WO2000064460A2

WO2000064460A2 - Composition for enhancing functional recovery of a mammal from central and/or peripheral nervous system injury of traumatic or pathological origin

Info

Publication number: WO2000064460A2
Application number: PCT/IB2000/000480
Authority: WO
Inventors: Diego Braguglia; Pedro Bittmann; Sophie Feldblum; Sonia Arnaud
Original assignee: Sulzer Orthopedics Ltd.
Priority date: 1999-04-23
Filing date: 2000-04-19
Publication date: 2000-11-02
Also published as: WO2000064460A3; AU3667100A

Abstract

Functional recovery of a mammal from central and/or peripheral nervous system injury of traumatic or pathological origin and including cerebral ischemia and stroke is enhanced by administration of a composition comprising a therapeutically effective amount of a natural protein mixture derived from bone, cartilage and/or tendon/ligament tissue and having bone morphogenic and/or osteoinductive characteristics. This natural protein mixture is preferably the mixture known as BP (Bome Protein) which mixture is derived from bovine bone material by extracting proteins from demineralized bone particles and purifying the extract by ultrafiltration, anion exchange, cation exchange and reverse phase high resolution liquid chromatography. In the rat model it is shown that the enhancement regarding BP-treated animals compared with sham-treated animals includes acceleration of recovery for reversible and irreversible cases of spinal cord lesions and a higher recovery level, i.e. a smaller irreversible functional loss for irreversible cases. Recovery is believed to be due to the beneficial effect of the natural protein mixture on cell survival after injury, on cell protection against exitotoxicity (e.g. glutamate intoxication) as well as on injury site conditions favoring axon regeneration. Compared with known administration of single species of recombinantly produced bone morphogenic proteins for similar aims, the natural protein mixture is effective in considerably lower doses.

Description

COMPOSITION FOR ENHANCING FUNCTIONAL

RECOVERY OF A MAMMAL FROM CENTRAL AND/OR

PERIPHERAL NERVOUS SYSTEM INJURY OF TRAUMATIC

OR PATHOLOGICAL ORIGIN

Field of the Invention

The invention relates to a composition according to the generic part of the first independent claim. The composition serves for enhancing functional recovery of a mammal from traumatic or pathological injury of the central and/or peripheral nerv- ous system (CNS and/or PNS).

Background of the invention

During development of the mammalian body, neurons extend their neurites and establish neuronal connections. Both in the CNS and in the PNS, neurites and dendrites abundantly elongate during development. For functional recovery after central or peripheral nervous system injury due to traumatic lesions or following pathological conditions neurite regrowth and reestablishment of connections similar to the corresponding processes during development is required in the adult body. However, in the adult CNS, axonal and dendridic regrowth is increasingly lost with evolutionary progression and the adult PNS only partially maintains the ability to rebuild damaged or destroyed connections. In mammals, neurite regrowth is limited to neuritic sprouting, whereas in lower organisms (vertebrates) regrowth of neuronal processes is possible also. It is well known that the failure of effective axonal regeneration in the mammalian central nervous system compared with the peripheral nervous system does not result from an intrinsic inability of central neurons to regenerate. In other environments, CNS neurons are able to regrow their axons to some extent. The non-permissive characteristics of the adult CNS are partially explained by the presence of membrane bound proteins, related to oligodendrocytes and central myelin that inhibit neurite growth. The failure in regeneration may also be due to the absence of growth promoting factors in the post injury neuronal microenvironment.

Furthermore, it is known that axonal regrowth is heavily influenced and controlled by glial cells. In mammals, glial cells are generally permissive for neurite outgrowth in the CNS during development and in adult PNS. Like in the mammalian PNS, glial cells in the CNS of some lower vertebrates retain their permission for neurite regrowth to the adulthood thus allowing partial reversion to the neurite outgrowth- promoting potential and fostering regeneration. In contrast, following lesions, glial cells in higher mammals are not supportive to neurite regrowth.

Neurotrophic factors are able to alter the cell-body response to injury by rescuing immature and mature axotomized CNS neurons from retrograde cell death, i.e. enable neurons to survive injury, which constitutes one of the primary conditions for regeneration. In the PNS, axotomy frequently leads to production of regeneration- associated proteins (i.e. α- and β-tubulin, actin, GAP-43 and others). This event is paralleled by successful axonal regeneration. Axotomy to CNS neurons does not induce any production of regeneration-associated molecules. Only when the lesion occurs close to the cell body, CNS neurons transiently upregulate regeneration associated genes. These observations suggest that a particular intervention (i.e. exposure of injured neurons to neurotrophins or other compounds) leading to upregulation of regeneration-associated genes may increase the ability of CNS neurons to initiate a regrowth response. Recent studies have shown that neurotrophic factors may have specific and overlapping influences both on the survival and on gene expression of axotomized CNS neurons. Neurotrophic factors appear to shift the balance toward regeneration.

Upon injury, the observed repair is obtained by two distinct events: regeneration of damaged pathways and induced sprouting of undamaged neurons. In vivo, regenerating and developing neurons differ in their requirements for growth. E.g. in axotomized rat retinal ganglion neurons, BDNF (brain derived neurotrophic factor) enhances expression of GAP-43 but not of Tal-tubulin, i.e. branching is enhanced by BDNF, but not long distance regrowth which requires Tal-tubulin. It should thus be theoretically conceivable to regulate growth of particular pathways after injury in very specific and distinct ways.

Neural pathways are tightly interconnected in vivo and therefore, axonal regrowth as well as axonal sprouting should be initiated and supported after injury.

Many in vivo experiments suggest that there may be opportunities for synergy be- tween multiple mechanisms to increase anatomical plasticity. Anatomical plasticity will be paralleled by improvements in motor functions. Of particular challenge is the observation that as the post-injury period increases, the capacity of growth factors or other compounds to promote axonal regeneration decreases. To date it is not clear for how long after initial injury mature CNS neurons maintain their ability to regrowth under favorable conditions. The observation that chronically injured CNS neurons presenting a blind-ended peripheral nerve bridge are able to regenerate even after long post-injury intervals, suggests that the window of opportunity for regrowth of injured CNS neurons may be wider than at first anticipated. As noted above, functional recovery from nervous system injury necessitates both activation of the intrinsic capacity for regeneration of the neurons and modification of the hostile environment at the site of injury. Strategies aiming at both include placing grafts of Schwann cells, grafts of peripheral nerve tissue or grafts of embry- onic spinal cord tissue at the lesion site. Major advances have also been made by combining a cellular terrain conductive to axonal growth with exogenous neurotrophic factors or with cells genetically engineered to secrete particular neurotrophins. Neurotrophins increase the regenerative capacity of the neurons acting on the cell body and they influence growth by affecting the cellular and molecular compo- sition of the transplanted tissue. Recently, the discovery of neuronal progenitor cells in the brain and in the spinal cord opened an additional potential way to replace cells lost by injury and/or to alter the terrain at the injury site.

After injury, the regeneration of the adult nervous system is also heavily influenced by numerous inhibitory cues. Among the principal players preventing regeneration are astrocytes- and myelin-associated inhibitors of axonal growth and oligodendro- cytes. Inhibitory cues are preventing axonal growth both at the lesion site and distally to the lesion. Moreover, these inhibitory cues are acting on regenerative axonal growth as well as on sprouting of undamaged axons. Astrocytes, particularly at the lesion site, are a major hurdle to axonal regeneration: they act as a physical barrier to growth and they secrete molecules (proteoglycans) in the extracellular matrix that are restricting axonal growth. When the injury is associated with a breakdown of the blood-brain barrier, chondroitin sulfate proteoglycans are heavily expressed, also preventing axonal regrowth. Serum factors and/or inflammatory cytokines are also very often involved in the molecular cascade that produces extracellular matrix in the immediate vicinity of the developing scar. Production of proteoglycans by astrocytes has been shown to be directly associated with failure of axonal regrowth. Of particular importance will be the determination of how different cell types are influenced for overcoming specific inhibitory components of the adult CNS. As shortly discussed above, recent advances in understanding the intrinsic capacity of mature mammalian CNS neurons to regenerate following traumatic injury or pathological conditions indicate that the windows of opportunity for enhancing neuronal regrowth after damage may be more numerous than previously thought. It seems likely that after injury both intrinsic and extrinsic environmental factors contribute to the normally observed lack of regeneration.

This means that it is not likely that any single regenerative intervention will be able to very successfully reverse the consequences of the injury. Instead of such single regenerative intervention, interacting interventions regarding exogenous neurotrophic supports and regarding the environment of the injury site are to be considered, i.e. interventions consisting of a plurality of intervention steps within a plurality of windows of opportunity which intervention steps may be governed by hierarchies and may be directed to particular neuronal populations having specific requirements for regeneration at very precise time points following injury.

For enhancing functional recovery from traumatic or pathological injury of the nervous system it seems necessary to take favourable action in a plurality of interacting fields of which the most important ones are likely to be the following ones:

a) the intrinsic capacity for axonal regeneration (axonal sprouting and axonal and dendritic regrowth) of mature neurons is to be revitalised and/or maintained;

a) neurons at the injury site are to be supported against cell death;

a) neurons at the injury site are to be protected against exitotoxicity such as e.g. glutamate intoxication;

a) injury site conditions are to be made favourable to axonal regeneration; a) axonal regeneration inhibitors or production thereof is to be deactivated;

f) axonal regeneration is to be controlled with an aim to achieve functional recovery.

Recent studies have demonstrated also the importance of having a normalized scor- ing table to allow drawing parallels between axonal growth/sprouting and post-lesion behavioral improvements in different experimental set-ups. Tests that monitor specific characteristics of locomotion or skilled (fore)limb function allow correlation between axonal regeneration and functional recovery. Monitoring of complex movements (i.e. rhythmic alternating movements such as stepping) allows a determi- nation of the potential mechanism underlying functional recovery.

Bone morphogenic proteins (BMPs) initially identified by their ability to induce ec- topic bone formation and representing a sub-family of structurally and functionally related proteins belonging to the transforming growth factor beta superfamily (TGF- β superfamily) are known to have an effect on other tissues than bone tissue, e.g. on kidney tissue and neural tissue. They are known to have an effect on the CNS during development (differentiation, cell growth, morphogenesis). In the mammalian adult brain, their receptors are still present in some neurons and their expression seems to be regulated after brain and spinal cord lesion.

The following paragraphs discuss a few published approaches to solving problems related to recovery of nerve tissue from malfunction caused by trauma, illness or degeneration by using bone morphogenic proteins. The publication WO-94/03200 (Creative Biomolecules Inc.) describes methods of therapeutic treatment for maintaining neural pathways in mammals in order to counteract the effects of traumatic or degenerative lesions of the brain and/or spinal cord (enhancing survival of damaged neuronal cells, redifferentiating transformed cells of neural origin, maintaining phenotypic expression of differentiated neuronal cells, repairing damaged neural pathways by stimulating axonal growth, alleviating immu- nologically-related/triggered nerve tissue damage). For achieving this goal, a morphogenic protein and/or a morphogen-stimulating agent alone or in combination with other molecules is administered locally or systemically. The morphogenic factors used are e.g. bone morphogenic factors such as OP-1 (osteogenic protein- 1 also known as bone morphogenic protein-7 or BMP-7), OP-2 and various other BMPs. Dosages are proposed to be within the range of 10 ng to 1 g per kg body weight and day, preferentially between 2 μg and 20 μg per kg body weight and day. It is stated that daily doses up to 80 μg OP-1/ kg body weight and day do not induce any ad- verse toxic effects in adult animals. Experiments disclosed describe repair of transected sciatic nerves on rats (guide channel loaded with 1 μg to 5 μg of recombinant OP-1). OP-1 treated animals could bridge the gap of 12 mm whereas sham-treated animals could not.

G. Perides et al. (Neuroscience Letters 187 (1995), p. 21-24) describe a neuroprotec- tive action of OP-1 in a rat model of cerebral ischemia. Rats were injected with 80 or 200 μg of OP-1/ kg body weight interperitoneally (IP) prior to injury. Treatment resulted in a marked reduction of cerebral infarct area compared with vehicle treated animals. Moreover, OP-1 injection after injury reduced mortality compared with untreated animals.

Publication WO-95/05846 (Genetics Institute Inc.) discloses methods and devices for inducing growth of neural cells and repairing neural defects in mammals by ad- ministering, preferably to the site of the lesion, one bone morphogenic protein or a plurality of bone morphogenic proteins (in particular recombinant human (rh) proteins, as rhBMP-2, rhBMP-4, rhBMP-5, rhBMP-6, rhBMP-7 and heterodimers thereof), optionally in combination with other factors e.g. contained in a suitable matrix within an artificial guide channel. The publication proposes as therapeutically effective amounts of the protein to be administered between 0.1 μg to 100 mg per kg body weight, preferably 0J to 100 μg. All the proteins used are purified proteins having a purity of more than 90%. The one described experiment regards severing the sciatic nerve of rats and subsequent repair using a suitable guide channel filled with a collagen sponge containing 0.5 μg of rhBMP-2 in which channel nerve ends are distanced by a gap of 15 mm. Regeneration across the gap was found in rhBMP-2 treated animals but not in sham-treated animals.

Publication WO-97/34626 (Creative Biomolecules Inc.) describes enhancing functional recovery (recovery of sensorimotor functions) following central nervous sys- tern ischemia or trauma, in particular stroke, by morphogens, such as e.g. OP-1, OP- 2, BMP-2, BMP-4, BMP-5, or BMP-6, or by morphogen inducers or agonists of morphogen receptors. The morphogens are administered optionally together with other growth factors, such as e.g. NGF, EGF, PDGF, IGF, FGF, TGF-α, TGF-β. In an experiment disclosed, rats received intracisternally injections of OP-1 (3 or 30 μg OP-1/ kg body weigh for a total of 1, 2 or 8 injections (total 3, 6 or 240 μg OP-1), respectively), after focal cerebral infarction (stroke) induced by middle cerebral artery occlusion (MCAO). The recovery was quantified by functional tests (e.g. limb placing experiments) of animals treated with OP-1 compared to vehicle treated animals for a period of one month after injury. Animals treated with OP-1 recovered faster than sham-treated animals. Higher OP-1 doses give more of the beneficial effect but are connected with pronounced negative effects, as shown by body weight loss. The same findings have been disclosed in the publication T. Kawamata et al. (Neu- roReport 9, 1441-1445, 1998).

The experiments disclosed above show positive results with administration of a bone morphogemc protein a) when contained in a graft in an amount of the order of one to a few micrograms, b) when administered intracistemally or intraperitoneally in an amount of the order of a few micrograms to about 100 μg per kg body weight whereby as far as the disclosed results go, administration of larger amounts is found to give better results at least as far as concerning the specific effect which is regarded.

All the above cited publications teach the use of bone morphogenic proteins for treating nerve tissue in particular for treating nerve tissue after injury. Thereby, in most cases it is suggested that other proteins with beneficial effects may be administered together with bone morphogenic protein(s).

From the publication WO-97/21447 (Stryker Corporation) it is known that the tissue inductive property of morphogenic proteins in a variety of different tissue types (e.g. bone, cartilage, tendon/ligament and neural tissue) is enhanced by stimulatory factors such as insulin-like growth factor (IGF-1), estradiol, fibroblast growth factor (FGF), growth hormone (GF), growth and differentiation factor (GDF), hydrocortisone, insulin, progesterone, parathyroid hormone (PTH), vitamin D or retinoic acid. The ef- feet is believed to be synergistic. Up to six fold enhancement is disclosed.

The proteins which are administered according to the state of the art as shortly discussed above are either isolated from natural sources or they are prepared by recombinant DNA techniques. Description of the invention

It is the object of the invention to provide a composition for enhancing functional recovery in mammals after central and/or peripheral nervous system injury of traumatic or pathological origin, the composition being based on osteoinductive proteins and constituting an improvement of known such compositions regarding the therapeutical effect and effectivity and also regarding unwanted side effects. It is a further object of the invention to provide a method for producing the inventive composition and a method for enhancing in a mammal functional recovery after central and/or peripheral nervous system injury of traumatic or pathological origin.

The above objects are achieved by the composition, by the method for producing the composition and by the method for enhancing functional recovery after central and or peripheral nervous system injury of a mammal as defined in the corresponding independent claims.

The invention is based on the idea of administering to a mammal suffering from central and or peripheral nervous system injury of traumatic or pathological origin a composition comprising a natural protein mixture having bone morphogenic or osteoinductive properties instead of administering pure protein species as proposed by the state of the art. The natural protein mixture is derived from natural tissue, e.g. bone, cartilage and/or tendon/ligament tissue, preferably it is derived from bovine bone tissue. Such natural protein mixtures are known to contain bone morphogenic proteins (BMPs) together with a large number of further proteins.

In the context of this specification, the expression "natural protein mixture" is used in the sense of a mixture containing a plurality of different protein species, each dif- ferent protein species being present in the mixture in one or a plurality of forms and in a relative amount, wherein the selection of protein species, the protein forms and the relative amounts of different protein species substantially correspond to the natural occurrence of the protein mixture. The mixture is preferably of natural origin, i.e. is derived from natural tissue. However once the composition of such a mixture is fully or at least nearly fully elucidated, such a mixture may also be a man-made mixture of recombinantly produced proteins mimicking the natural mixture by showing substantially the same protein selection, protein forms and relative protein amounts.

In the context of the present specification the expression "nervous system injury of traumatic or pathological origin" is to be understood in a very broad manner. It is to include besides injury caused by external forces but also cerebral ischemia and stroke and it is to include acquired and hereditary pathological conditions.

The natural protein mixture preferably contained in the inventive composition for enhancing functional recovery after central and/or peripheral nervous system injury of a human or animal patient is produced by the method as disclosed in the publication US-5290763 (Poser and Benedict) which publication is enclosed herein by reference. The natural protein mixture according to US-5290763 is primarily characterized by the specific process for extracting and purifying it and further by a specific amino acid composition and a specific composition of proteins belonging to specific molecular weight ranges as disclosed in the publication. Furthermore, the protein mixture according to US-5290763 is known to contain the bone morphogemc proteins BMP-2, BMP-3, BMP-4, BMP-5, BMP-6 and BMP-7 (OP-1) and cartilage derived morphogenic proteins (CDMPs), as detectable by interaction with correspond- ing antibodies, and further to contain transforming growth factors-beta TGFβ-1, TGFβ-2 and TGFβ-3, fibroblast growth factor (FGF-1), Osteocalcin, Osteonectin, Albumin, Transferin, Apo-Al-LP and Noggin (antagonist to TGFβ/BMP family members). Up to now, up to 58 percent of the weight of the mixture has been associated with known protein structures, the other 42 or more percent are yet unknown.

The bone morphogemc or osteoinductive properties of the natural protein mixture are shown by inserting in a skin pocket in the ventral thorax region of rats a collagen sponge disc containing the natural protein mixture. After 21 days the rats are sacrificed, the test materials removed, fixed, x-rayed and evaluated histologically for os- teogenesis and calcified tissue proliferation. This bioassay and corresponding scoring characteristics are described in detail in the publication US-5290763 which is in- eluded here by reference.

According to the invention, the natural protein mixture is dissolved in a suitable solvent (e.g. physiological saline solution) in a concentration of between 1 and lOOμg per ml solution and is administered to the cerebrospinal liquid (in the case of a CNS- injury) or as close as possible to the site of injury (in the case of a PNS-injury).

In the rat model, the enhancement of functional recovery from traumatic spinal cord injury being achieved by administering the inventive composition includes faster recovery for reversible and irreversible cases and higher degrees of recovery for irreversible cases, compared with sham-treated animals. The enhancement achieved is in the same order as recovery enhancement achieved by administration of TCP (thi- enylcyclohexypiperidine), an NMDA receptor antagonist known for its favorable effect regarding nerve tissue recovery but also known for its toxicity in humans. Furthermore, the enhancement is achieved by administering doses of the inventive composition having seemingly no negative side effects. In particular, it is found that administration into the cerebrospinal liquid of the inventive composition after central nervous system injury enhances functional recovery to a degree which is much higher than expected from the contents of bone morphogenic proteins detectable in the mixture by corresponding antibodies.

The unexpected effect of the natural protein mixture compared with the effect of single proteins isolated from natural mixtures or produced recombinantly or compared with the effect of "man-made" protein mixtures may have a plurality of reasons:

• Different proteins contained in the natural protein mixture of the inventive composition may act in different fields in which intervention is necessary or highly advantageous for successful neuronal regeneration or functional recovery respectively (see listed fields a to f further up).

• Action of proteins present in the natural protein mixture of the inventive composition may be amplified by synergisms with other proteins present in the mixture.

• There may be yet unknown proteins contained in the mixture which have beneficial effects.

• Proteins contained in the natural protein mixture of the inventive composition may be present in an inactive form and possibly in a form which is not detectable with the aid of antibodies. They may be present in association with protection and/or carrier proteins and therefore enjoy protection in the in vivo surroundings from destruction, inactivation or attraction to receptors at other sites than the injury site and they may be separated from such association and thereby may be rendered active only by factors or enzymes naturally produced at the site of injury. Such effects are observed in vital tissue also wherein tissue inductive proteins or growth factors are present in the tissue incorporated in large mole- cules which have at least partly protein structures and from which specific proteins having a specific effect are "cut" out when needed.

• It may also be that the natural protein mixture of the inventive composition contains further proteins bound to or accompanying other pro- teins having beneficial effects which further proteins facilitate solution in the cerebrospinal liquid and/or transport in the cerebrospinal liquid or in the tissue aimed towards the injury site.

• Furthermore, the natural protein mixture of the inventive composition may contain proteins controlling the process of recovery, i.e. the action profile of the beneficial effects of specific proteins over time in a way comparable to similar processes observed in natural healing processes or even in the embryonic state.

It is found that the natural protein mixture derived from demineralized bone tissue (BP) as disclosed in the Publication US-5290763 is able in vitro to differentiate rat adrenal pheochromocytoma PC 12 cells into their neuronal phenotype, to enhance survival of chick ciliary ganglions and to rescue motoneurons from induced gluta- mate exitotoxicity (see Examples 1 to 3 further below). This demonstrates the favorable effect of BP on the intrinsic capacity of neural cells and on their ability to survive and withstand intoxication.

Furthermore, in vivo experiments (see Examples 4 to 13 further below) show in the rat model a marked enhancement of functional recovery after spinal cord injury for BP-treated animals in comparison with sham-treated animals. This allows the conclusion that the effect of BP is not limited to survival characteristics of neurons but affects the injury cite in a way favourable to controlled neuron regeneration. In particular, BP is shown to in vivo actively accelerate the recuperation of animals which had their spinal cord compressed to induce a reversible paraplegia (see Examples 4 to 9 further below). Recuperation is paralleled by a shortening of the time needed to reach particular behavioral stages, by a volume reduction of the cystic cavity induced by the spinal cord compression and by a decrease in the lesion induced astroglial cellular response. In a more severe model of irreversible paraplegia (see Examples 10 to 13 further below), treatment with BP demonstrates that upon treatment it is feasible to regain almost all of the original motor functions which are normally irreversibly lost in absence of treatment. In this severe model it is shown that regain of the motor functions is also paralleled by a volume reduction of the cystic cavity, by a shortening of the time needed to reach defined behavioral stages and by a reduction of the lesion induced astrocytic response.

The in vivo experiments also show that the administered amount of inventive composition or of the natural protein mixture respectively corresponds to a few tens of nanograms of each one of the bone morphogenic or osteoinductive protein species detectable in the mixture and therefore represents doses which are much smaller (at least by a factor one hundred) than those previously disclosed, e.g. in the publication WO-97/34626 (Creative Biomolecules Inc.) for administration of single protein species. Thereby, the recovery results achieved with the inventive composition and the named doses of the natural protein mixture correspond with the recovery results as disclosed for the larger doses of isolated protein species as published in WO- 97/34626 (Creative Biomolecules Inc.).

Furthermore, it is found that administration of the inventive composition or the natural protein mixture respectively in larger doses than the above stated few micrograms per kg of body weight does not further enhance recovery. On the contrary, the observation of the body weight during the recovery period seems to imply that administra- tion of larger doses is not only less beneficial but may even have negative side- effects.

In view of the similarities of requirements regarding neural regeneration from traumatic injury and from pathological conditions allows the conclusion that similar effects of the inventive composition as found for functional recovery after spinal cord injury are likely to be found for recovery from other sorts of injury of the central or peripheral nervous system and in particular from cerebral ischemia or stroke as well as from aquired or hereditary neurodegenerative diseases such as e.g. Parkinson's disease.

Production of the inventive pharmaceutical composition comprises the steps of providing tissue particles by e.g. cleaning, grinding and demineralizing the tissue from which the natural protein mixture is to be produced, extracting proteins from the tissue particles, purifying the extracted proteins in a plurality of purification steps including ultrafiltration, anion exchange, cation exchange and reverse phase high resolution liquid chromatography, freeze drying the protein mixture and finally dissolving it in a suitable solvent for administration. In particular, the method for producing the inventive composition does not comprise a step of isolating from the protein mixture specific protein species to homogeneity. There is no need for neither protein isolation nor for protein activation.

Examples

The invention is described in further detail in connection with the following Examples and Figures, wherein: Fig. 1 shows a bar graph quantification (in %) of neurite extension and differentiation in PC 12 cells triggered by BP after 4 days and 8 days (d) of treatment (control: NGF triggered differentiation);

Fig. 2 shows a light microscopy picture of PC 12 cells treated with BP for 4 days (control: untreated cells);

Fig.3 shows a bar graph quantification (in %) of cell survival enhanced by BP in E9 chick ciliary ganglions after 4 days of BP treatment (control: untreated cells);

Fig. 4 shows in a bar graph BP induced protection (in % of surviving cells) of telen- cephalic neurons after glutamate induced toxicity at 5 days post intoxication

(control: no intoxication);

Fig. 5 shows in a line graph enhanced functional recovery upon single intrathecal BP injection in rats (reversible paraplegia model; control: TCP-treated animals and sham-treated animals);

Fig. 6 shows in a line graph enhanced functional recovery upon repeated intrathecal BP injections in BP-treated rats (reversible paraplegia model; control: TCP- treated animals and sham-treated animals);

Fig. 7 shows in a line graph the weight (in g) profile of BP-treated rats in the reversible paraplegia model (control: sham-treated animals);

Fig. 8 shows in a bar graph the inclined plane score at 45° and 60° of BP-treated rats in the reversible paraplegia model (control: TCP-treated animals and sham- treated animals); Fig. 9 shows in a bar graph for BP-treated rats in the reversible paraplegia model the time (in days) needed to reach various behavioral stages (control: TCP- treated animals and sham-treated animals);

Fig. 10 is a bar graph quantification of the volume of cystic cavities (in μm²) in BP- treated rats in the reversible paraplegia model (control: TCP-treated animals and sham-treated animals);

Fig. 11 shows in a bar graph the effect of BP on the astroglial cellular density in the white and gray matter of rats in the reversible paraplegia model( control: TCP-treated animals and sham-treated animals);

Fig. 12 shows in a line graph showing the functional recovery enhancement upon repeated intrathecal BP injections in rats (irreversible paraplegia model; control: TCP-treated animals and sham-treated animals).

Fig. 13 shows in a line graph the weight loss (in %) profile of BP-treated rats in the irreversible paraplegia model (control: TCP-treated animals and sham- treated animals);

Fig. 14 shows in a bar graph the time needed (in days) by BP-treated rats in the irreversible paraplegia model to reach various behavioral stages (control: TCP- treated animals and sham-treated animals);

Fig. 15 is a bar graph quantification of the volume reduction of the cystic cavity (in μm²) in BP-treated rats in the irreversible paraplegia model (control: TCP-

Ireated animals and sham-treated animals);

Fig. 16 shows in a bar graph the effect of BP on the astroglial cellular density in the white and gray matter of BP-treated rats in the irreversible paraplegia model (control: untreated animals, TCP-treated animals and sham-treated animals). Example 1

Example 1 demonstrates that a natural protein mixture derived from bovine bones according to the process as disclosed in the Publication US-5290763 (Bone Protein or BP) enhances neurite extension and differentiation in rat adrenal pheochromocy- toma PC12 cells.

Rat adrenal pheochromocytoma PC12 cells (ATCC No. CRL-1721) were obtained from the American Type Tissue Collection. PC 12 cells were cultured in RPMI 1640 media (Life Technologies) in the presence of 10% horse serum and 5% fetal calf serum. Cells were exposed to Bone Protein (BP) for 4 and 8 days at a concentration of 10 μg/ml and 50 μg/ml. NGF (supplemented at 10 μg/ml) served as positive control. BP and NGF were freshly added to the culture medium every second day. At day 4 and 8 after addition of BP the percentage of differentiated cells was determined by counting the cells under a Zeiss Axiovert 25 light microscope. Fig. 2 shows a photomicrograph of PC 12 cells after 4 days in culture in the presence (lower panel) or absence (upper panel) of 10 μg/ml BP.

Quantification as presented in Fig. 1 shows a significant differentiation of PC 12 cells into their neuronal phenotype after 4 days and 8 days. None of the PC 12 cells differentiated into the neuronal phenotype when BP was omitted from the culture media (data not shown). Student t-test, *p < 0.05

Example 2

Example 2 demonstrates that Bone Protein (BP) enhances survival of E9 chick ciliary ganglions in vitro. Ciliary ganglia are motor nuclei of the CNS that are dependent on external trophic support for survival in culture. This effect is relevant to potential trophic effects on motoneurons in vivo.

Chick ciliary ganglia were purified and cultured as previously described (Zurn et al., 1996, J. of Neurosci. Res., 44, 133 - 141 and references herein). BP protein cocktail (10 μg/ml and 50 μg/ml) was administered every second day. On day 4, living chick ciliary ganglions were counted under a Zeiss Axiovert light microscope as previously described (Zurn et al., 1996, J. of Neurosci. Res., 44, 133 - 141 and references herein).

Quantification as presented in Fig. 3 demonstrates that BP significantly enhances survival of dissociated E9 chick ciliary ganglions in vitro. Increasing BP dosage to the culture medium (up to 50 μg/ml) did not improve the protective effect. Student t- test; *p< 0.05.

Example 3

Example 3 demonstrates that Bovine Protein (BP) is able to rescue telencephalic neu- rons from glutamate induced toxicity.

Rat embryonic telencephalic motoneurons were isolated and cultured as previously described (Bloch-Gallego et al., 1991, Development 111, 221 - 232, Henderson et al., 1993, Nature 363, 266 - 270 and references herein). Motoneurons were exposed to 1 mM Glutamate for 1.5 hrs. Glutamate was washed out by replacing the culture media supplemented with BP (10 μg/ml and 50 μg/ml). 24 hours after intoxication, living motoneurons were counted under a light microscope as previously described (Bloch-Gallego et al., 1991, Development 111, 221 - 232, Henderson et al., 1993, Nature 363, 266 - 270 and referenced herein).

As demonstrated in Fig. 4, BP is able to significantly rescue motoneurons from induced glutamate toxicity. 43.64% of neurons survived after 24 hours in control cul- tures. When BP is supplemented to the culture media at 10 μg/ml, 68.68% of neurons survived after 24 hrs. This effect was lost when BP was present in higher concentrations. Student t-test. *p< 0.05

Example 4

Example 4 demonstrates that Bone Protein (BP) when injected in the intrathecal space of a rat that underwent a spinal cord compression leading to reversible paraplegia is enhancing the functional recovery of the animal.

Spinal cord compression in rats (adult female Sprague-Dawley, 260 - 300 gr) was induced by angioplasty balloon inflated with 12 μl water and kept in place for 5 minutes. Lesion model and protocol were performed as disclosed in the publication FR- 9413905.

Anesthesia: 30 min. prior to anesthesia, the animals received a pre-medication of xylazine (5 mg/kg i.m.) and were then anesthetized with 30 mg/kg i.p Pentobarbital and 126 mg/kg i.p. Chloral hydrate.

Spinal cord compression: in order to reproduce the most common type of human spinal cord injury due to contusion, the anesthetized animals were traumatized using an inflatable balloon initially intended for the occlusion of intracranial vascular malformations, according to a method adapted to the rat (Martin et al., 1992, J. Neurosci. Res. 32, 539 - 550). After opening the skin, the paravertebral muscles were retracted from the posterior arches of the vertebrae. A laminectomy of vertebra at T12 thoracic level was performed in order to lay bare the spinal cord. Duration and degree of compression can be adjusted according to the desired severity of the lesion (Reversible paraplegia: 12 μl inflation volume during 5 minutes of compression). The animals were placed in individual cages in a heated environment (28° to 30°C) with food and water ad libitum. An antibiotic (Gentamycine 2 mg/kg i.m.) was adminis- tered during the five days following surgery. The occurrence of hematuria in some animals required a prolongation of antibiotic protection.

BP injection: 30 minutes after spinal cord compression, BP was injected in the intrathecal space (T12 thoracic level) using a Hamilton syringe (day of BP injection indicated by an arrow in Fig. 5). 20 μl of each BP concentration or vehicle were in- jected as a single bolus. As control, animals received a single TCP (1 mg/kg) i.p. injection.

Perfusion: At the end of the experiment, one week after their locomotor recovery, animals were deeply anesthetized with Pentobarbital. Prior to perfusion, animals received an intra-cardiac injection of heparin (370 Ul/kg) and 1% sodium nitrite (0.8 ml) in 0.01 M phosphate buffer (pH 7.2). The animals were perfused through the heart with 300 ml of 4% paraformaldehyde (PF) in phosphate buffered saline (PBS, 130 mM NaCl, 10 mM Na-phosphate, pH 7.0). The spinal cord was removed and maintained in a solution of 4% PF at 4°C until processing of the tissue.

Post operative monitoring or neurological function: The animals were tested daily for their performance on the inclined plane method of Rivilin and Tator (1977, J. Neurosurg. 77, 585 - 590), as well as for their locomotor activity according to a modified version of the Tarlov score. The inclined plane test consisted in placing the animal on a board with an adjustable angled between 0 to 90 degrees. The rats were positioned with their body axis parallel to the long axis of the board. When the board was raised the animal used both its fore - and hindlimbs to maintain its posture. The maximal angle at which the animal can stand for 5 seconds constitutes the inclined plane score. Animals were tested respectively at angles of 30°, 45° and 60° and were rated with the following scores: 30° = 0, 45° = 1 and 60° = 2

The animals were also observed for spontaneous activity in the open field and hindlimb function was scored in the following way: stage 1 score 0 bilateral paralysis; absence of hindlimb movement and weight support stage 2 score 1 unilateral paralysis stage 3 score 2 hindlimb flexion but still no weight support stage 4 score 3 crawling, alternate movement of hindlimbs without bearing stage 5 score 4 gait support and alternate stepping pattern; walking with mild deficit stage 6 score 5 normal walking patter

Cumulative performance (CP): Derived from the Tarlov scale, the functional re- covery score was expressed as a score summing the performance of the animal on the inclined plane test and its motor behavioral stage.

As demonstrated in Fig. 5, BP is able to increase temporarily the cumulative performances (as a direct measure of the functional recovery) of animals that received a single intrathecal bolus injection (indicated by the arrow) 30 minutes after spinal cord compression. These results suggest that a repeated injections of BP can lead to improved cumulative performance scores. Example 5

Example 5 demonstrates that repeated injected of Bovine Protein (BP) in the intra- ventricular space of a rat that underwent spinal cord compression leading to reversible paraplegia is enhancing the functional recovery of the animal.

Spinal cord compression and animal surgery were performed as described in Example 4. Chronic implantation of an intraventricular canula was performed prior to spinal compression. Briefly after anesthesia, the animals were placed into a stereotaxic frame and a canula (Phymed, France) was inserted into the fourth ventricle at the coordinates [Bregma -8.3 mm, laterality 0.0 mm, depth 5.0 mm]. Vehicle, BP and TCP were injected into the cerebro-spinal fluid trough the canula, daily during the first 10 days after spinal contusion (indicated by the arrows). The injected volumes represent 0.7 μg/kg, 7.0 μg/kg and 70.0 μg/kg of BP and 1 mg/kg of TCP.

Post operative monitoring of neurological functions were performed as described in Example 4. Since body weight loss is one major indicator of toxic events. Animals were weighed daily. The corresponding weights are shown graphically in Fig. 9.

As demonstrated in Fig. 6, repeated daily BP injections for 10 days following spinal cord compression dramatically improved the cumulative performances of the treated animals. BP at 7.0 μg/kg was the most efficient treatment in improving cumulative performances. BP at 0.7 μg/kg and at 70.0 μg/kg had no effect on cumulative per- formances. As demonstrated in Fig. 7, BP at 7.0 μg/kg was effective in reversing the body weight loss induced by the spinal cord compression. No such reversion was found in animals treated with 0.1 μg/kg and 70.0 μg/kg. 25

Example 6

Example 6 demonstrates that repeated injections of Bovine Protein (BP) into the in- traventricular cavity of a rat that underwent spinal cord compression leading to reversible paraplegia is shortening the time needed to achieve the maximum score in the inclined plane test.

Spinal cord compression and intraventricular BP injections were performed as described in Example 5; inclined plane tests and scoring as described in Example 4.

As demonstrated in Fig. 8, repeated intraventricular injection of 7.0 μg/kg of BP significantly shortens the time needed by the animal to obtain the maximum score on the inclined plane tests when the plane was tilted to 45° and 60°. Animals with BP- injections of 0.7 μg/kg and 70.0 μg/kg had similar scores as sham-treated animals.

Example 7

Example 7 demonstrates that Bovine Protein (BP) when repeatedly injected in the intraventricular cavity of a rat that underwent spinal cord compression leading to reversible paraplegia shortens the time needed by the animal to achieve defined behavioral stages in an open field.

Spinal cord compression and repeated intraventricular BP injections were performed as described in Example 5. Post operative monitoring of neurological functions were performed as described in Example 4. As demonstrated in Fig. 9, animals receiving repeated intraventricular injections of 7.0 μg/kg of BP showed a remarkable shortening of the time needed to achieve precisely defined behavioral stages (as described in example 4). Treatment of animals with 0.7 μg kg and 70.0 μg/kg of BP had similar profiles as sham-treated animals.

Example 8

Example 8 demonstrates that Bone Protein (BP) when repeatedly injected in the intraventricular cavity of a rat that underwent spinal cord compression leading to reversible paraplegia is able to reduce the volume of the cystic cavity induced by the compression.

Spinal cord compression and repeated intraventricular BP injections were performed as described in Example 5 and animal perfusion was performed as described in Example 4.

Tissue preparation: three blocks of 1 cm length of the fixed spinal cord, corresponding to the lesioned (thoracico-lumbar) area, above the lesion (thoracic) and below the lesion (lumbar) were cut. Vibratome longitudinal spinal sections (50 mm) were cut from these blocks and immersed in PBS containing 0.01% Benzalkonium chloride. A total number of 20 serial sections were collected. Half of the sections sampled from different blocks were counterstained with a solution of 0.9 % Cresyl violet. The other half of the sections was used for glial fibrillary acidic protein (GFAP)-immonodetection. When remaining blood cells were visible, the sections were preincubated in 0.5% hydrogen peroxide in phosphate buffer 60 mM for 15 minutes and then washed twice in PBS 60 mM for 5 min. To increase tissue permeability, the sections were incubated for 5 min. in 0.25% trypsin in the same buffer 27

followed by 2 washes in PBS 60 mM. The sections were incubated in sodium boro- hydride 10 mM for 8 min. previous to incubation for 24 hours at 4°C in rabbit- antisera directed against GFAP, at a dilution of 1:3000 with the addition of 1% nonspecific goat serum (NSS) and 0.1% triton X-100 in PBS (60 mM). GFAP detection was done in Tris-NaCl buffer (TBS 50 mM). After rinses in TBS, sections were successively incubated in goat-anti-rabbit-serum (GAR) (1/200) and in rabbit anti- peroxidase (PAP) diluted 1:200 in Tris saline containing 1% NSS, for 1 hour at room temperature. Under visual control, the peroxidase deposit was revealed with 0.05% diaminobenzidine and 0.01% hydrogen peroxyde in TBS-NaCl 50 mM for 12 min. Finally, sections were transferred into phosphate buffer 60 mM, mounted on glass slides, dried overnight in the hood, cleared in Histosolv π (Labonord) and mounted with Bukitt (Labonord). Control sections were incubated in Pro-immune serum omitting the primary antibody.

Morphometric Analysis: Several parameters were measured: 1) the length of the lesion (cm) at macroscopic level, 2) the maximal surface area (mm²) of the cystic cavity, using a Cohu CCD camera, a frame Grabber Scion LG3 and the NTH image software. Similarly to behavioral analysis, we have pooled the data of length and surface of the lesioned area in establishing a score grid.

Densitometric Analysis of GFAP- immunoreativity: GFAP-immunoreactivity was quantified under a 10X microscopic magnification (Nikon). The image was transmitted onto a Cohu CCD camera and the NIH image software was used for densi- tometry quantification. Longitudinal sections were selected at the lesion site (n=3). Within section, due to the gradient of labeling, two areas were considered: I. the white and gray matter surrounding the cystic cavity and 2. the white and gray matter away from the cavity. From each area 4 individual measurements were made. The analysis was done in 4 animals from each group. As demonstrated in Fig. 10, BP when repeatedly injected intraventricularly at 7.0 μg/kg remarkably reduces the volume of the cystic cavity induced by spinal cord compression. The reduction in the cystic cavity is paralleled by a reduction of the adverse astroglial answer to the injured tissue as demonstrated in Fig. 11.

Example 9

Example 9 demonstrates Bovine Protein (BP) when repeatedly injected in the intraventricular cavity of a rat that underwent spinal cord compression leading to reversible paraplegia is able to reduce to almost normal level the density of astrocytes in the lesion tissue.

Spinal cord compression and repeated intraventricular BP injections were performed as described in Example 5 and animal perfusion was performed as described in Example 4. Tissue preparation, cellular counting and cavity measurements were performed as described in Example 8.

As demonstrated in Fig. 11, treatment with BP (repeated injections of 7.0 μg/kg) is able to dramatically reduce the increase in the astroglial cellular density induced by the compression of the spinal cord. Upon treatment, the animals receiving BP show an astroglial cellular density comparable to that of healthy, uninjured animals. This Example demonstrates that BP actively acts on the environment and prepares a responsive terrain to nerve regeneration. 29

Example 10

Example 10 demonstrates that Bovine Protein (BP) when repeatedly injected in the intraventricular space of a rat that underwent spinal cord compression leading to irreversible paraplegia is enhancing the functional recovery of the animal.

Spinal cord compression and animal surgery was performed as described in Example 4. To induce an irreversible paraplegia, the angioplasty balloon was inflated with 14 μl water and compression executed during 5 minutes. Chronic implantation of an intraventricular canula was performed prior to spinal compression as described in Example 5. Vehicle, BP and TCP were injected daily during the first 10 days after spinal contusion into the cerebrospinal fluid trough the canula as described in Example 5. The injected volumes represented 7.0 μg BP/kg (amount of BP showing the best results in the reversible paraplegia model) and 1 mg TCP/kg.

Post operative monitoring of neurological functions were performed as described in Example 4.

As demonstrated in Fig. 12, injections of 7.0 μg BP/kg (indicated by the arrows) was able to promote an almost complete functional recovery of the treated animals. The final cumulative performances of the BP-treated animals were better than those of the sham-treated animals. Fig. 12 demonstrates that BP is not only able to accelerate recovery (as demonstrated in example 5 and Fig. 6) but is also able to induce nerve regeneration not occurring without BP-treatment, i.e. is able to promote recovery to a higher recovery degree. As demonstrated in Fig. 13, BP actively counteracts the weight loss profile induced by the severe spinal cord compression, and in Fig. 14, 30 -

shows that BP-treated animals are able to regain normal alternate walking while sham-treated are not.

Example 11

Example 11 demonstrates that Bovine Protein (BP) when repeatedly injected in the intraventricular cavity of a rat that underwent spinal cord compression leading to irreversible paraplegia is shortening the time needed by the animal to achieve defined behavioral stages in an open field.

Spinal cord compression and repeated intraventricular BP injections were performed as described in Example 10. Post operative monitoring of neurological functions were performed as described in Example 4.

Fig. 14 shows that animals receiving a repeated intraventricular injection of 7.0 μg/kg of BP need a remarkably shorter time to achieve precisely defined behavioral stages. Untreated animals failed to recover further than the crawling stage (alternate movement of the hindlimb without bearing).

Example 12

Example 12 demonstrates that Bovine Protein (BP) when repeatedly injected in the intraventricular cavity of a rat that underwent spinal cord compression leading to irreversible paraplegia is able to reduce the volume of the cystic cavity induced by the compression. 31

Spinal cord compression and repeated intraventricular BP injections were performed as described in Example 10 and animal perfusion was performed as described in Example 4. Tissue preparation and detection was performed as described in Example 8.

Fig. 15 shoes that repeated intraventricular injections of 7.0 μg/kg of BP remarkably reduce the volume of the cystic cavity induced by spinal cord compression leading to irreversible paraplegia. The reduction of the surface is paralleled by a reduction of the adverse astroglial answer to the injured tissue as demonstrated in Fig. 16.

Example 13

Example 13 demonstrates that Bovine Protein (BP) when repeatedly injected in the intraventricular cavity of a rat that underwent spinal cord compression leading to irreversible paraplegia is able to reduce to almost normal level the density of astrocytes in the lesion tissue.

Spinal cord compression and repeated intraventricular BP injections were performed as described in Example 10 and animal perfusion was performed as described in Ex- ample 4. Tissue preparation and detection was performed as described in Example 8.

Fig. 16 shows that treatment with BP (repeated injections of 7.0 μg/kg) is able to dramatically reduce the increase in the astroglial cellular density induced by a compression of the spinal cord leading to irreversible paraplegia. Animals receiving BP show an astroglial cellular density comparable to that of healthy, uninjured animals. This Example demonstrates that BP acts on the injury environment and prepares a responsive terrain to nerve regeneration.

Claims

32C L A I M S

1. Composition for enhancing functional recovery of a mammal from central and/or peripheral nervous system injury of traumatic or pathological origin and including cerebral ischemia and stroke, the composition being characterized by comprising a natural protein mixture derived from bone, cartilage and/or tendon/ligament tissue and having bone morphogenic or osteoinductive characteristics.

2. Composition according to claim 1, characterized in that the natural protein mixture is derived from bovine bone tissue.

3. Composition according to claim 2, characterized in that the natural protein mixture comprises at least one bone morphogenic protein (BMP) detectable by a corresponding antibody.

4. Composition according to claim 3, characterized in that the natural protein mixture comprises BMP-2, BMP-3, BMP-4, BMP-5, BMP-6, BMP-7, CDMPs, TGF-βl, TGF-β2, TGF-β3, FGF-1, osteocalcin, osteonectin, albumin, transferin. apo-Al-LP and noggin.

5. Composition according to one of claims 1 to 4, characterized in that the natural protein mixture is dissolved in a solvent suitable for administration to the cerebrospinal liquid or close to an injury site. - 33

6. Composition according to claim 5, characterized in that the solvent is physiological saline solution.

7. Composition according to one of claims 1 to 6, characterized in that the natural protein mixture is contained in the solvent in a concentration of between 1 and 100 μg per ml solution.

8. Method for producing a composition for enhancing recovery of a mammal from central or peripheral nervous system injury of traumatic or pathological origin and including cerebral ischemia and stroke, the method comprising, for producing a natural protein mixture having osteoinductive or bone morphogenic proper- ties, a step of providing tissue particles derived from bone, cartilage and/or tendon/ligament tissue, a step of extracting proteins from the tissue particles, a plurality of purification steps including ultrafiltration, anion exchange, cation exchange and reverse phase high resolution liquid chromatography and a step of freeze drying and no step of isolating single protein species, the method further comprising a step of dissolving the natural protein mixture in a suitable solvent for administration.

9. Method according to claim 8, characterized in that the tissue particles are particles of mammalian tissue.

10. Method according to claim 8 or 9, characterized in that the step of providing tissue particles comprises cleaning, grinding and demineralizing bone tissue. 34 -

11. Use of a natural protein mixture derived from bone, cartilage and/or tendon/ligament tissue and having bone morphogenic or osteoinductive properties for preparing a composition for enhancing functional recovery of a mammal from central or peripheral nervous system injury of traumatic or pathological origin and including cerebral ischemia and stroke.

12. Use of a natural protein mixture containing BMP-2, BMP-3, BMP-4, BMP-5, BMP-6, BMP-7, CDMPs, TGF-βl, TGF-β2, TGF-β3, FGF-1, osteocalcin, os- teonectin, albumin, transferin, apo-Al-LP and noggin for preparing a composition for enhancing functional recovery of a mammal from central or peripheral nervous system injury of traumatic or pathological origin and including cerebral ischemia and stroke.

13. Use of a natural protein mixture derived from bone, cartilage and/or tendon/ligament tissue, having bone morphogenic or osteoinductive properties and further containing noggin, for preparing a composition for enhancing functional recovery of a mammal from central or peripheral nervous system injury of traumatic or pathological origin and including cerebral ischemia and stroke.

14. Use of a natural protein mixture derived from bone, cartilage and/or tendon ligament tissue and having bone morphogenic or osteoinductive properties for preparing a composition to be administered to the cerebrospinal liquid or in- jected close to an injury site of a mammal for enhancing recovery from central or peripheral nervous system injury of traumatic or pathological origin and including cerebral ischemia or stroke. 35 -

15. Use of a natural protein mixture derived from bone, cartilage and/or tendon/ligament tissue and having bone morphogenic or osteoinductive properties for preparing a composition for reducing the volume of a cystic cavity induced at the site of a traumatic spinal cord lesion of a mammal.

16. Use of a natural protein mixture derived from bone, cartilage and/or tendon/ligament tissue and having bone morphogenic or osteoinductive properties for preparing a composition for reducing astroglial response induced by a spinal cord injury of a mammal.

17. Use of a natural protein mixture derived from bone, cartilage and/or ten- don/ligament tissue and having bone morphogenic or osteoinductive properties for preparing a composition for accelerating functional recovery of a mammal from a spinal cord injury, cerebral ischemia or stroke.

18. Use of a natural protein mixture derived form bond, cartilage and or tendon/ligament tissue and having bone morphogenic or osteoinductive properties for preparing a composition for raising the level of recovery of a mammal from partially reversible spinal cord injury, cerebral ischemia or stroke.

19. Use of a natural protein mixture derived from bone, cartilage and/or tendon/ligament tissue and having bone morphogenic or osteoinductive properties for preparing a composition for protecting neurons against exitotoxicity or glu- tamate intoxication.

20. Use of a natural protein mixture derived from bone, cartilage and/or tendon/ligament tissue and having bone morphogenic or osteoinductive properties - 36 -

for preparing a composition for treating a mammal suffering from a neurodegen- erative disease or from Parkinson's disease.

21. Method for enhancing in a mammal functional recovery from central or peripheral nervous system injury of traumatic or pathological origin and including cerebral ischemia and stroke, the method comprising a step of administering to the mammal a composition comprising a solvent and a therapeutically effective amount of a natural protein mixture derived from bone, cartilage and/or tendon/ligament tissue and having bone morphogenic or osteoinductive properties.

22. Method according to one of claims 21, characterized in that the nervous system injury regards the central nervous system and the composition is administered to the cerebrospinal liquid of the mammal.

23. Method according to claim 21, characterized in that the nervous system injury regards the peripheral nervous system and the composition is injected close to the injury site.

24. Method according to one of claims 21 to 23, characterized in that the composition is administered in successive administrations during a recovery period.

25. Method according to claim 24, characterized in that the successive administrations are daily administrations. - 37

26. Method according to one of claims 21 to 25, characterized in that the therapeutically effective amount of the natural protein mixture is in the range of 1 to 10 μg per kg body weight and per day.