although the number of cells extending processes 1 to five cell diameter in length changes little with increasing concentrations of conditioned medium, the number extending processes greater than five cell diameter in length increases greatly. Figure lb depicts dose-response curves of two separate experiments showing neurite outgrowth in response to increasing concentrations of conditioned medium. Data represent the percentage of cells with processes greater than five cell diameter in length, a cut-off point selected based upon the histogram date in Figure la. In both experiments, maximal outgrowth is attached in response to conditioned medium at a 10% concentration (i.e., total protein concentration of about 10 μg/ml). Figure lc is a graph of cell survival as a function of conditioned medium concentration in two independent experiments.

As indicated in the histograms of Figure la, in the absence of conditioned medium, 4% of cells had neurites in the range of 1 to five cell diameters and fewer than 1% had processes any longer than this. With the addition of conditioned medium at a 5% concentration, the process length distribution shifted markedly: 7% of cells now had neurites 5 to 10 cells in length and 2% had even longer processes. With higher concentrations of conditioned medium (15%), there were few cells left with axons in the 1 to five cell diameter. As in all subsequent experiments, the results shown are the means of greater than or equal to 4 wells for each sample + S.E.M. On the basis of the distribution patterns found here, most subsequent results have been represented as the percentage of cells with axons greater than five cell diameters, a cut-off point that discriminates responsive and non-responsive groups well.

Figure lb shows the dose-response curves of two consecutive experiments using different preparations of conditioned medium and retinas. For conditioned medium concentrations up to 10%), the number of cells with axons greater than five cell diameters in length increases, then levels off (outgrowth in response to 5% conditioned medium vs. L-15 + Medium E alone, p<0.001 for both experiments; growth with 10% conditioned medium vs. 5% conditioned medium, p<0.02 for both; error bars not shown if less than 1%>; one of these 2 experiments was carried out early in the study and was not blinded). The maximum number of cells responding differs somewhat between the two experiments, perhaps reflecting differences in the percentage of RGCs in the two preparations. The inset demonstrates that conditioned medium has little effect on cell viability. These data, taken from two studies, show the number of viable cells counted in 14 consecutive microscope fields (i.e., 1 well diameter), averaged over 4 wells for each condition, at increasing concentrations of conditioned medium. Data are normalized by the number of viable cells in the negative control (to account for differences in plating densities in the two experiments). Although viability appeared to be elevated in response to 15% conditioned medium in one experiment, this failed to achieve statistical significance (15% conditioned medium vs. L-15, p = 0.21), and was not seen in the other experiment.

Identification of Retinal Ganglion Cells.

Retrograde labeling was used to investigate outgrowth in retinal ganglion cells per se. Application of 4-Di-10 ASP to a small region of the optic tectum resulted in the retrograde labeling of 4-5%> of the viable cells in culture. These cells were ellipsoid, measuring 8-10 x 16-18 μm, similar to the dimensions of RGCs reported by Schwartz & Agranoff (1981) Brain Res. 206:331-343. The cells labeled with 4-di-10 ASP showed little spontaneous outgrowth in the presence of L-15 and Medium E alone; in response to 10% conditioned medium, however, these cells showed twice the level of neurite outgrowth observed in the overall cell population. The survival of RGCs is unaffected by conditioned medium. For both sets of retinas, the number of retrograde labeled cells in culture, divided by the total number of viable cells, was 4 to 5% irrespective of the presence or absence of conditioned medium. For outgrowth in RGCs vs. total cells, p < 0.005 for both sets of retinas.

Retinal ganglion cells were labeled by applying the lipophilic dye 4-di- 10 ASP to the optic tectum 7 days prior to culturing. After 6 days in culture, labeled cells extended one or two processes in response to conditioned medium. These cells generally formed one or two long, thin processes. For both sets of retinas, the viability of retrograde labeled cells relative to the overall cell population showed no change with the addition of conditioned medium. Thus, conditioned medium stimulates neurite outgrowth from RGCs selectively, and this effect is not a consequence of enhancing the survival of this cell type. This study also provides criteria (diameter, number of processes) for distinguishing the ganglion cells in the mixed, 5,6-CFDA-stained cultures. In the mixed cultures, neurite outgrowth is counted only from cells that match the criteria for RGCs and find that 15%>-25% of the total population extends neurites greater than five cell diameter. Since it can be determined from the retrograde labeling study that approximately one-third of neurons identified as RGCs are growing vigorously under these conditions, it follows that RGCs constitute 45-75%> of the cells in the mixed cultures.

Tissue Specificity of Conditioned Media.

Unlike media conditioned by the goldfish optic nerve, media conditioned by an equal mass of goldfish skeletal muscle, gill, or liver showed little neurite-promoting activity (all samples differ from optic nerve conditioned medium at p<O.Ol; experiment not blinded).

Example 3: Isolation and characterization of neurotropic factors.

Optic Nerve Conditioned Medium Contains Two Trophic Factors.

Preliminary separations carried out by ultrafiltration showed that all of the trophic activity passed through ultrafiltration devices with molecular weight cut-offs of 100 and 1000 kDa; in addition, a great deal of the activity passed through filters with cut-offs of 10,000 and 3,000 Da. On a size-exclusion column with a molecular weight cut-off of 6 kDa (i.e., a desalting column: Bio-Rad), neurite-promoting activity was found to be present both in fractions containing protein (as assessed by spectroscopy at O.D. 280) and in low-molecular weight fractions (containing salts, as assessed by measuring conductivity). On the basis of these observations, conditioned media was prepared from optic nerves dissected either from normal goldfish or from animals 3 or 7 days after bilateral optic nerve crush, then used ultrafiltration to separate the conditioned medium samples into fractions less than 3,000 Da and greater than 3,000 Da in size. Conditioned medium obtained from either intact or injured optic nerves yielded both high and low molecular weight neurite-promoting factors (all samples show higher growth than the L-15 control at a level of p<0.002). To simplify the figure, data have been represented by first subtracting the level of growth in negative controls grown with Medium E and L-15 alone (3 + 0.2% [mean + S.E.M.] in this particular experiment), then normalizing by the net growth in positive controls treated with unfractionated conditioned medium (21.3 + 2.3% in this instance). The data in Figure 2a suggest that most of the activity in unfractionated conditioned medium can be attributed to the smaller factor, though this is less evident in the material collected at three days post-injury. Qualitatively, the presence of high- and low-molecular weight trophic factors has been observed repeatedly in conditioned medium prepared from either intact or injured optic nerves.

Size Fractionation.

Several methods were used to determine the size of the active factors. Conditioned medium was first separated by centrifugal ultrafiltration using filters with molecular weight cut-offs of 10, 100, and 1000 kDa (Amicon, Beverly, MA). Filtrates and retentates were tested in the bioassay. Next, conditioned medium was passed through a 6 kDa desalting column (BioRad) and fractions were monitored by absorbance at 280 nm (for protein) and by measuring conductivity (for low-molecular weight fractions containing salts). Fractions containing high and low molecular weight constituents were evaluated by bioassay and were both found to be active, as shown in Figure 2a. Fractions greater than 6 kDa were pooled and concentrated 10- to 100-fold using a Centriprep-3™ filter (Amicon) with a 3 kDa cut-off. This material was then separated by high performance liquid chromatography (HPLC, Beckman Instruments) using a Biosep Sec™-S3000 N-capped bonded silica column (Phenomenex, Torrance, CA). Column fractions were screened in the bioassay, as shown in Figure 2c. The low molecular weight material (less than 6 kDa) was characterized further by being passed through a Centriprep-3~ filter (Amicon) or a Microsep™ (Filtron, Northborough, MA) centrifugal filter with a 1 kDa cut-off.

Heat and Protease Treatment.

To determine whether neurite-promoting factors in conditioned medium are polypeptides, high and low molecular weight fractions were heated to 95°C for 15 min or exposed to 0.1% trypsin. Soybean trypsin inhibitor at 0.125%) was added either together with the trypsin or after 1 or 2 hr incubation with trypsin. Samples were then screened in the bioassay. In addition, the less than 3 kDa fraction was exposed to pronase (Sigma) at 10 U/ml for 8 h (490C), or to proteinase K (PK, Boehringer Mannheim, Indianapolis, IN: 50 μg/ml, 56°C, 1 h). Following incubation, the enzymes were separated from low molecular weight components using a Centriprep-3= filter and the filtrates were bioassayed. Controls included heating active fractions without enzymes to verify that heat per se did not cause inactivation; and incubating the enzymes by themselves at 56° for I h, filtering, then adding the filtrate to the less than 3 kDa fractions to verify that the proteases were not generating autolytic fragments that might affect cell growth.

Sensitivity of the two factors to heat and proteases indicated that both factors are polypeptides. After heating at 95°C for 15 min, unfractionated conditioned medium (5%) lost half of its activity; the high molecular weight fraction, by itself (at a 20% concentration), lost nearly all of its activity. The low molecular weight factor by itself, treated for 1 h at 56°C or for 15 min at 95°C, lost only 30% of its activity.

Exposure to trypsin for 1 or 2 h diminished the activity of unfractionated conditioned medium by about 60%, although the low molecular weight fraction by itself showed little loss in activity after trypsin digestion. In the control, soybean trypsin inhibitor added at the same time as trypsin prevented the loss of activity. The sensitivity of the low molecular weight factor to proteases was examined further by treating it with pronase (8 h, 40°C) or with proteinase K (1 h, 56°C). Following the incubations, low molecular weight components were separated from the enzymes by ultrafiltration and then tested in the bioassay. Controls were incubated without the enzymes present. Pronase, like trypsin, had little effect on the activity of the low molecular weight factor(s). Proteinase K, however, diminished its activity by 80% (treatment at 56°C with and without proteinase K significant at p = 0.004). Thus, both the large and the small factors appear to be polypeptides.

Anion Exchange Chromatography.

Anion-exchange chromatography was carried out on diethylaminoethyl cellulose columns (DE-52, Whatman, Hillsboro, OR). DE-52 beads were equilibrated with 25 mM HEPES at pH 8.4, then added at a ratio of 0.5 ml hydrated beads: 10 ml conditioned medium (which had been desalted using a 6 kDa cut-off size-exclusion column). After an overnight incubation (4°C), the mixture was pipetted into a 5 mm I.D. glass Econo- column (BioRad). The unbound fraction and the first rinse of the column with 25 mM HEPES were pooled. Bound proteins were then eluted with successive 3 ml steps of 0.1, 0.2, 0.5 and 1.0 M NaCl in 25 mM HEPES. Fractions were divided into aliquots and stored at -80°C for bioassays. Ion-exchange chromatography was also carried out on the high molecular weight fraction of conditioned medium at pH 10.

The Smaller Trophic Factor Passes Through a 1 , 000 Da Filter.

Further fractionation of conditioned medium using a Microsep™ filter with a 1,000 Da cut-off yields a high level of activity in the filtrate, as shown by Figure 2b; less than 1,000 Da fraction versus L-15 control, p = 0.01; less than 1,000 Da fraction versus total conditioned medium, p = 0.06. Since the ability of a molecule to pass through this pore size depends on its shape as well as its size, however, this is not absolutely determined of size.

Size Estimate of the Larger Trophic Factor.

Figure 2c shows the elution profile of the high molecular weight fraction of conditioned medium when separated by size-exclusion HPLC. As seen by SDS-polyacrylamide gel electrophoresis (Caday et al. 1989) and the present chromatogram, conditioned medium contains a complex mixture of proteins. Fractions were initially bioassayed in groups of two; if pooled fractions showed any activity, they were rescreened individually, or in pairs again if not. High levels of activity were observed in fractions 12 and 13 in both the initial and in the secondary screens (Figure 2d; fraction 12 vs. L-15, p == 0.01; fraction 13 vs. L-15, p = 0.053; all others N.S.). The active factor has a similar retention time as cytochrome C (12 kDa; Figure 2d), so the size is estimated to be in the range of between 8,000 and 15,000 Da. In some experiments, an additional peak of activity has been observed at 70 to 100 kDa, but this has not been reproducible. This larger molecule may be unstable and degrade to form the 12,000 Da factor, or it may be a multimeric complex that dissociates under certain experimental conditions.

Charge and Substrate Binding.

Ion-exchange chromatography was used to purify the larger factor further. Separation of the high molecular weight component of conditioned medium by DEAE-anion-exchange chromatography was performed. At pH 8.4, neurite-promoting activity was recovered in the unbound fraction, whereas at pH 10, the active factor bound to the column and eluted with 0.2 M NaCl. Neither component of conditioned medium acts as a substrate-bound trophic factor. Conditioned medium was separated into high and low molecular weight fractions by ultrafiltration (3,000 Da cut-off). Prolonged exposure of the larger protein to elevated pH appears to reduce its activity significantly.

Polylysine-coated plates were incubated overnight with either the high or low molecular weight fractions of conditioned medium, either at full-strength or at a 1 :10 dilution. After rinsing wells to remove unbound material, no neurite-promoting activity . was retained after rinsing plates with L-15. Thus, neither the large nor the small factor acts as a substrate-bound growth factor.

Two-phase extraction of conditioned medium

Figure 3 is a graph of neurite outgrowth showing that the low molecular weight factor, AF-1, can be isolated using a two-phase solvent extraction system. The negative control is culture medium alone; the positive control, lane 2, is the low molecular weight fractions of the molecules secreted by the glial sheath cells into culture medium, conditioned medium less than 3,000 Da that induces high levels of axonal growth. When this material is mixed with an organic solvent at pH 7.5, isobutanol, little activity remains in the aqueous phase (pH 7.5 Aq). When the organic phase is then mixed with a low pH buffer (pH2 Aq), the biologically active molecule goes into the aqueous phase and nothing remains in the organic phase, (pH2 org). This provides a simple method for the initial purification of the low molecular weight trophic factor AF-1.

Figure 4 is a graph of neurite outgrowth showing that when the partially purified extract containing the low molecular weight trophic factor, AF-1, is separated by reversed phase HPLC, the active component appears in the column fractions designated FC, FD, and FE. As described above, the negative control (L-15) is the tissue culture medium alone; the positive control is the unfractionated low molecular weight component of the molecules secreted by optic nerve glia (conditioned medium less than 3,000 Da, 10%) concentration); FA-FI indicate column fractions from the high performance liquid chromatography separation.

Intra- and Extracellular Concentration of the Two Factors

Whether the two neurite-promoting factors are actively secreted was then determined by comparing their activity in conditioned medium and in the high-speed supernatant fraction of optic nerve homogenates. Samples were used in the bioassay at concentrations of 10% and 20%, adjusting the protein concentration of the optic nerve cytosol to match that of the conditioned medium (i.e., a 10%) concentration is equivalent to a protein concentration prior to ultrafiltration of about 10 ~g/ml protein for each. This is based on the hypothesis that most of the proteins found in conditioned medium arise from cell lysis and not by active secretion).

The low molecular weight factor is considerably more concentrated in conditioned medium than in the ON Cyto (p = 0.002). These data have been replicated in two more experiments and suggest that the smaller molecule is actively secreted. The larger factor is present at equal concentrations intra- and extracellularly.

Effect of Cell Density

To determine whether cell density affects the response of retinal ganglion cells to conditioned medium, retinas were plated at either the standard density used throughout, these studies (about 70 cells/mm2) or at increasingly lower densities. If RGCs are responding to a secondary factor released by another type of cell that is the direct target of conditioned medium, then as the number of these other cells decrease and the concentration of a secondary factor decreases, one would expect to find a diminished response of RGCs at lower cell densities. At one-third the standard cell density (about 2five cells/mm2), retinal neurons appeared to have a slightly higher response to conditioned medium than at the standard density (N.S.), and with another 3-fold dilution, outgrowth was only 30% lower than at the normal plating density [p = 0.18, not significant]. At 1/27 the standard density, outgrowth did show a significant decrease (p = 0.05).

Absence of a Priming Effect

Retinas dissected from either previously intact fish or fish that had undergone optic nerve surgery 14 days previously to initiate the regenerative response were dissociated and cultured in the presence of either control (L-15) medium alone, unfractionated conditioned medium at a 10%) concentration, or the low molecular weight fraction of conditioned medium at a 10%) concentration. In all cases, the response of RGCs was unaffected by the 'priming' lesion. This was done to determine whether the neurite-promoting factors in conditioned medium would allow 'naive' retinal ganglion cells to grow to the same extent as 'primed' cells in which the regenerative process had been initiated in vivo. Four retinas from previously intact fish were pooled, as were 4 retinas from fish that had undergone bilateral optic nerve surgery 10 days previously. Like 'naive' retinal ganglion cells, 'primed' RGCs showed little spontaneous outgrowth in control media. In the presence of either total (unfractionated) conditioned medium or the low molecular weight fraction alone (<3 kDa), neurons from 'naive' and 'primed' retinas showed identical levels of neurite outgrowth.

Activity of Other Molecules on Dissociated Retinal Cultures

Whether other molecules that have been reported to alter the growth characteristics of retinal explant cultures would be active in this system was then determined. Taurine, retinoic acid, and NGF have all been reported to influence neurite outgrowth from retinal explants.

Lima et al. (1989) reported that taurine, in the presence of laminin, augments neurite outgrowth from retinal explants primed to regenerate in vivo, but has little effect on previously intact retina. Taurine also contributes to the differentiation of rod cells (Altschuler et al. (1993) Development, 119:1317-1328). In this cell system, taurine, at concentrations between 1 μm and 10 mM, had no effect at all, nor did retinoic acid at concentrations between 10~9 and 10"4.

Retinoic acid (RA), a prominent factor in. cell differentiation, has been found to enhance the expression of the intermediate filament proteins ONi (gefiltin) and ON2 in goldfish retinal explants without affecting outgrowth per se (Hall et al. 1990). In the present study, RA (between 10"" to 10"4 M) had no measurable effect.

Preliminary experiments found no effects of NGF on dissociated retinal neurons at 50 nM and weak stimulation at 500 nM; the results show an absence of NGF activity at 100 nM, 20 times the dosage that enhances axonal outgrowth in primed explant cultures (Turner et al. 1982).

Although nerve growth factor (NGF) failed to elicit neurite outgrowth in these cultures, it may nevertheless contribute to optic nerve regeneration in an indirect fashion. NGF activity has been demonstrated in the goldfish brain (Benowitz & Greene (1979) Brain Res. 162:164-168), and the presence of an NGF-like molecule in optic nerve conditioned medium is supported by preliminary western blot studies showing that antibodies to mouse NGF cross-react with a protein of 12-13 kDa, the expected size of the β-NGF monomer. At low concentrations (i.e. 5 ng/ml), the β-subunit of mammalian NGF augments neurite outgrowth from goldfish retinal explants that had been primed to grow in vivo, while antibodies to NGF suppress outgrowth from primed retinal explants maintained in the presence of serum (Turner et al. 1982). However, NGF has little effect on explants of unprimed retina (turner et al. 1982) or on the rate of axonal outgrowth in vivo (Yip & Grafstein, 1982). Thus, although NGF is likely to play a modulatory role in this system, it does not seem to induce axonal outgrowth directly. NGF may stimulate glial cells to release factors that in turn act upon neurons.

The relationship of the two factors to one another is not clearly understood at this time. AF-1 can each induce neurite outgrowth independently in these assays, and their effects do not appear to be synergistic. Nevertheless, it is possible that in vivo they function in a complementary fashion. It is also possible that AF-1 derives from degradation of AF-2. Since the cultures contain a variety of cell types, it remains possible that AF-1 may not act directly upon retinal ganglion cells, but rather upon another cell type as an intermediary. As observed by Schwartz & Agranoff (1981), RGCs are the dominant cell type in dissociated goldfish retinal cultures prepared as described above, and support cells are not abundant. Thus, rather than suppressing proliferation of support cells, cell density was systematically reduced to decrease the concentration of any secondary factors that might be released by another cell type while holding the concentration of conditioned medium constant. Since neurite outgrowth did not decline significantly over a 9-fold decrease in cell density, the RGCs would appear to be responding to conditioned medium directly. At a cell density of c. 3 cells/mm^, however, outgrowth did show a significant decline. This could result from a decreased concentration of a trophic factor, released from another cell type in culture or from RGCs themselves, this is required to maintain the cells in a state in which they can respond to conditioned medium.

These findings lend further support to the specificity of AF-1 in inducing axonal outgrowth from RGCs.

Example 4: Isolation and Characterization of Rat AF-1

To isolate a mammalian AF-1, rat Schwann cells, derived from postnatal sciatic nerves, were grown to confluence in the presence of a pituitary-derived factor using the method of Brockes (1979) Brain Res.165 (1): 105-18. Cells were rinsed three times and incubated in saline for 1-2 hours. Conditioned media, containing factors secreted by Schwann cells, were passed through ultrafiltration devices with molecular weight cutoffs of 3 and 1 kilodaltons. Biological activity of the mammalian AF-1 was examined using our standard bioassay, e.g., axonal outgrowth from dissociated goldfish retinal ganglion cells.

In parallel studies, it was investigated whether Schwann cells in vivo in the intact or injured sciatic nerve also release an AF-1-like factor. The sciatic nerve of rats was surgically injured or left intact. At various times afterwards, nerves were dissected out, and incubated in phosphate buffered saline for 2 hours. Conditioned media was processed by ultrafiltration to separate high and low molecular weight components.

Physical properties of Schwann cell low molecular weight growth factor and goldfish AF-1

As shown in table I, similar to the goldfish AF-1, the mammalian AF-1 is readily extracted into 95% ethanol, after being lyophilized, thereby readily separating it from inorganics. Other properties established for mammalian AF-1 are that it fails to bind to a C18 reversed-phase HPLC column, that on a GlO-Sepharose size-exclusive column, it has a characteristic elution time, and that after being chemically derivatized with AQC it forms a compound that elutes from a reverse-phase HPLC column at 23 minutes.

TABLE I Property AF-1 Schwann cell factor

Size by ultrafiltration < lkDa < lkDa

Peak elution volume, G10 size-exclusion 6 minutes 6 minutes

C18 column elution unbound unbound

C18 column after AQC derivatization 23 minutes 23 minutes

Solubility in 95% EtOH very soluble very soluble

Induces axon outgrowth +++ +++

Secretion of an AF-1 molecule in vivo after peripheral nerve injury

Conditioned media from rat sciatic nerves collected at various times after injury were subjected to ultrafiltration to collect low molecular weight factors (< lkDa) that were tested in a standard bioassay. Whereas only very low levels of axon-promoting activity was found in media conditioned by the intact sciatic nerve, media conditioned by sciatic nerve 12 hours after surgery showed high levels of activity: this activity peaked at 1-2 days and then declined, but still remained significantly above the background level at 14 days after nerve injury.

Activity in mammalian bioassay systems AF-1 derived from mammalian Schwann cells was tested to determine whether it can promote axon regeneration in the injured mammalian nervous system. For these studies, the experimental paradigm described in Benowitz et al. (1999) Proc. Natl. Acad. Sci. U S A. 96(23): 13486-90 was used. Briefly, the corticospinal tract (CST) of mature rats was transected on one side of the brain. The CST in humans is the principal pathway for the voluntary control of the limbs and the trunk, sending its projections from the sensorimotor cortex to the spinal cord. AF-1 was applied to the uninjured side of the brain via a minipump for 2 weeks, after which the trajectories of new axons were investigated using the anterograde tracer biotin dextran amine (BDA). Mammalian AF- 1 induced even more axon growth than inosine. AF-1 caused numerous axon branches to grow from the intact CST over to the denervated side. It is likely that these new axons form functional connections (synapses) in the appropriate region on the denervated side, since varicosities were observe that resemble synaptic endings in the correct lamina of the spinal cord.

Example 5: Isolation and Characterization of Bovine AF-1

To isolate a mammalian AF-1 from another source, vitreous fluid from the eye of a cow was used. Briefly, vitreous fluid was extracted from the bovine eye and diluted in saline (3:1 saline:vitreous fluid). High molecular weight components were separated out using ultrafiltration (and a membrane of a 1 kDa cut-off) and the low-molecular weight filtrate was lyophilized. AF-1 was extracted into acetonitrile (CAN: in ratio of 10 ml CAN: 1 ml diluted vitreous fluid) and the soluble fraction was dried by speed- vac. The lyophilized ACN extract was solubilized in acetonitrile (3:1 acetonitrile:water; vol. < 0.5 ml) and the sample was applied to a normal-phase LC-NH₂ column. The resulting elution profile is shown in Figure 5 A.

The column fractions obtained from the foregoing chromatography were analyzed in using axon growth bioassay described above. The results, shown in Figure 5B, demonstrated that, whereas most components applied to the column elute within the first 8 minutes and have no biological activity in the bioassay, a late-eluting component (retention time 9-10 minutes) contains all of the axon-inducing activity. The biologically active fraction was re-chromatographed on the same column, yielding several sharp peaks. The fraction eluting at 9.8 minutes, but not the adjacent fractions, contained axon-promoting activity in the axon growth bioassay. This material is being analyzed using mass spectrometry, amino acid analysis, and N-terminal sequencing. Thus far, analysis has revealed the presence of a principal component with a mass of approximately 382.9 daltons. Example 6: Isolation and Characterization of Bovine AF-1

We have found the vitreous fluid of newborn calves to be a rich source of AF-1. Molecules in the vitreous fluid are extracted into saline (4:1 dilution) at 4°C. The saline extract (BV 1 :4) is passed through a lkDa molecular weight cut-off filter to remove components >1 kDa. The low molecular weight fraction is then lyophilized, and concentrated approximately 30:1 by redissolving in a small volume of water. This sample is then chromatographed on a Sephadex™-G10 column attached to an HPLC instrument, and the eluted material is monitored at 214 nm and 280 nm. Fractions are then bioassayed using our standard dissociated goldfish retinal ganglion cell cultures (described in Schwalb et al., 1995). The fractions containing the axon-promoting activity are then re-pooled and separated again on the same column to purify the axon-promoting activity further. This material is then applied to a LC-NH₂ column (normal-phase), using as a mobile phase 75%> acetonitrile in water. From our previous studies, AF-1 was found to elute with a long retention time, thereby separating it from most other components of the BV. Combining the two columns (G10 Sepharose™ and LC-NH₂ yields a highly enriched AF-1. This material is then analyzed by mass-spectrometry (MS-MS) to identify the structure of the components present. We then synthesize the molecules found to be present (or purchase commercially if they are available), and test their biological activity in the goldfish retinal ganglion cell assay. If the predicted molecular structure mimics that of AF-1, this will provide strong evidence that we have the correct structure for AF- I .

Equivalents

Those skilled in the art will recognize, or be able to ascertain using no more than routine experimentation, many equivalents to the specific embodiments of the_. invention described herein. Such equivalents are intended to be encompassed by the following claims.

Claims

We claim:

1. An isolated neurotrophic factor of the type that:

(a) is present in medium in which Schwann cells have been cultured;

(b) stimulates axonal outgrowth of naive goldfish retinal ganglion cells;

(c) stimulates axonal outgrowth of embryonic rat spinal cord neurons; and

(d) passes through a centrifugal filter with a 1 KDa cut-off.

2. The neurotrophic factor of claim 1 , that further:

(a) fails to bind to a C18 reversed-phase HPLC column;

(b) forms a compound that elutes from a reverse-phase HPLC column at 23 minutes, after being chemically derivatized with AQC; and

(c) has an elution time of 6 minutes on a GlO-Sepharose size-exclusive column.

3. The neurotrophic factor of claim 1 , that is mammalian AF- 1.

4. The neurotrophic factor of claim 1 , that is human AF- 1.

5. The neurotrophic factor of claim 1 , that is bovine AF- 1.

6. The neurotrophic factor of claim 1 , that is rat AF- 1.

7. The neurotrophic factor of claim 1 further comprising a pharmaceutically acceptable carrier for administration to a subject.

8. The neurotrophic factor of claim 7, wherein the pharmaceutically acceptable carrier is a dispersion system.

9. The neurotrophic factor of claim 8, wherein the pharmaceutically acceptable carrier comprises a lipid-based carrier.

10. The neurotrophic factor of claim 9, wherein the pharmaceutically acceptable carrier comprises a liposome carrier.

11. The neurotrophic factor of claim 10, wherein the pharmaceutically acceptable carrier comprises a multivesicular liposome carrier.

12. The neurotrophic factor of claim 7, wherein the pharmaceutically acceptable carrier comprises a polymeric matrix.

13. The neurotrophic factor of claim 12, wherein the polymeric matrix is selected from the group consisting of naturally derived polymers, such as albumin, alginate, cellulose derivatives, collagen, fibrin, gelatin, and polysaccharides.

14. The neurotrophic factor of claim 12, wherein the polymeric matrix is selected from the group consisting of synthetic polymers such as polyesters (PLA, PLGA), polyethylene glycol, poloxomers, polyanhydrides, and pluronics.

15. The neurotrophic factor of claim 12, wherein the polymeric matrix is in the form of microspheres.

16. The neurotrophic factor of claim 1 , further comprising a compound selected from the group of agents that prevent free radical formation, free radical scavengers, and counter growth-inhibiting molecules.

17. A method comprising administering to a subject a therapeutically effective amount of a neurotrophic factor, thereby producing a neurosalutary effect in said subject.

18. The method of claim 17, wherein said neurotrophic factor is AF- 1.

19. The method of claim 17, further comprising administering to said subject a macrophage-derived factor.

20. The method of claim 19, wherein said macrophage-derived factor is oncomodulin.

21. The method of claim 19, wherein said macrophage-derived factor is TGF- β.

22. The method of claim 17, further comprising administering to said subject a cAMP modulator.

23. The method of claim 22, wherein said cAMP modulator is non- hydrolyzable cAMP analogues, adenylate cyclase activators, macrophage-derived factors that stimulate cAMP, macrophage activators, calcium ionophores, membrane depolarization, phosphodiesterase inhibitors, specific phosphodiesterase IV inhibitors, beta2-adrenoreceptor inhibitors or vasoactive intestinal peptide.

24. The method of claim 17, wherein the neurosalutary effect is produced in said subject by modulating neuronal survival.

25. The method of claim 17, wherein the neurosalutary effect is produced in said subject by modulating neuronal regeneration.

26. The method of claim 17, wherein the neurosalutary effect is produced in said subject by modulating neuronal axonal outgrowth.

27. The method of claim 17, wherein the neurosalutary effect is produced in said subject by modulating axonal outgrowth of central nervous system neurons.

28. The method of claim 27, wherein the central nervous system neurons are retinal ganglion cells.

29. The method of claim 17, wherein the neurotrophic factor is administered by introduction into a region of neuronal injury.

30. The method of claim 17, wherein the neurotrophic factor is introduced into the cerebrospinal fluid of the subject.

31. The method of claim 17, wherein the neurotrophic factor is introduced to the subject intrathecally.

32. The method of claim 17, wherein the neurotrophic factor is introduced into a region selected from the group consisting of a cerebral ventricle, the lumbar area, and the cisterna magna of the subject.

33. The method of claim 17, wherein the neurotrophic factor is administered to the subject in a pharmaceutically acceptable formulation.

34. The method of claim 33, wherein the pharmaceutically acceptable formulation is a dispersion system.

35. The method of claim 33, wherein the pharmaceutically acceptable formulation comprises a lipid-based formulation.

36. The method of claim 35, wherein the pharmaceutically acceptable formulation comprises a liposome formulation.

37. The method of claim 35, wherein the pharmaceutically acceptable formulation comprises a multivesicular liposome formulation.

38. The method of claim 33 , wherein the pharmaceutically acceptable formulation comprises a polymeric matrix.

39. The method of claim 33, wherein the pharmaceutically acceptable formulation is contained within a minipump.

40. The method of claim 33, wherein the pharmaceutically acceptable formulation provides sustained delivery of the neurotrophic factor for at least one week after the pharmaceutically acceptable formulation is administered to the subject.

41. The method of claim 33 , wherein the pharmaceutically acceptable formulation provides sustained delivery of the neurotrophic factor for at least one month after the pharmaceutically acceptable formulation is administered to the subject.

42. The method of claim 17, wherein the subject is a mammal.

43. The method of claim 42, wherein the mammal is a human.

44. The method of claim 17, wherein said subject is suffering from a neurological disorder.

45. The method of claim 44, wherein said neurological disorder is a spinal cord injury.

46. The method of claim 45, wherein the spinal cord injury is characterized by monoplegia, diplegia, paraplegia, hemiplegia and quadriplegia.

47. The method of claim 44, wherein said neurological disorder is epilepsy.

48. The method of claim 47, wherein the epilepsy is posttraumatic epilepsy.

49. The method of claim 44, wherein said neurological disorder is Alzheimer's disease.

50. A method comprising administering to a subject a therapeutically effective amount of a neurotrophic factor in combination with a therapeutically effective amount of a macrophage-derived factor, thereby producing a neurosalutary effect in said subject.

51. A method comprising administering to a subject a therapeutically effective amount of a neurotrophic factor in combination with a therapeutically effective amount of a macrophage-derived factor and a therapeutically effective amount of a cAMP modulator, thereby producing a neurosalutary effect in said subject.

52. A method comprising administering a neurotrophic factor to a subject suffering from a neurological disorder, thereby treating said subject suffering from a neurological disorder.

53. The method of claim 52, further comprising making a first assessment of a nervous system function prior to administering the neurotrophic factor to the subject and making a second assessment of the nervous system function after administering the neurotrophic factor to the subject.

54. The method of claim 53, wherein the nervous system function is a sensory function, cholinergic innervation, or a vestibulomotor function.