WO1992003536A1 - Autotransplantation of schwann cells to promote nervous system repair - Google Patents

Abstract

The present invention relates to methods of promoting nervous system repair comprising transplanting autologous Schwann cells into a region of nervous tissue injury. In particular embodiments of the invention, Schwann cells for autologous grafting may be harvested from a patient in need of such treatment and then propagated in culture. The present invention provides for cell culture methods which yield essentially pure populations of Schwann cells in substantial numbers which may preferably be derived from segments of adult peripheral nerve.

Description

AUTOTRANSPLANTATION OF SCHWANN CELLS TO PROMOTE NERVOUS SYSTEM REPAIR

1. INTRODUCTION

The present invention relates to methods of promoting nervous system repair comprising transplanting autologous

Schwann cells into regions of injury in the central and/or peripheral nervous system. Autologous transplantation of

Schwann cells may be used to facilitate neuronal regrowth, diminish neuron loss secondary to axonal injury, decrease denervation atrophy of muscle, provide for more effective remyelination, and to obviate the problem of immune rejection

10 associated with transplantation of heterologous Schwann cells.

2. BACKGROUND OF THE INVENTION

15 Many of the most clinically devastating consequences of neurologic disease or injury are associated with the extremely limited regenerative capacity of the central nervous system. De la Torre (1982, Brain Research Bulletin :545-552) concludes that regeneration of the nervous system is impeded by a variety of factors, including the absence of

20 a favorable tissue milieu, the development of physical barriers following injury (e.g. gliosis, necrosis, or the excessive production of collagen) , the lack of contact guidance to permit transected axons to connect correctly to distal targets, and inadequate cell and periaxonal vascular

25 supply.

Regrowth in the nervous system has been associated with Schwann cells, which are found in the peripheral nervous system (PNS) and form a yelin sheath around peripheral axons

30 by wrapping their membranous processes around the axon in a tight spiral (Kandel, in "Principles of Neural Science", Kandel and Schwartz, eds., Elsevier Press, N.Y., p. 1.). Oligodendrocytes are the CNS counterparts of Schwann cells, but differ from Schwann cells in that oligodendrocytes

35 envelop several axons, whereas Schwann cells envelop only one axon. The yelin sheath improves nerve conduction velocity. Loss of the myelin sheath (demyelination) drastically alters nerve impulse conduction. Restoration of myelin (remyelination) is one of the simplest types of nervous system repair.

One type of nervous system repair is believed to involve the migration of cells, such as Schwann cells, into areas of injury (Blakemore, 1983, in "Spinal Cord Reconstruction", Kao et al., eds., Raven Press, NY, pp. 281- 291) . Migration and axonal regeneration may require the presence of appropriate extracellular matrix (ECM) protein. The usefulness of peripheral nerve in CNS repair may be limited by the absence of appropriate ECM for the migration of Schwann cells into the CNS. In an attempt to provide ECM suitable for regeneration, de la Torre (1982, Brain Research Bulletin £:545-552) inserted cell-free bovine-derived collagen into transected rat spinal cord and observed evidence of some degree of nerve regeneration through the graft. Similarly, Yannas et al. (1989, J. Cell. Biol. 107:732) inserted a collagen glycosaminoglycan (CG) bridge ensheathed in a silicone tube into a gap in transected rat sciatic nerve, and observed axon regeneration through the CG bridge.

Blakemore (1984, J. Neurolog. Sci. 6 :265-276) created areas of primary demyelination in spinal cord, and then placed autologous peripheral nerve tissue in the subarachnoid space over the lesions; subsequently, remyelination was observed to be limited to axons in the vicinity of blood vessels. Blakemore (Id.) suggested that Schwann cells migrated from the transplanted tissue into the lesion via the perivascular space, but failed to remyelinate axons distant from blood vessels because of the absence within the CNS of suitable extracellular matrix (ECM) . A number of investigators have attempted to improve the regenerative capabilities of the central nervous system. It has been found that providing central nervous system (CNS) lesions with a peripheral nerve environment may promote axonal regeneration of central neurons (David and Aguayo, 1981, Science 214:931-933; Davis et al., 1985, J. Neurosci. 5:2662-2671; Kleitman et al., 1988, Exp. Neurol. 102:298- 306). Vidal-Sanz et al. (1987, Jr. Neurosci. 7:2894-2909) observed that peripheral nerve grafts could be used to guide lesioned optic nerve axons from the eye to the tectum to form terminal synapic contacts in the superior colliculus.

Wrathall et al. (1982, Acta Neuropathol. (Berl.) 57:59-69) had found that implanted non-neuronal cells (thought to be primarily SCs) decreased the time required for axons to grow into nerve grafts placed in severe contusion lesions In cat spinal cord. As evidence of the role of Schwann c xls in regeneration, Berry et al. (1988, Brain Research Bulletin ^^:223~²³¹) observed that optic nerve axons appeared to regenerate into sciatic nerve isographs only in the presence of Schwann cells. Baehr and Bunge (1989, Exp. Neurol. 106:27-40) found that activated Schwann cells appeared to support adult rat retinal ganglion cell survival and axonal regrowth significantly better than quiescent Schwann cell monolayers, suggesting that Schwann cells activated by neurite regeneration (either proliferating or mature Schwann cells) may be more effective in promoting adult rat retina regeneration than are Schwann cells prepared without axon contact. Peripheral nerve grafts placed into transected spinal cord have been shown to induce axonal regeneration (Richardson et al., 1980, Nature 284:264-265) that otherwise would not have occurred (Ramon et al., 1928, Hafner Publishing Co. , London & New York) . Kromer and Cornbrooks (1985, Proc. Natl. Acad. Sci. 82:6330-6334) produced lesions in the septo-hippocampal system of adult rats, bridged that gap with collagen-supported cultures consisting of Schwann cells, extracellular matrix, degenerating neuronal processes and myelin, and observed evidence of regeneration of CNS neurons in vivo.

The ability of Schwann cells to express molecules with neurotrophic activities and/or mediating neurite attachment and growth is well established (Bunge and Bunge, 1983, TINS, 6:499-505; Muir et al., 1989, Neurochem. Res., 14:1003-1012; Rende et al., 1990, Soc. Neurosci. Astr. 16:807) . Schwann cells in the distal nerve stump re-express both nerve growth factor (NGF) (Lemke and Chao, 1988, Development 102:499-504; Heumann et al., 1987, J. Cell Biol. 104:1623-1631) and NGF receptors (NGFr) (Taniuchi et al., 1986, Proc. Natl. Acad. Sci. U.S.A. 83:4094-4098; Taniuchi et al. , 1988, J. Neurosci. 8:664-681). The secreted NGF binds to low-affinity NGFrs on the surface of Schwann cells and to high-affinity NGFrs on the tip of extending neurites, attracting neurites toward more distally located Schwann cells, and thus toward the target tissue (Taniuchi et al., 1988, J. Neurosci. 8:664-681). The NGF-NGFr complex is then internalized and transported to the cell body to exert its neurotrophic activity (Johnson et al., 1987, J. Neurosci. 7:923-929). Schwann cells from transected nerves and outgrowing neurites also re-express cell adhesion molecules (CAMs) , including nerve-CAM (N-CAM) and LI (Nieke and Schachner, 1985, Differentiation 30:141-151; Daniloff et al., 1986, J. Cell Biol. 103:929-945; Martini and Schachner, 1988, J. Cell Biol. 106:1735-1746). Both molecules may promote adhesion between Schwann cells and axons (Seilheimer and Schachner, 1988, J. Cell Biol. 107:341-351), facilitate the extension of regenerating neurites on the surface of Schwann cells (Daniloff et al., 1986; J. Cell Biol.

103:929-945; Bixby et al., 1988, J. Cell Biol. 107:353-361;

Kleitman et al., 1988, Exp. Neurol. 102:298-306; Seilheimer and Schachner, 1988, J. Cell Biol. 107:341-351) and participate in myelination processes (Wood et al., 1990, J. Neurosci. 10:3635-3645). Finally, Schwann cells secrete proteases and proteases inhibitors. Both types of molecules are involved in providing a better anchorage of neurite membranes to the environment, thus promoting the advance of growth cones (Monard, 1988, TINS 11:541-544). Collagen grafting into spinal cord lesions has been tried by several investigators (de la Torre et al., 1982, Brain Res. Bull. 9:545-552; Gelderd, 1990, Brain Res. 511:80-92; Marchand and Woerly, 1990, Neuroscience 36:45- 60) , all of whom reported some axonal ingrowth. In one of these studies (Gelderd, 1990, Brain Res. 511:80-92), for example, collagen allowed ingrowth of centrally derived axons, although some entered along blood vessels, a very common route for ingress of axons that become yelinated by SCs. This collagen was more three-dimensional than the ammonia-gelled collagen we utilized; collagen preparations vary in organization (Bunge, et al., 1987, In: Progress in Brain Research; Seil, Herbert & Carlson (eds.), Elsevier, 71:61-74) and ability to support central neurite growth (Kleitman et al., 1988, J. Neurosci. 8:653-663) depending upon the gelling procedure employed.

The Martin group concluded that many axons in their SCs grafts originated in dorsal root ganglia based on the neuropeptide content of the axons (Martin et al., 1991, Neurosci. Lett. 124:44-48). Kleitman et al. (1988, J. Neurosci. 8:653-663) observed that Schwann cell surfaces but not the ECM organized by Schwann cells supported neurite outgrowth from embryonic rat retina. It has further been suggested that Schwann cells may elaborate a factor which promotes neurite outgrowth; Schwann cells have been shown to synthesize nerve growth factor (NGF; Richardson and Ebendal, 1982, Brain Res. 24j5:57-64; Bantlow et al., 1987, EMBO J. 6:891-899; Heu ann et al., 1987, Proc. Natl. Acad. Sci. U.S.A. 84:8735-8739) .

Thanos et al. (1989, Eur. J. Neurosci. 3₁:19-26) observed that severed sciatic nerve produced a factor bearing similarities to a recently identified brain derived neurotrophic factor (BDNF) .

Citation of a reference herein is not meant to be construed as an admission that such reference is available as prior art to the instant invention.

3. SUMMARY OF THE INVENTION The present invention relates to methods of promoting nervous system repair comprising transplanting autologous Schwann cells into a region of nervous tissue injury. In particular embodiments of the invention, Schwann cells for autologous grafting may be harvested from a patient in need of such treatment and then propagated in culture. The present invention provides for cell culture methods which yield essentially pure populations of Schwann cells in substantial numbers which may preferably be derived from segments of adult peripheral nerve. In a preferred aspect involving the use of cultured cells, cells are passaged for a minimal number of times before use. According to the invention, autologous Schwann cells may be applied directly to areas of nerve injury or disease or may, alternatively, be incorporated into a vehicle comprising support matrix material for transplantation. In particular embodiments of the invention, autologous Schwann cells may be comprised in a prosthetic device which may be used to bridge lesions in the peripheral or central nervous systems.

The autologous transplantation methods of the invention may be used to promote neuronal regrowth, reduce neuronal loss secondary to axonal injury, decrease denervation atrophy of muscle, and to obviate the problem of immune rejection associated with transplantation of heterologous Schwann cells. Furthermore, the methods of the present invention may be applied to the treatment of central as well as peripheral nervous system lesions.

4. DESCRIPTION OF THE FIGURES Figure 1. Photomicrograph of a substantially pure population of human Schwann cells prepared from adult peripheral nerve.

Figure 2. Cable cross-sectional area (A) and number of myelinated axons (B) of transverse sections taken 2,4,6 and 8 mm from the proximal suture in empty, F-CD-φ , F-40,

F-80, F-120, CD-80 channels and in sciatic nerve autografts 3 weeks postimplantation. Data represents means + SEM. The

Games-Howell test was used to test statistical significance between treatments. *:p<0.05 indicates statistical significance between either sciatic nerve autografts, F-40,

F-80 or F-120 channels as compared to SC-0 and empty channels and **:p<0.05 indicates statistical significance between sciatic nerve autografts and F-120 channels.

Figure 3. Light micrographs of toluidine blue-stained transverse sections taken 4 mm distally to the proximal suture in (A) CD-80, (B) F-80 channels and (C) sciatic nerve autografts 3 weeks postimplantation. A) Numerous reactive β cells are seen among Matrigel patches (MP) . Note the absence of myelinated axons. B and C) Myelinated axons (MA) with associated Schwann cells are seen. Note the presence of degenerating figures (DF) in the nerve autografts. In A, B, and C, scale bar = 10 μm.

Figure 4. Light micrographs of toluidine blue-stained transverse sections taken 4 mm distally to the proximal suture in (A) empty, (D) F-CD-φ and (G) F-80 channels 3 weeks postimplantation. Note the increase in cable surface area between the regenerated cables. At higher magnification, light micrographs of transverse sections taken respectively 4 and 8 mm distally to the proximal suture in (B, C) empty, (E,F) F-CO-φ and (H,I) F-80 channels showing nerve microfascicles with blood vessels (BV) and numerous myelinated axons (MA) . At the channels' midpoint, the myelinated axon density was lower in F-CO-φ channels (E) as compared to empty (B) and F-80 channels (H) . At the distal nerve stump level, myelinated axons are observed in both empty (C) and F-80 channels (I) , but not in F-CO-φ channels (F) . Note the difference in microfasciculation between empty and F-80 channels. In A, D and G, scale bar = 200 μm and in B, C, E, F, H, and I, scale bar = 5 μm. Figure 5. Transmission electron micrographs taken at the midpoint of a cable regenerated in a SC-80 channels 3 weeks post-implantation. A) Both unmyelinated (UMA) and myelinated (MA) axons are associated with Schwann cells. At higher magnification (B) , myelinated (MA) and unmyelinated (UMA) axons are seen with Schwann cells (SC) with their typical basal laminae (BL) and collagen fibrils (CF) . In A, scale bar = 1 μ and in B, scale bar = 0.5 μm.

Figure 6. Number of blood vessels per mm 2 of cable surface area in transverse sections taken 2, 4, 6 and 8 mm from the proximal suture in empty, F-CO-φ , F-40, F-80, F-120 and CD-80 channels and in sciatic nerve autografts 3 weeks post-implantation. Data represents means + SEM. The Games- Howell test was used to test statistical significance between treatments. **:p<0.05 indicates statistical significance between F-CO-φ channels and either F-40, F-80, F-120, CD-80 or empty channels, *:p<0.05 indicates statistical signficance between sciatic nerve autografts and either type of channel.

5. DETAILED DESCRIPTION OF THE INVENTION

According to the present invention, a patient's own cells may be employed to foster regeneration of either the peripheral or the central nervous system. In particular, small segments of the patient's own peripheral nervous system ay be obtained by biopsy and the cell type known to favorably influence neural regeneration (the cells of Schwann) , may be separated from a variety of other cells that are present in the peripheral nerve trunk. The present invention provides, for the first time, methods of preparing essentially pure populations of Schwann cells from adult nerve; prior to the present invention, substantially purified Schwann cell populations could be prepared from embryonic but not from mature peripheral nerve. After separating the cells of Schwann, a variety of agents may be used to expand their numbers by causing them to proliferate in tissue culture dishes. In preferred embodiments of the invention, after substantial proliferation has been achieved, the cells may be utilized either in suspension or may be comprised in a cell- containing prosthesis for insertion into damaged areas of the central or peripheral nervous system. The construction of this prosthesis may involve the combination of these cultured cells with a variety of organic or inorganic materials to assemble a prosthesis of the correct shape to promote regeneration of specific parts of the nervous system. The prostheses may then be surgically implanted into the nervous system, either to enhance remyelination of demyelinated nerve fibers, to influence neural survival or to enhance the regeneration of nerve fibers. Alternatively, the cells may be treated while in tissue culture by methods which allow the introduction of new genetic material in order to provide them with added growth potential or with the biological activities which they may not otherwise express, thereby enhancing the regenerative or functional activity of nerve cells. For example, Schwann cells may be infected with a retrovirus carrying the genetic code for the production of the enzyme tyrosine hydroxylase, allowing the cells to produce and release L-dopa, a compound which acts to relieve the symptoms of Parkinson's disease. If transplanted into the striatu , these cells may alleviate the movement disorder of this common neurological disease. Because the cells of Schwann appear to be active in influencing the regeneration in both the central and peripheral nervous system neurons, the cellular prostheses of the invention may be useful in the treatment of a variety of central (as well as peripheral) nervous system lesions.

The advantages of the present invention include the following. By providing for nerve prostheses comprising autologous cells, the present invention makes available a quantity of autologous graftable material surpassing the amount of nerve tissue available for grafting in a particular patient. Furthermore, the nerve prostheses of the invention comprise few or no fibroblasts, thereby avoiding the formation of fibroblast-associated scar tissue, which is believed to interfere with nerve regeneration and repair. For purposes of clarity of presentation, and not by way of limitation, the detailed description of the invention is divided into the following subsections:

(i) clinical applications of autotransplantation of Schwann cells; (ii) harvesting cells for autotransplantation; (iii) vehicles for autotransplantation; (iv) autotransplantation of cells for nervous system repair; and (v) a specific embodiment: the use of 5 autotransplantation in the repair of a cauda equina lesion.

5.1. CLINICAL APPLICATIONS OF AUTOTRANS- PLANTATION OF SCHWANN CELLS _Q The present invention may be utilized in the treatment of any nervous system injury, which may be construed to refer to any disease or disorder which results in either a disconnection of axons, a diminution or degeneration of neurons, or demyelination. Nervous system lesions which may 5 be treated in a patient (including human and non-human mammalian patients) according to the invention include, but are not limited to, the following lesions of either the central or peripheral -lervous systems:

(i) traumatic lesions, including lesions caused by physical injury or associated with surgery, for example lesions which sever a portion of the nervous system, or compression injuries; (ii) ischemic lesions, in which a lack of oxygen in a portion of the nervous system results in neuronal injury or death, including cerebral infarction or ischemia, or spinal cord infarction or ischemia; (iii) malignant lesions, in which a portion of the nervous system is destroyed or injured by malignant tissue which is either a nervous system associated malignancy or a malignancy derived from non-nervous system tissue; (iv) infectious lesions, in which a portion of the nervous system is destroyed or injured as a result of infection, for example, by an abscess or associated with infection by human immunodeficiency virus, Lyme disease, tuberculosis, syphillis, herpes zoster or herpes simplex virus infection; (v) degenerative lesions, in which a portion of the nervous system is destroyed or injured as a result of a degenerative process including, but not limited to, degeneration associated with Parkinson's disease, Alzheimer's disease, Huntington's chorea, or amyotrophic lateral sclerosis; (vi) lesions associated with nutritional diseases or disorders, in which a portion of the nervous system is destroyed or injured by a nutritional disorder or disorder of metabolism including, but not limited to; vitamin B12 deficiency, folic acid deficiency, Wernicke disease, tobacco-alcohol amblyopia, Marchiafava-Bignami disease (primary degeneration of the corpus callosum) , and alcoholic cerebellar degeneration; (vii) neurological lesions associated with systemic diseases, including, but not limited to, diabetes (diabetic neuropathy, Bell's palsy) , systemic lupus erythematosis, carcinoma, or sarcoidosis; (viii) lesions caused by toxic substances including alcohol, lead, or particular neurotoxins; and (ix) demyelinated lesions in which a portion of the nervous system is destroyed or injured by a demyelinating disease including, but not limited to, multiple sclerosis, human immunodeficiency virus-associated myelopathy, transverse myelopathy of various etiologies, progressive multifocal leukoencepholopathy, and central pontine myelinolysis.

5.2. HARVESTING CELLS FOR AUTOTRANSPLANTATION According to the present invention, Schwann cells for autotransplantation may be harvested from regions in which removal of nerve tissue is preferably associated with as little clinical consequence as is practical. If the patient who is to receive the autotransplant has damaged tissue as a result of injury or surgery, Schwann cells may be harvested from the damaged tissue. Alternatively, nerve tissue to be used as a source of Schwann cells may be harvested from peripheral nerve, and preferably sensory nerve, which may be obtained by a surgical procedure such as nerve biopsy.

Potential sources of peripheral nerve which may be used according to the invention include, but are not limited to, the sural nerve of the ankle, the saphenous nerve, or the brachial or antibrachial nerve of the upper limb, and are preferably sensory nerves.

Additional cell types may also be used for autotransplantation according to the invention, including, but not limited to, oligodendrocytes, retinal cells, glial cells (e.g., olfactory glial cells or glial cells from enteric nerve plexices) , or astrocytes. Such cells may be obtained by biopsy or may be recovered from resected tissue.

5.2.1. CULTURING SUBSTANTIALLY PURE POPULATIONS OF SCHWANN CELLS

According to preferred embodiments of the invention, cells harvested from a patient for autotransplantation may be allowed to proliferate in culture so as to expand their numbers or so as to activate the cells prior to transplantation. Cells harvested from a patient may be cultured together with other cell types, or may be cultured so as to select for the growth or survival of one specific population of cells.

In a specific embodiment of the invention, substantially pure cultures of Schwann cells may be obtained using the following procedure. A portion of peripheral nerve may be cultured as an explant having preferably dimensions of about 1 X 1 mm in Dulbecco's Modification of Eagle's medium (MEM; or a similar basic culture medium) supplemented with 10% fetal bovine serum on plastic tissue culture substratum in a 7% CO atmosphere at 37 C. During the course of about 4 to 5 weeks, fibroblasts may be observed to grow out from the explants, which may be periodically transferred to fresh culture containers at a frequency of about once a week for a period of about 3 to 5 weeks. Periodic dissociation and subculture of these explants has shown that the number of fibroblasts within the explant decline as the explant is repeatedly transferred to fresh culture containers, each time leaving fibroblasts behind which have exited the explant and formed an outgrowth. Following substantial exit of fibroblasts, the explants may then be dissociated using, for example, about 1.25 units/ml of dispase and about 0.05 percent collagenase with 15 percent fetal calf serum and 50 mM HEPES buffer (pH 7.4) in Dulbecco's MEM. The dissociated cells should be greater than about 90 percent Schwann cells. Following dissociation, the cells may be used for autotransplantation or, preferably, may be allowed to proliferate in culture. Proliferation may be induced by culturing the purified Schwann cells in the presence of co- cultured axons or glial growth factor (GGF) and forskolin at concentrations of about 20 μg/ml and 2μM, respectively, preferably on a polylysine over plastic substratum. Other agents which may induce Schwann cells to proliferate and which may be used according to the invention include, but are not limited to cholera toxin, laminin, fibronectin, nerve growth factor, platelet derived growth factor, fibroblast growth factors, and transforming growth factor beta; other proliferation inducing agents may become known in the art and are envisioned by the present invention.

It may be desirable to evaluate cultured Schwann cells (produced by the method described supra or any method known in the art) for the maintenance of differentiated Schwann cell functions, including, but not limited to, the ability to associate with sensory neurons in culture or the ability to produce myelin. This may be accomplished by establishing substantially pure populations of sensory neurons in culture, seeding the Schwann cells onto these neuronal populations and observing the cell interactions over a period of several weeks. 5.2.2. ADDITIONAL METHODS FOR SELECTING CELLS FOR AUTOTRANSPLANTATION

In addition, cells which may be used for autotransplantation may be selected from mixed populations of cells using cell selection techniques widely known in the art. Such techniques may include, but not be limited to, the selective proliferation of the cell type of interest using specific growth factors or defined culture conditions, the selection of cells of interest by exposing the cells to antibody which selectively binds to the cell type of 0 interest, such that antibody binds to the cell type of interest and such binding is detectable, using, for example, rosetting techniques, magnetic beads, colorimetric techniques, fluorescence techniques including fluorescence activated cell sorting, or affinity chromatography using, for 5 example, a Staphylococcus protein A surface to collect antibody-bound cells, to name but a few of the techniques available. Alternatively, the cell type of interest may be selected by selectively destroying cell types which are not of interest, using, for example, compounds or culture 0 conditions which inhibit the growth or are toxic to undesirable cell types, or by using antibodies to destroy undesirable cell types by, for example, complement mediated cell lysis or antibody dependent cell-mediated cytotoxicity.

For example, and not by way of limitation, Schwann 5 cells may be selected using antibodies to the NGF receptor, such as the antibody disclosed in Peng, et al. (1982, Science 215: 1102-1104) . Alternatively, fibroblasts from peripheral nerve could be selectively destroyed, either via antibody directed toward a fibroblast specific antigen such as, for _Q example, the thy-1 antigen and may either be separated from Schwann cells by techniques including, but not limited to, fluorescence activated cell sorting or complement mediated lysis, or may be selectively destroyed on the basis of higher proliferative activity relative to Schwann cells by agents 5 selectively toxic to actively dividing cells, including, but not limited to, methotrexate and cytosine arabinoside or fluorodeoxyuridine, or may be destroyed during passaging using active complement in the presence of Thy-1 antigen.

5.3. VEHICLES FOR AUTOTRANSPLANTATION

According to the present invention, cells may be autotransplanted into a nervous system lesion directly after harvesting or, alternatively, after culturing. The cells may preferably be comprised in a pharmacologically suitable carrier such as a physiologically compatible vehicle, including, but not limited to, culture medium, including Dulbecco's Modified Eagle's Medium, RPMI 1640, Fisher's, Iscove's, or McCoy's medium to name but a few; or preferably, a solid, semisolid, gelatinous, or viscous support medium including, but not limited to, collagen, collagen- glycosaminoglycan, fibrin, polyvinyl chloride, polyamino acids such as polylysine or polyornithine, hydrogels, agarose, dextran sulfate or silicone, to name but a few. The support medium may, in specific embodiments, comprise growth factors or relevant extracellular matrix proteins, such as laminin. For example, gelatinous vehicles include, but are not limited to, collagen, collagen-glycosaminoglycan, hydrogels, agarose, dextran sulfate or silicone.

If a solid, semisolid, or gelatinous support is used as a vehicle, cells may be introduced into a liquid phase of the vehicle which is subsequently treated such that it becomes more solid. For example, cells may be added to unpolymerized vehicle which is then caused to polymerize. In a specific embodiment of the invention, not by way of limitation, cells may be added to type I collagen solubilized in acetic acid in H₂0. When this mixture is brought to neutral pH by addition, for example, of NaOH, and to isotonicity by the addition of salts it should gel in several minutes time at room temperature. Prior to gelling, cells can be intermixed into the mixture and thereby may be dispersed throughout the matrix after gelling. If the vehicle permits, cells may simply be mixed with the vehicle. In embodiments of the invention in which the vehicle has a solid structure, the vehicle may be molded into a shape which may conform to the shape of the lesion. In embodiments of the invention in which the vehicle does not have a solid structure, the vehicle comprising cells may either be applied directly to the lesion or may be partially or completely enclosed in a second material which may allow the vehicle to retain a particular shape while permitting desirable regeneration through the vehicle. The second material may be applied before or after insertion of the vehicle into the patient. For example, and not by way of limitation, a semisolid vehicle comprising autologous cells may be applied to a spinal cord lesion which extends to the surface of the cord, creating a gap. A second substance, such as a plastic or hydrogel, may then be placed at the surface of the cord over the gap so as to confine the vehicle and cells within the lesion. In other specific examples, a solid tube filled with semi-solid vehicle comprising autologous cells may be used to bridge a gap in a peripheral nerve, an optic nerve, or other parts of the nervous system.

5.4. AUTOTRANSPLANTATION OF CELLS FOR NERVOUS SYSTEM REPAIR

According to the present invention, autologous cells may be introduced into a nervous system lesion using any method known in the art. In preferred embodiments of the invention, the autologous cells comprise Schwann cells or more preferably, substantially pure populations of Schwann cells.

Methods of introducing autologous cells include, but are not limited to, surgical techniques which expose the neurological lesion and permit introduction of autologous cells in a suitable vehicle, as well as techniques which can inject autologous cells in suitable vehicle into a neurological lesion without exposing the lesion.

For example, and not by way of limitation, if a portion of the spinal cord has been destroyed by a crush injury, surgical techniques including laminectomy may be used to expose the affected region of the spinal cord. After contusion lesions the damaged central region of the cord is often removed by macrophage action, leaving a fluid-filled cyst. This cyst may be visualized by imaging techniques (such as ultrasound) and an injection of cells plus vehicle may then be made directly into the cystic defect. This injection may contain autologous Schwann cells comprised in a semi-solid vehicle such as solubilized collagen (see supra) which should solidify shortly after mixing. The initially semi-solid character of the vehicle permits effective filling of the lesioned area; subsequent solidification may obviate the need to otherwise fix the vehicle in place.

In another example, where a lesion occurs in a structurally clearly defined nerve, such as the facial nerve (cranial nerve VII) or the optic nerve (cranial nerve II) a solid vehicle (such as a cylindrical sleeve comprising a solid polymer exterior and an interior comprising a gel and autologous cells preferably Schwann cells) may be molded to conform in size and shape to a segment of the affected nerve, thereby forming a nerve prosthesis. The affected nerve may then be surgically exposed and the lesioned area removed and replaced with the nerve prosthesis, which may be sutured or otherwise fixed in place.

In yet another example, if a neurologic lesion occurs within another structure, including, but not limited to, an infarcted area or area of demyelination deep within the spinal cord or cerebral hemisphere, autologous cells in suitable, preferably liquid, vehicle may be injected into the affected area, preferably using radiologic guidance, being careful to limit the injected material to a volume easily accommodated by the tissue so as to avoid increased pressure within the tissue and/or intracranially.

5.5. SPECIFIC EMBODIMENT: THE USE OF AUTOTRANS¬ PLANTATION OF SCHWANN CELLS IN THE REPAIR OF A CAUDA EQUINA LESION

The following is a specific, non-limiting example of how the present invention may be utilized to influence the regeneration of the nervous system in a specific human patient. For example, the present invention may be used to treat a patient who has received a severe vertebral injury resulting in the fracture of vertebral bone which has severely crushed that portion of the nervous system within the vertebral canal referred to as the cauda equina. If severe, such an injury may result in motor paralysis, sensory loss, and loss of visceral functioning. Furthermore, the crushing injury may be expected to cause a severe fibrosis of the nerve roots within the cauda equina. Subsequent to the injury these roots may no longer effectively conduct nerve impulses; not only is the patient paralyzed below the level of injury but, in addition, the substance of the muscles supplied by this part of the nervous system begins to be lost as the muscle mass of the legs and lower trunk undergoes progressive disuse and denervation atrophy. There is no known effective treatment for this and a variety of other kinds of nerve injury that result from this type of trauma. Similar considerations may apply if the injury were to the spinal cord itself.

The patient may be examined to ascertain, utilizing both neurological and modern imaging techniques, that the neural injury is to the cauda equina. According to this specific embodiment, the spinal cord itself is not damaged; rather, damage is limited to the nerve roots of the cauda equina. A biopsy of the patient's peripheral nerve may be taken by removing, for example, a 3-5 inch segment of the sural nerve posterior to the lateral malleolus on the lateral aspect of the ankle. After removal from the body the nerve segment may be placed immediately in sustaining fluids such as a tissue culture medium and cooled. Subsequently, the area of incision may be sutured. The patient may expect to suffer a small area of anesthesia at the lateral aspect of the foot, but there should be no motor deficiency, inasmuch as this is primarily a sensory nerve at the level that the nerve has been removed. For this particular patient, the surgery should have no immediate consequences inasmuch as the patient may already be anesthetic below the level of the cauda equina lesion. If the lesion in the cauda equina region involved nerve roots that provide nerve fibers to the sural nerve it may be desirable to use an alternate biopsy site in as much as the sural nerve may have undergone degenerative changes secondary to the injury. In this case, it may be advantageous to use a cutaneous branch of one of the sensory nerves of the arm (see supra) .

The segment of nerve may be immediately taken to a tissue culture facility and treated in one of several alternative ways. The nerve is first stripped of its epineural lining with the use of very fine forceps. Preferably, the nerve may be cut into a multitude of small pieces approximately 0.5mm in diameter which may be placed in tissue culture on plastic surfaces. These may be maintained under standard tissue culture conditions in a medium comprising a minimal essential medium and 10% serum. Alternatively, the fragments of nerve may be treated with digestive enzymes to allow direct dissociation of the cells. The mixed cellular elements obtained may then be propogated in tissue culture.

If the approach is to undertake the culture of fragments of the nerve without dissociation, these fragments may be carried in the culture dishes as described (supra) for approximately one week. During this time degenerative changes may occur in the nerve resulting from its amputation from connection to nerve cell bodies. Axons, as well as the surrounding myelin sheaths, may degenerate, but the Schwann cells which have been related to myelin sheaths (as well as Schwann cells which would be related to unmyelinated fibers) may, during this period of time, undergo a gradual increase in number. Periodic transplantation of these fragments of nerve to new culture dishes may allow many of the fibroblasts to escape from the explant, leaving the explant enriched in Schwann cells. After several weekly sequential passages of these nerve fragments, it may be noted that the outgrowth is progressively enriched in the Schwann cells. These fragments may then be harvested by detaching them from the culture dish and exposing the cells to proteolytic enzymes which have little effect on the living cells but which are effective in dissociating the connective tissue framework within these fragments. From this dissociation several thousand to several hundred thousand Schwann cells may be obtained.

This Schwann-cell enriched population may then be expanded from several thousands of cells into populations which may number in the hundreds of thousands to several million. A number of methods may be used to stimulate Schwann cell proliferation, including a) the use of a protein factor called glial growth factor (used in conjunction with forskolin; b) the use of fragments of axolem a obtained by fractionation techniques; c) the use of a variety of growth factors such as transforming growth factor beta, or acidic or basic fibroblast growth factor which are known to be effective on Schwann cells of certain species, especially when combined with agents which raise intracellular cyclic AMP; and d) the use of certain molecules as substratum (for example it has been shown that certain Schwann cells proliferate in the presence of serum, when they are grown on substratum containing the extracellular matrix molecule laminin, but not when they are grown on collagen substratum) . During their proliferation phase the cells may be optionally treated to introduce new genetic material to enable them to produce desired proteins. For example, it may be possible to introduce a gene encoding a specific cell adhesion molecule which is known to promote nerve fiber growth, or a gene for a specific growth factor molecule (such as the gene for neurotrophin-3 (NT-3)) believed to sustain neuronal health. It is anticipated that under ideal conditions expansion may occur at the rate of 1-2 doublings per week so that it may take several weeks to obtain the required number of cells. During this time, the purity of the cells may be monitored by utilizing antibodies which specifically mark Schwann cells (such as antibodies to S100 protein) and by undertaking complement mediated cell lysis of non-Schwann cell contaminants (as described supra) at the time when the cells are transferred from one dish to another as the numbers expand.

Next, the cultured cells may be combined with an encasing matrix, thereby providing a tubular prosthesis which has the correct shape and length to insert into an area of neural injury. Various organic polymers (poly amino acids) or hydrogels are examples of materials known to be compatible with body tissues, and which are also known to provide hospitable housing for live cells, as well as other materials discussed in Section 5.3, supra, may be used. n the cauda equina patient, a considerable length of a tubular prosthesis may be needed which would provide an envelope which is porous to allow nutrients to be exchanged between the external aspect of the tubular prosthesis and the enclosed Schwann cells. The envelope may desirably have characteristics which would accept suturing. An estimate of the length of the prosthesis may be undertaken, knowing the number of nerve roots that are involved in the crush injury and the overall dimensions of the lesion which contain these crushed roots. For example, if 12 nerve roots were to be repaired over a distance of two inches, a total of 24 inches of prosthesis several millimeters in diameter would be required, much more material than would be available from the use of an autograft in which the sural nerve would be directly used as an implant into the region of lesion. After the length and diameter of the prosthesis has been constructed, the patient may be prepared for surgery, and the injury site may be exposed by laminectomy. At the site of nerve root crush the entangled roots and the fibrous scar in which they are embedded may be physically removed. The individual nerve roots may then be identified so that each nerve root as it egresses from the cord above would be connected to the proper nerve root as it courses toward its destination distal to the site of injury below. The availability of extensive prosthetic material according to the invention may allow the prosthesis to be sewn between the divided central and peripheral aspect of the nerve root. In this way the region of injury may be removed from the body and a bridge inserted to replace the region of injury for each of several critical nerve roots. After closing the dura over this repair site, the patient may then be expected to experience a relatively long period of recovery in which nerve fibers may grow across the region of bridging to reach targets which would have previously been denervated. In particular, ventral roots may be observed to effectively reach the distal segment and to follow that distal nerve segment to the regions of the muscles which would have been denervated. It may be expected that this process may take several months or longer, inasmuch as the distance of nerve growth required in this case may be between 10 and 15 inches and the rate of nerve growth as it would be observed in standard peripheral nerve injury is at the rate of about one inch per month. There are additional specific embodiments in which this type of Schwann cell containing prosthesis may be used to foster nerve regeneration in which the distances for regeneration may be much shorter. Of particular interest are those patients in which spinal cord injury has occurred in the cervical region. Many of these patients have partial or complete loss of motion within the muscles of their arms. The distance between the site of injury and the portion of the spinal cord which may be intact and able to signal movement within the upper extremity is relatively short; in certain cases only one or two inches. The ability to obtain some regrowth to attain some control over motor neurons distal to a cervical site of injury may be extraordinarily beneficial to a patient who may have little useful movement in certain of the arm muscles. In these cases, the lesion often results in the loss of central cord tissue which becomes a fluid-filled cystic space. This space would provide a repository for the cellular implant, which in this case may not require an enveloping sleeve in as much as the retained peripheral cord tissue would contain the injected material.

6. EXAMPLE: ISOLATION AND CHARACTERIZATION OF SCHWANN CELLS DERIVED FROM ADULT NERVE

Adult rat sciatic nerves were removed from adult Sprague-Dawley rats and enzymatically dissociated with 1.25 units/ml of dispase and 0.05 percent collagenase with 15 percent fetal calf serum, and 50 mM HEPES buffer (pH 7.4) in Dulbecco's MEM either immediately or following a period of "in vitro Wallerian degeneration" as -1 X 1 mm explants in tissue culture. Immediate dissociation appeared to yield populations of cells containing less than 20% Schwann cells. Contaminating cells were mostly endoneurial fibroblasts. Conversely, serial explanation of explants over 4-5 weeks was found to decrease fibroblast contamination as these grew out fro the explants. After these explantations, dissociation yielded up to 98% pure Schwann cells. These techiques have been found to be applicable to human derived tissue with similar results. (Figure 1) . Schwann cells from this source retained functional capacity in that they associated with sensory neurons in culture and formed myelin under appropriate conditions. These cells proliferated in response to axonal mitogens as well as in the presence of GGF and forskolin. It has further been observed that serial explantation appears to yield a purer population of Schwann cells than primary dissociation.

EXAMPLE: INDUCTION OF AXON GROWTH INTO SCHWANN CELL IMPLANTS GRAFTED INTO LESIONED ADULT RAT SPINAL CORD

Polymerized collagen rolls enclosing Schwann cells (SCs) raised in culture were grafted into cystic cavities formed after lesioning the thoracic spinal cord of adult rats. Axons were already present within the graft by 14 days after implantation and both ensheathed and myelinated axons were numerous by 28 days. This axonal ingrowth was maintained over longer survival periods. The axons within the graft always appeared related to Schwann cells. Acellular collagen rolls did not show axonal ingrowth. These Schwann cell-collagen implants resemble peripheral nerve grafts in their ability to induce axonal regeneration into the graft.

We report here preliminary results of experiments designed to achieve the periphral nerve graft effect in spinal cord with only purified populations of SCs on collagen substrata. The rationale was to a) design a SC graft that can be prepared by tissue culture techniques to replace peripheral nerve in experimental studies of regeneration of central axons and b) determine if transplanted SCs induce growth of spinal cord axons _in vivo. Purified populations of SCs were obtained from dorsal root ganglia of 16-day rat embryos according to the method of Wood (1976, Brain Res. 115:361-375) . Folymerized collagen substrata carrying SCs either dissociated (Porter et al., 1986, J. Neurosci. 6:3070-3078) or predifferentiated to form basal lamina tubes and myelin sheaths (Eldrige et al., 1989, J. Neurosci. 9:625-638) but secondarily separated from nerve cell bodies to create bands of Bungner (Paino and Bunge, 1990, Soc. Neurosci. Abstr. 16:1282) were rolled (Kuhlengel et al., 1990, J. Comp. Neurol. 293:63-73) to obtain a cylindrically shaped graft (about 1mm in diameter and 4-6mm in length) . The SCs were enclosed within these spiralled structures. These rolls were coated with fresh rat plasma to stablize the graft by precluding it from unrolling and possibly to enhance their adhesion to host tissue after grafting.

Spinal cord lesions were produced in adult female Sprague-Dawley rats by the photochemical method (Cameron et al., 1990, Exp. Neurol. 109:214-223; Prado et al. , 1987, J. Neurosurg. 67:745:753; Watson et al., 1986, Brain Res. 367:296-300). Briefly, an argon-dye laser (Innova 70 System, Coherent, Inc. , Palo Alto, CA) beam lOOmW was focused transversely over the T8 lamina of the vertebral column immediately after a photosensitizing agent (rose bengal dye) had been injected intravenously. At 150 seconds, 50% duty cycle irradiation produced a photothrombotic lesion that consistently led to cystic degeneration within the dorsal half of the spinal cord; an elliptical shaped cystic cavity developed, reaching its rostrocaudal extent of 4-6mm after one week (Cameron et al., 1990, Exp. Neurol. 109:214-223). The lesion continued to develop, manifesting a somewhat defined glial border after two weeks (Salvatierra et al., 1990, Soc. Neurosci. Abstr. 16:1282) . The SC cylindrical grafts were implanted into these cystic cavities at 5 or 28 days after lesioning. Laminectomies were performed at T7 and T8 and a small longitudinal incision was made dorsolaterally in the dura and cyst wall on the left side. Two grafts were introduced through this slit and positioned with their axes oriented mainly in the rostrocaudal direction. No immunosuppression was used. Animals were perfused with fixative (Kuhlengel et al., 1990, J. Comp. Neurol. 293:74-91) at 14, 28, and 90 to 180 days after implantation and processed for light microscopy (semi-thin plastic sections stained with toluidine blue or paraffin sections stained with the Sevier-Munger silver stain) and for electron microscopy. Data was obtained from 20 animals receiving SCs and 3 control animals receiving collagen rolls with no cells. Additional control data from non-implanted but photoche ically lesioned animals were obtained from 4 animals prepared for this study as well as from other animals investigated in previous and other current studies (Cameron et al., 1990, Exp. Neurol. 109:214-223; Prado et al., 1987, J. Neurosurg. 67:745-753; Salvatierra et al., 1990, Soc. Neurosci. Abstr. 16:1282).

The grafts filled the lesion cavity and closely apposed much of the cavity laterally. The graft reached at least portions of the rostral and caudal ends of the cavity. Macrophage accumulation sometimes prevented the graft from contacting host tissue.

Accellular collagen grafts showed some invasion by fibroblast-like cells at 28 days after implantation, but myelinated or ensheathed axons were not detected between collagen layers by light or electron microscopy. In contrast, axonal ingrowth was abundant by the same time period in SC-containing grafts. Such axonal ingrowth could be detected by 14 days after SC implantation, consisting mostly of SC-ensheathed (unmyelinated) axons and a few thinly myelinated axons. The number of axons at this 14-day period varied substantially between different grafts, indicating that axons had only recently started to enter the implant. By 28 days much of the axonal ingrowth had occurred; most of the grafts consistently showed large numers of myelinated as well as unmyelinated axons.

At 90 to 180 days after implantation, axonal ingrowth of the same magnitude as that present at 28 days was observed. Myelinated axons were outnumbered by those only ensheathed by SCs at all time periods. Axons within the implant always appeared in relation to SCs, either in contact with them or surrounded by their basal lamina. As seen by silver staining, the axons followed parallel paths, at least for some distance; the spiralled configuration of the collagen may have constrained many of them to a roughly linear trajectory along the graft. Only at the graft-host interface was profuse axonal branching observed. Astrocytic gliosis at the interface was minimal.

These results demonstrate that purified populations of SCs induced substantial axonal growth when grafted into lesioned adult rat spinal cord. On the other hand, the implantation of collagen substratum alone did not generate axonal ingrowth. Both cellular and acellular implants closely apposed the host cord and appeared to be well tolerated by the host. Martin et al. 1991, Neurosci. Lett. 124:44-48) have recently reported substantial axonal growth into SC populations injected into lesioned rat cord.

Axons and SCs did not enter the acellular collagen implants, even though collagen has been shown to be a favorable terrain for neurite extension in vitro (Bunge et al., 1987, In: Progress in Brain Research, eds: Seil,

Herbert, Carlson; Elsevier, 71:61-74).

The origin and fate of the axons that enter the

SC-collagen grafts are being studied. The appearance of numerous SC- yelinated axons in dorsal regions of lesioned but not implanted cord suggests some ingrowth from dorsal root ganglia (Salvatierra et al., 1990, Soc. Neurosci. Abstr. 16:1282). Silver stained sections suggest that at least some axons traveling through the grafts originate in the spinal cord; the presence of axons in close relation to Schwann-like cells in zones of degenerating corticospinal tract caudal to the lesion suggests that some axons are able to re-enter host tissue and extend for some distance when accompanied by SCs. Our present results utilizing purified SC populations clearly establish the ability of SCs to influence the capacity of regeneration in the spinal cord.

8. EXAMPLE: SEMI-PERMEABLE GUIDANCE CHANNELS

SEEDED WITH CULTURED ADULT SYNGENEIC SCHWANN CELLS ENHANCE PERIPHERAL NERVE REGENERATION

In the present experiments, the regenerative potential of Schwann cells seeded in permselective guidance channels was evaluated in transected rat sciatic nerve across an 8mm nerve gap. Immunological sequelae were evaluated by comparing Schwann cells from syngeneic (a model for autologous) and heterologous rat strains. Schwann cells from either outbred (CD) or inbred (Fisher) adults rats were suspended in a laminin-containing hydrogel at a final density of 80 x 10 cells/ml (CD) or 40, 80, or 120 x 10 cells/ml (Fisher; F-40, F-80, F-120 channels, respectively) . Channels containing Schwann cells were compared to nerve autografts, empty channels or channels filled with the laminin-containing hydrogel alone. By three weeks post implantation, regenerating axons had grown into all grafts. Sciatic nerve autografts supported extensive regeneration, containing 4000-5000 myelinated axons at the graft midpoint. The ability of tubes containing Fisher Schwann cells to foster regeneration was dependent on the density of Schwann cells. At the graft midpoint, F-120 channels contained nearly as many myelinated axons as sciatic nerve autografts, and significantly more than F-80, F-40 or control channels. The nerve cable in Schwann cell containing tube consisted of larger, more organotypic fascicles than acellular control channels. In contrast, heterologous (CD) Schwann cells elicited a strong immune reaction which impeded regeneration; myelinated axons were seen no more than 2mm into the graft channel. The present study shows that cultured adult syngeneic Schwann cells seeded in high density in permselective synthetic guidance channels support extensive peripheral nerve regeneration. Successful regeneration depends upon immune compatibility between donor and host, indicating the importance of developing large populations of autologous adult human Schwann cells for clinical use in the repair of nerve injury.

8.1. MATERIALS AND METHODS

A large portion of the experiments detailed below were carried out by Dr. P. Aebischer of Brown University. Schwann cell cultures: Schwann cells from adult_,rat sciatic nerves were isolated according to a modified technique by (Morrissey et al., 1991, J. Neurosci., in press). Schwann cells were harvested from peripheral nerves following a period of "in vitro Wallerian degeneration". Sciatic nerves either from adult male inbred Fisher 344 rats (Taconic, Germantown, NY) weighing 250-300 g or from adult male outbred CD rats (Charles River Laboratories, MA) weighing 400 to 450 g were collected into Dulbecco's Modified Eagles

Medium (DMEM) (Gibco Laboratories, Long Island, NY) , cleaned from their epmeurium and chopped into 1 mm 2 pieces. The nerve pieces were placed on Vitrogen^®-coated (Collagen Corporation, Palo Alto, CA) Petri dishes (Fisher Scientific,

Medford, MA) in DMEM supplemented with 10% fetal calf serum

(FCS) (Gibco Lab) and penicillin/streptomycin (1,000 U/ml)

(DMEM-FCS) . Every 5 days, the nerve pieces were transferrd into new Petri dishes until most fibroblasts had migrated out of the nerve explants, usually after 3 to 4 passages. The cells were dissociated by incubating the nerve chunks in Ca -Mg -free Hanks Balanced Salt Solution (HBSS) containing 0.3% trypsin (Sigma, St. Louis, MO), 0.1% collagenase (Sigma) and 0.1% hyaluronidase (Sigma) at 37°C for 2 hours. The cells were then washed and cultured in DMEM-FCS. The culture medium was replaced with mitogenic medium containing DMEM, FCS, forskolin (2 μM) (Sigma) and pituitary extract (10 μg/ml) (Collaborative Research Inc., Bedford, MA) a day later (Porter et al., 1986, J. Neurosci. 6:3070-3078). For the next 4-6 days, Schwann cells were cultured with mitogenic medium at 37"C in a humidified atmosphere with 5% CO..

Guidance channel preparation: Guidance channels consisted of a 60:40 acrylonitrile vinylcholoride (PAN/PVC) copolymer tubing with a 1.12 mm ID and a 0.126 μm-thick wall. These tubes were fabricated using a dry-jet wet spinning technique (Cabasso 1980, Encyclopedia of Chemical Technology, Kirk-Othner, 12:492-517; Aebischer et al., 1991, Biomaterials 12:50-56). The tubes featured a smooth inner skin connected to a partially fenestrated outer skin by a trabecular network. The inner skin provided a permselective barrier, with a molecular weight cut-off of 50,000 Da. Prior to seeding, the tubing was cleaned and sterilized as previously described (Aebischer et al., 1988, Brain Res. 454:179-187).

Seeding with Schwann cells: Rat syngeneic (inbred) Schwann cells were collected by centrifugation after treatment with 0.05% trypsin and 0.02% EDTA (Sigma) for 5 min at 37°C. The cells were pelleted, washed in DMEM and the number of live cells was evaluated with trypan blue exclusion using a hemocytometer. Pelleted cultured Schwann cells were gently suspended in a 70:30 (v:v) solution of DMEM:laminin-containg hydrogel (Matrigel ) (Collaborative

Research Inc.) at a final density of either 40 x 10 cells/ml (F-40 channel) , 80 x 10 cells/ml (F-80 channel) or

120 x 10 cells/ml (F-120 channel) • Ten cm-long tubes were filled with the Schwann cell suspension, avoiding air bubbles. Channels were also seeded with non-syngeneic

Schwann cells. Cultured outbred Schwann cells were collected as described above and suspended in a 70:30 solution of DMEM-FCS:laminin-containing hydrogel at a final density of 80 x 10 cells/ml (CD-80 channels) . The filled tubes were then placed in DMEM at room temperature to allow for gellation of the cell suspension and cut into lOmm-long pieces. These tubes were closed at each end using PAN/PVC copolymer glue. The seeded channels were kept in DMEM for

24 hours at 37"C. Control channels were filled with a 70:30 β solution of DMEM:Matrigel in a similar way (F-CO-φ channel) . The closing caps were removed at the time of implantation.

Arrangement of the cells in the channels: Prior to implantation, tubes collected i) after filling, ii) on the day of surgery and iii) 5 days post-seeding were fixed overnight in 2.5% paraformaldehyde in 0.l M PBS at pH 7.4 and observed under either scanning electron microscopy (SEM) or light microscopy. For SEM, the channels were post-fixed, dehydrafted in dimethylaminomethyl phenol (DMP) , critical point dried and observed using a Hitachi 2700 microscope.

For light microscopy, the samples were dehydrafted, β infiltrated overnight in Historesin (Cambridge Instruments,

Deerfield, IL) and embedded. Five μm-thick longitudinal sections of the seeded tubes were cut using a Reichert-Jung

2050 microtome (Cambridge Instruments) . Sections were counter-stained with cresyl violet. Evaluation of the Schwann cell cultures: The purity of the cultures was quantified on the days of implantation. At the same time the channels were filled, cells were seeded on uncoated glass coverslips placed in a

₅ 6-well plastic plate at a density of 10 cells/well in DMEM-FCS. The purity of the cultures was determined by immunostaining using both Schwann cell and fibroblast markers. Schwann cells were labelled with both mouse monoclonal 217c antibody, a selective marker for NGF receptors (Kumar et al., 1990, J. Neurosci. Res. 27:408-417) and polyclonal S-100 antibody (Dakopatts, Santa Barbara,

CA) . The 217c antibody was generously provided by Dr. J. de Vellis (University of California at Los Angeles, CA) . Contamination with fibroblasts was assessed using a rabbit polyclonal fibronectin antibody, generously supplied by Dr. R. Morris (National Institute for Medical Research, London, England) .

For Schwann cell labelling, cells were fixed using a 3% solution of paraformaldehyde in 0.01 M PBS at pH 7.4 and incubated at room temperature with either the 217c (1:100) or S-100 (1:200) antibody with 0.2% Triton X-100 in buffer. Cells were then respectively incubated with secondary rabbit anti-mouse antiserum (1:50) (Sternberger-Meyer, Jarrettsville, MD) or goat anti-rabbit antiserum (1:100) (Sternberger-Meyer) followed by mouse (1:50) (Sternberger-Meyer) or rabbit (1:100) peroxidase anti-peroxidase (PAP) (Dakopatts) . For fibroblast identification, cells were fixed with 95% ethanol: 5% glacial acetic acid (v:v) solution at -20°C for 20 min and incubated overnight at 4°C with rabbit anti-fibronectin antiserum (1:20,000) with 0.2% Triton X-100 in buffer.

Secondary goat anti-rabbit antiserum (1:200) and rabbit PAP (1:100) were then used. All reactions were visualized with a solution of 0.05% diaminobenzidine tetrahydrochloride (Sigma) and 0.01% hydrogen peroxide (Sigma). The cells were then counterstained with cresyl violet. The percentage of 217c-positive, S-100-positive and fibronectin-positive cells was evaluated at a magnification of 160x on 25 randomly taken fields. Characterization of the cells after seeding:

Tubes collected on the day of implantation were fixed overnight in 3% paraformaldehyde in 0.1 M PBS at pH 7.4.

The specimens were frozen on dry ice and 20 μm-thick longitudinal sections were obtained using a Reichert-Jung 1800 cryocut (Cambridge Instruments) . The sections were mounted on APTES-coated glass slides and reacted with the

217c antibody as described above.

Guidance channel implantation: Inbred Fisher 344 rats weighing 100-125 g were anesthetized with pentobarbital ® (Nembutal ) and their left sciatic nerve exposed as previously described (Guenard et al., 1991, Biomaterials,

12:259-263). The nerve was sectioned 3 mm distal to the tibio-peroneal bifurcation and a 4-5 mm-long piece of nerve was resected and discarded. Both the proximal and distal nerve stumps were secured 1 mm within the channels, leaving an 8 mm-long nerve gap. The wound was then closed. Rats were implanted for 3 weeks with empty (n=6) , F-CD-φ (n=6) , F-40 (n=7) , F-80 (n=4), F-120 (n=3) and CD=80 (n=6) channels. Animals were housed in plastic cages in rooms with 12 hours on-off light cycles and received food and water ad libitum.

Sciatic nerve autografting: The left sciatic nerve was exposed and cut both 3 and 11 mm distal to the tibio-peroneal bifurcation. The 8 mm-long piece of sciatic nerve was removed and rotated to align the distal end of the nerve segment with the proximal nerve stump and the proximal end of the nerve piece with the distal nerve stump. The nerve autograft was secured in place using 4 to 6 10.0 nylon suture epineurial stitches. Six rats received sciatic nerve autografts for 3 weeks.

Implant retrieval and evaluation: Rats were

® deeply anesthetized with pentobarbital (Nembutal ) , transcardially perfused with buffered fixative and both the channels and autografts exp1anted (Guenard et al., 1991, Biomaterials, 12:259-263). Samples were fixed overnight in 3% paraformaldehyde and 2.5% glutaraldehyde in 0.1 M PBS at pH 7.4, transversely cut at their midpoint, post-fixed, dehydrated in graded ethanol, inflitrated and embedded in Spurr resin. Specimens were sectioned 2, 4, 6 and 8 mm distal from the proximal suture (sections S2, 4, 6 and 8, respectively) using a Sorvall MT-5000 microtome. Semi-thin sections were stained with toluidine blue and ultra-thin sections with Reynold's lead citrate and uranyl acetate for transmission electron microscopy (TEM) . The cross-sectional area (CSA) of the regenerated tissue, number of myelinated

2 axons and number of blood vessels per mm of CSA were quantified using a computerized morphometric analysis system

(CUE-2, Olympus Corp., Lake Success, NY) interfaced with a

Zeiss IM35 microscope. Ultra-thin sections were examined at sections S4 and S8 for the presence of Schwann cells and unmyelinated axonal profiles using a Philips 410 electron microscope. All data are presented as mean + SEM. The data were subjected to analysis of variance by Games-Howell test for treatment comparisons. Statistical significance was assigned for p<0.05.

8.2. RESULTS Evaluation of the cultures: Dissociation of nerve explants yielded cell populations that were 97% 217c- and S-100-positive (97.2 + 0.6%, 97.0 + 1.0%, respectively). These cells were morphologically divided into 2 groups. Most of the cells were small, elongated and spindle-shaped. These Schwann cells had a tendency to line up side-by-side or end-to-end and formed interconntected networks. Some slightly larger cells, tripolar in shape with a fibroblast- like appearance, were also observed. These Schwann cells may have lost their characteristic spindle shape due to the presence of forskolin in the culture medium (Porter et al., 1986, J. Neurosci. 6:3070-3078). A few 217c/S-l00 negative cells with a large purple-stained ovoid nucleus were scattered among Schwann cells; these cells were fibronectin-positive.

Arrangement of the cells in the channels: Cell viability as observed by trypan blue exclusion was above 95% at the time the channels were seeded. Scattered round cells filled the lumen of the tubes. After a day in culture, a cellular cable free of attachment from the channel's wall was observed at both SEM and light microscopy. Cells were aligned along the longitudinal axis of the tubes, arranging themselves side-by-side and end-to-end. A similar pattern was observed after 5 days in culture. The aligned cells were 217c-positive, indicating that seeded Schwann cells expressed NGF receptors at the time of implantation.

In vivo Studies: Upon retrieval, all channels showed a minimal tissue reaction. The outer surface of the tubes was covered with a layer of macrophage-like cells and a few layers of fibroblast-like cells. A cable bridging both nerve stumps regenerated in all the tubes, whether they

® were empty, filled with Matrigel or filled with either

® outbred or inbred Schwann cells suspended in Matrigel . At any point along the tubes, the regenerated nerve cables were round in shape, and separated from the inner wall of the channels by an acellular gel. The morphology of the regenerated nerve cables differed whether non-syngeneic (outbred) or syngenic (inbred) Schwann cells were implanted. Non-syngeneic Schwann cells: The cable surface area (CSA) of the regenerated cables decreased up to section S6, then increased toward section S8 (Figure 2A) . At section S2, the regenerated cables consisted of blood vessels and nerve microgascicles with unmyelinated and myelinated axons and associated Schwann cells surrounded by an epineurial-like sheath. Myelinated axons extended up to section S2 in all the tubes but were never seen at level S4

(Figure 2B) . From section S4 to S8, the regenerated cables consisted of connective tissue containing blood vessels,

® acellular Matrigel patches and numerous reactive cells including macrophage-like cells and mast cells (Figure 3A) .

No unmyelinated axons were observed at S4. There was no signi .fi.cant difference i.n number of blood vessels per mm2 of CSA in CD-80 channels as compared to the other type of channels, except F-CD-^ channels (Figure 6) .

Syngβneic Schwann cells: Cable cross-sectional area: In all the channels, the cable surface area (CSA) decreased up to section S4, then increased toward the distal nerve stump (Figure 2A) . Cables extending in both F-120 and F-80 channels were larger than cables regenerated in empty, F-CD-φ and F-40 channels (Figure 3A) . There was no significant difference in CSA between cables regenerated in F-40 tubes as compared to empty and F-CO-φ channels (Figure 2A) .

Morphology: In all groups of channels, cables contained nerve microfascicles and blood vessels surrounded by a thin epineurial-like layer (Figures 4A, 4D, and 4G) . At any point along the tubes, the percentage of the CSA covered with an epineurial-like structure was similar in all channels. At any section along empty, F-40, F-80 and F-120 channels, microfascicles contained both unmyelinated and myelinated axons with associated Schwann cells (Figures 3B, 4B, 4C, 4H, 41, 5A and 5B) . In contrast, in F-CO-φ channels, unmyelinated axons bridged the nerve gap but myelinated axons only elongated up to section S6 in 2 cases, and S4 in the remaining 4 cases (Figures 4E and 4F) . In addition, the organization of the regenerated tissue varied among the channels. Tissue regenerated in F-40, F-80 and F-120 channels was morphologically more organized as compared to tissue regenerated in both empty and F-CO-φ tubes (Figure 4B, 4C, 4E, 4F versus figures 4H and 41) , with less microfasciculation. In empty and F-CO-φ channels, the microfascicles were well separated by numerous fibroblast-

® like cells. Occassionally, small patches of Matrigel were seen in the center of cables regenerated in F-CD-^ channels.

Myelinated axons: The number of myelinated axons decreased along the length of all of the channels (Figure 2B) . At any section along the tubes, cables regenerated in

F-40, F-80, and F-120 channels contained more myelinated axons as compared to empty and F-CD-^ channels, except at S8 where F-40 channels contained less myelinated axons than empty channels (Figure 2B) . As the Schwann cell seeding density increased the myelinated axon population increased (Figure 2B) ; F-120 channels contained significantly more myelinated axons as compared to F-80 channels at both S2 and S4 and the myelinated axon population was signficantly higher in F-80 channels as compared to F-0 channels from S6 to S8. Finally, cables extending in F-CO-φ tubes contained less myelinated axons as compared to cables regenerated in empty tubes (Figure 3B) ; however, the difference was not statistically significant.

Blood vessels: The number of blood vessels per mm 2 of CSA did not vary significantly between empty, F-40,

F-80 and F-120 channels (Figure 6) . However, at any section along the tubes except S8, cables regenerated in F-CD-φ channels contained signficantly more blood vessels as compared to empty, F-40, F-80 and F-120 channels (Figure 6) . Sciatic nerve autografts: At any point along the autografts, 2 nerve bundles surrounded by a thin epineurial-like layer were observed. Each bundle consisted of numerous nerve microfascicles containing unmyelinated and myelinated axons with their associated Schwann cells; degenerating figures were often seen (Figure 3C) . The number of myelinated axons decreased from section S2 to section S8 (Figure 2B) . The myelinated axon population in the autografts did not vary signficantly from that in F-120 channels at both S2 and S4 levels; however, the difference was significant at sections S6 and S8. The number of blood

₂ vessels per mm of CSA in the nerve autograft was significantly smaller as compared to the nerve regenerated through any type of channel (Figure 6) .

8.3. DISCUSSION

The present study shows that cultured adult syngeneic Schwann cells harvested after their isolation from degenerated sciatic nerve segments and suspended in a laminin-containing hydrogel are able to form an oriented central cable in synthetic guidance channels. Such Schwann cell cable enhanced peripheral nerve regeneration through permselective guidance channels in a seeding density- dependent fashion. Our findings show that myelinated axon population increased in channels containing syngeneic Schwann cells as compared to channels filled with laminin- containing hydrogel. Although the laminin-containing hydrogel itself impeded regeneration, as previously reported (Valentini et al., 1987, Exp. Neurol. 98: 325-338), cultured adult Schwann cells were able to reverse the inhibitory effect of the hydrogel and enhance the regenerative processes, suggesting that the transplanted Schwann cells interacted directly with the outgrowing axons. Shine et al. (1985, J. Neurosci. Res. 14:393-401) also reported successful regeneration through blind-ended polyethylene channels filled with PNS cultures containing both Schwann cells and dorsal root ganglia (DRG) . However, DRG neurons are known to be mitogenic for Schwann cells and may have stimulated the proliferation of endogenous Schwann cells. Furthermore, it was not possible to differentiate the effects of the various implanted cells on regeneration, making the interpretation of those results difficult.

The physical organization of the Schwann cells in the tubes may have also influenced the outcome of regeneration. At the time of implantation, channels were filled with an organized cellular cable aligned along their main axis. Schwann cells themselves were also oriented in the same direction, lining end-to-end and side-by-side. It is likely that the presence of this pre-existing Schwann cell cable enhanced regenerative processes. It has been demonstrated that the formation of an organized central fibrin cable between the stumps of transected peripheral nerves is critical for successful regeneration (Williams and Varon, 1985, J. Comp. Neurol. 12:851-860; Aebischer et al., 1990, Brain Res. 531:211-218). Furthermore, prefilling silicone elastomer channels with an organized fibrin matrix enhances the rate of regeneration of peripheral nerves (Williams, 1987, Neurochem. Res. 12:851-860). Thus, by forming an organized cable, the lined Schwann cells may have served as a scaffold for elongating axons and initial regenerative events were not needed for neurite outgrowth. The Schwann cell seeding density in the tubes influenced the outcome of regeneration. The number of myelinated axons in cables regenerated through channels seeded with the low Schwann cell density (F-40 channels) were close to the myelinated axon counts in the empty channels whereas the myelinated axon counts increased signficantly when the seeding density was doubled (F-80 channels) or tripled (F-120 channels) . Furthermore, in F-120 channels, up to their midpoint, the number of myelinated axons was close to the myelinated axon population in the sciatic nerve autografts.

Non-syngeneic adult Schwann cells seeded in per selective guidance channels impeded regeneration. Myelinated axons did not extend farther than 2 mm within channels seeded with non-syngeneic Schwann cells whereas myelinated axons and bridged the nerve gap in channels seeded with syngeneic Schwann cells at a similar density.

(Kalderon, 1988, J. Neurosci. Res. 21:501-512) also reported no stimulatory effect of the seeded Schwann cells on peripheral nerve regeneration. Cultured neonatal Schwann cells from outbred rats seeded in silicone elastomer channels did not influence the success of regeneration. The lack of potential beneficial effects of transplanted Schwann cells isolated from outbred rat strains on nerve regeneration may have been due to the observed immune reaction. Schwann cells in allografts have been shown to express MHC class I and II antigens (Ansselin and Pollard, 1990, J. Neurol. Sci. 96:75-88). These can act as antigen presenting cells and trigger an immune response which ultimately destroys the transplanted Schwann cells. No immune reaction was observed with implanted syngeneic Schwann cells, indicating that their stimulatory effect on regeneration was clear. After Schwann cell-axonal contact occurred, some of the Schwann cells may have differentiated into myelin-forming cells and participated in the myelination of the regenerating axons.

We conclude that Schwann cells isolated from adult rat sciatic nerves are able to enhance regenerative processes in the PNS and that the density of the Schwann cells in the channels determine the outcome of regeneration. This study suggests that Schwann cells could be used clinically for the repair of peripheral nerves. Transplantation of autologous Schwann cells would allow the avoidance of immunological rejection.

The present invention is not to be limited in scope by the embodiments disclosed in the examples which are intended as illustrations of a few aspects of the invention and any embodiments which are functionally equivalent are within the scope of this invention. Indeed, various modifications of the invention in addition to those shown and described herein will become apparent to those skilled in the art and are intended to fall within the scope of the appended claims. Various publications are cited herein, which are incorporated by reference in their entireties.

Claims

WHAT IS CLAIMED IS:

1. A method of promoting nervous system repair comprising transplanting autologous Schwann cells into a region of injury in the central or peripheral nervous system.

2. The method of claim 1 in which the Schwann cells are harvested from a peripheral nerve of a patient in need of such treatment.

3. The method of claim 2 in which the peripheral nerve is the sural nerve.

4. The method of claim 2 in which the peripheral nerve is the saphenous nerve.

5. The method of claim 2 in which the peripheral nerve is the brachial nerve.

6. The method of claim 2 in which the peripheral nerve is the antibrachial nerve.

7. The method of claim 1 in which the Schwann cells are harvested from damaged tissue.

8. The method of claim 1 in which the Schwann cells have been cultured prior to transplantation.

9. The method of claim 2 in which the Schwann cells have been cultured prior to transplantation.

10. The method of claim 3 in which the Schwann cells have been cultured prior to transplantation.

11. The method of claim 4 in which the Schwann cells have been cultured prior to transplantation.

12. The method of claim 5 in which the Schwann cells have been cultured prior to transplantation.

13. The method of claim 6 in which the Schwann cells have been cultured prior to transplantation.

14. The method of claim 7 in which the Schwann cells have been cultured prior to transplantation.

15. The method of claim 8 in which the Schwann cells have been cultured prior to transplantation by a method comprising (i) culturing a peripheral nerve explant on a culture substrate such that fibroblasts exit the explant and form outgrowths on the culture substrate; (ii) transferring the explant to a fresh culture container repeatedly until fibroblasts have substantially exited the explant; and (iii) dissociating the explant to form a population of cells which is greater than about ninety percent Schwann cells.

16. The method of claim 9 in which the Schwann cells have been cultured prior to transplantation by a method comprising (i) culturing a peripheral nerve explant on a culture substrate such that fibroblasts exit the explant and form outgrowths on the culture substrate; (ii) transferring the explant to a fresh culture container repeatedly until fibroblasts have substantially exited the explant; and (iii) dissociating the explant to form a population of cells which is greater than about ninety percent Schwann cells.

17. The method of claim 10 in which the Schwann cells have been cultured prior to transplantation by a method comprising (i) culturing a peripheral nerve explant on a culture substrate such that fibroblasts exit the explant and form outgrowths on the culture substrate; (ii) transferring the explant to a fresh culture container repeatedly until fibroblasts have substantially exited the explant; and (iii) dissociating the explant to form a population of jells which is greater than about ninety percent Schwann cells.

18. The method of claim 11 in which the Schwann cells have been cultured prior to transplantation by a method comprising (i) culturing a peripheral nerve explant on a culture substrate such that fibroblasts exit the explant and form outgrowths on the culture substrate; (ii) transferring the explant to a fresh culture container repeatedly until fibroblasts have substantially exited the explant; and (iii) dissociating the explant to form a population of cells which is greater than about ninety percent Schwann cells.

19. The method of claim 12 in which the Schwann cells have been cultured prior to transplantation by a method comprising (i) culturing a peripheral nerve explant on a culture substrate such that fibroblasts exit the explant and form outgrowths on the culture substrate; (ii) transferring the explant to a fresh culture container repeatedly until fibroblasts have substantially exited the explant; and (iii) dissociating the explant to form a population of cells which is greater than about ninety percent Schwann cells.

20. The method of claim 13 in which the Schwann cells have been cultured prior to transplantation by a method comprising (i) culturing a peripheral nerve explant on a culture substrate such that fibroblasts exit the explant and form outgrowths on the culture substrate; (ii) transferring the explant to a fresh culture container repeatedly until fibroblasts have substantially exited the explant; and (iii) dissociating the explant to form a population of cells which is greater than about ninety percent Schwann cells.

21. The method of claim 14 in which the Schwann cells have been cultured prior to transplantation by a method comprising (i) culturing a peripheral nerve explant on a culture substrate such that fibroblasts exit the explant and form outgrowths on the culture substrate; (ii) transferring the explant to a fresh culture container repeatedly until fibroblasts have substantially exited the explant; and (iii) dissociating the explant to form a population of cells which is greater than about ninety percent Schwann cells.

22. The method of claim 1 in which the Schwann cells are comprised in a gelatinous vehicle.

23. The method of claim 2 in which the Schwann cells are comprised in a gelatinous vehicle.

24. The method of claim 7 in which the Schwann cells are comprised in a gelatinous vehicle.

25. The method of claim 8 in which the Schwann cells are comprised in a gelatinous vehicle.

26. The method of claim 9 in which the Schwann cells are comprised in a gelatinous vehicle.

27. The method of claim 15 in which the Schwann cells are comprised in a gelatinous vehicle.

28. The method of claim 16 in which the Schwann cells are comprised in a gelatinous vehicle.

29. The method of claim 21 in which the Schwann cells are comprised in a gelatinous vehicle.

30. The method of claim 1 which is used to treat a spinal cord injury.

31. The method of claim 2 which is used to treat a spinal cord injury.

32. The method of claim 7 which is used to treat a spinal cord injury.

33. The method of claim 8 which is used to treat a spinal cord injury.

34. The method of claim 9 which is used to treat a spinal cord injury.

35. The method of claim 15 which is used to treat a spinal cord injury.

36. The method of claim 16 which is used to treat a spinal cord injury.

37. The method of claim 21 which is used to treat a spinal cord injury.

38. The method of claim 1 which is used to treat a brain injury.

39. The method of claim 2 which is used to treat a brain injury.

40. The method of claim 7 which is used to treat a brain injury.

41. The method of claim 8 which is used to treat a brain injury.

42. The method of claim 9 which is used to treat a brain injury.

43. The method of claim 15 which is used to treat a brain injury.

44. The method of claim 16 which is used to treat a brain injury.

45. The method of claim 21 which is used to treat a brain injury.

46. A pharmaceutical composition comprising cells which are greater than about ninety percent Schwann cells in a suitable pharmacologic carrier.

47. The pharmaceutical composition of claim 46 which further comprises collagen.

48. The pharmaceutical composition of claim 46 which further comprises agarose.

49. The pharmaceutical composition of claim 46 which further comprises dextran sulfate.

50. The pharmaceutical composition of claim 46 which further comprises collagen-glycosaminoglycan.

51. The pharmaceutical composition of claim 46 which further comprises silicone.

52. The pharmaceutical composition of claim 46 which further comprises an organic polymer.

53. The pharmaceutical composition of claim 52 in which the organic polymer comprises polyvinyl chloride.

54. The pharmaceutical composition of claim 52 in which the organic polymer comprises polylysine.

55. The pharmaceutical composition of claim 52 in which the organic polymer comprises polyornithine.

56. The pharmaceutical composition of claim 46 which is prepared by a method comprising (i) culturing a peripheral nerve explant on a culture substrate such that fibroblasts exit the explant and form outgrowths on the culture substrate; (ii) transferring the explant to a fresh culture container repeatedly until fibroblasts have substantially exited the explant; and (iii) dissociating the explant to form a population of cells which is greater than about ninety percent Schwann cells.

57. The pharmaceutical composition of claim 47 which is prepared by a method comprising (i) culturing a peripheral nerve explant on a culture substrate such that fibroblasts exit the explant and form outgrowths on the culture substrate; (ii) transferring the explant to a fresh culture container repeatedly until fibroblasts have substantially exited the explant; and (iii) dissociating the explant to form a population of cells which is greater than about ninety percent Schwann cells.

58. The pharmaceutical composition of claim 48 which is prepared by a method comprising (i) culturing a peripheral nerve explant on a culture substrate such that fibroblasts exit the explant and form outgrowths on the culture substrate; (ii) transferring the explant to a fresh culture container repeatedly until fibroblasts have substantially exited the explant; and (iii) dissociating the explant to form a population of cells which is greater than about ninety percent Schwann cells.

59. The pharmaceutical composition of claim 49 which is prepared by a method comprising (i) culturing a peripheral nerve explant on a culture substrate such that fibroblasts exit the explant and form outgrowths on the culture substrate; (ii) transferring the explant to a fresh culture container repeatedly until fibroblasts have substantially exited the explant; and (iii) dissociating the explant to form a population of cells which is greater than about ninety percent Schwann cells.

60. The pharmaceutical composition of claim 50 which is prepared by a method comprising (i) culturing a peripheral nerve explant on a culture substrate such that fibroblasts exit the explant and form outgrowths on the culture substrate; (ii) transferring the explant to a fresh culture container repeatedly until fibroblasts have substantially exited the explant; and (iii) dissociating the explant to form a population of cells which is greater than about ninety percent Schwann cells.

61. The pharmaceutical composition of claim 51 which is prepared by a method comprising (i) culturing a peripheral nerve explant on a culture substrate such that fibroblasts exit the explant and form outgrowths on the culture substrate; (ii) transferring the explant to a fresh culture container repeatedly until fibroblasts have substantially exited the explant; and (iii) dissociating the explant to form a population of cells which is greater than about ninety percent Schwann cells.

62. The pharmaceutical composition of claim 52 which is prepared by a method comprising (i) culturing a peripheral nerve explant on a culture substrate such that fibroblasts exit the explant and form outgrowths on the culture substrate; (ii) transferring the explant to a fresh culture container repeatedly until fibroblasts have substantially exited the explant; and (iii) dissociating the explant to form a population of cells which is greater than about ninety percent Schwann cells.

63. The pharmaceutical composition of claim 53 which is prepared by a method comprising (i) culturing a peripheral nerve explant on a culture substrate such that fibroblasts exit the explant and form outgrowths on the culture substrate; (ii) transferring the explant to a fresh culture container repeatedly until fibroblasts have substantially exited the explant; and (iii) dissociating the explant to form a population of cells which is greater than about ninety percent Schwann cells.

64. The pharmaceutical composition of claim 54 which is prepared by a method comprising (i) culturing a peripheral nerve explant on a culture substrate such that fibroblasts exit the explant and form outgrowths on the culture substrate; (ii) transferring the explant to a fresh culture container repeatedly until fibroblasts have substantially exited the explant; and (iii) dissociating the explant to form a population of cells which is greater than about ninety percent Schwann cells.

65. The pharmaceutical composition of claim 55 which is prepared by a method comprising (i) culturing a peripheral nerve explant on a culture substrate such that fibroblasts exit the explant and form outgrowths on the culture substrate; (ii) transferring the explant to a fresh culture container repeatedly until fibroblasts have substantially exited the explant; and (iii) dissociating the explant to form a population of cells which is greater than about ninety percent Schwann cells.

66. An isolated population of cells which is (i) greater than about ninety percent Schwann cells and (ii) which is prepared by a method comprising (i) culturing a peripheral nerve explant on a culture substrate such that fibroblasts exit the explant and form outgrowths on the culture substrate; (ii) transferring the explant to a fresh culture container repeatedly until fibroblasts have substantially exited the explant; and (iii) dissociating the explant.

67. The method of claim 2 in which the patient is an adult.

68. The pharmaceutical composition of claim 46 in which the Schwann cells are Schwann cells of an adult.

69. The population of cells of claim 66 in which the peripheral nerve explant is from an adult.