AU9368498A - Neuronal rescue agent - Google Patents

Description

WO 99/15192 PCT/NZ98/00139 NEURONAL RESCUE AGENT This invention is directed to a new therapeutic use of activin and its analogs. More 5 particularly, it is directed to the use of activin and its analogs as neuronal rescue agents and/or neuronal phenotype restoratives. BACKGROUND 10 Members of the transforming growth factor (TGF)-3 family display multiple roles as hormonal, paracrine and autocrine regulators of cellular function, growth and differentiation. Amongst them, activin is emerging as an important factor in an increasingly diverse range of biological processes including hormone production and secretion, modulation of testicular and ovarian cell function, induction of 15 erythropoiesis, initiation of early embryonic development and more recently as a promoter of wound repair. As another example of its various properties, activin has been shown to be neuro prophylactic and to promote survival of neurons subjected to a subsequent toxic 20 event (see, for example, Krieglstein et al; "TGF-0 superfamily members promote survival of midbrain dopaminergic neurons and protect them against MPP+ toxicity" EMBO Journal Vol. 14 No. 4 pp 736-742 (1995)). However, there has been no suggestion to date that activin has application as a neuronal rescue agent. 25 As used herein, neuronal "rescue" is distinct from "prophylaxis". A neuronal "rescue" agent is one which, when administered after an insult prevents neurons from dying which would otherwise be destined to die. In contrast, a neuro prophylactic agent is one which protects neurons against insult but only where the agent is present at the time of or before the insult. 30 It is the applicants finding that activin is a neuronal rescue agent as well as being neuro-prophylactic which primarily underlies the present invention. A number of known compounds are neuro-prophylactic but are not neuronal rescue 35 agents, eg. flunarazine (Gunn and Gluckman, (1991)), GM1 ganglioside (Simon et al., WO 99/15192 PCT/NZ98/00139 2 (1993)), NGF-P (Rabizadeh et al., (1994)), bFGF (Mattson et al., (1993)), TNFox (Barger et al., (1995)) and TNFP (Barger et al., (1995)). Other compounds are neuronal rescue agents without being neuro-prophylactic, eg. IGF-1 (Guan et al., (1993), Gwag et al., (1995)) and BDNF (Gwag et al., (1995)). Still a further group of 5 compounds are both (for example GPE). The fact that a compound is neuro prophylactic therefore cannot be predictive of that compound being effective as a neuronal rescue agent. SUMMARY OF THE INVENTION 10 In a first aspect, the present invention provides a method of treating a patient to rescue neurons otherwise destined to die as the result of a prior neuronal insult which comprises administering to said patient activin or an analog thereof after said insult in an amount sufficient to prevent the neurons from dying. 15 In a further aspect, the present invention provides a method of treating a patient to rescue neurons otherwise destined to die as the result of a prior neuronal insult which comprises increasing the active concentration of activin within said patient after said insult such that the neurons are prevented from dying. 20 "Neuronal insult" is used herein in its broadest possible sense and includes neuronal insults due to trauma, toxins, asphyxia, hypoxia-ischemia (HI) and disease. 25 "Activin" as used herein means activin A, activin B or activin AB of mammalian origin and preferably of human, porcine, bovine or murine origin. "Analog" is used herein to mean a variant of activin through insertion, deletion or substitution of one or more amino acids but which retains at least substantially 30 equivalent biological activity to activin. The applicants have found that the neuronal rescue role of activin is mediated through the activin type II receptor. Specifically contemplated analogs are therefore those which bind to and activate the activin type II receptor. 35 WO 99/15192 PCT/NZ98/00139 3 In a further aspect, the invention provides a method of treating a patient to rescue neurons otherwise destined to die as the result of prior neuronal insult which comprises activating the activin type II receptors of neuronal cells of a patient who has suffered a prior neuronal insult. 5 Activation can be through administration of a ligand which binds to, and activates, the receptor. Preferably, activin type II receptor activation is effected through administration of 10 activin or an analog thereof. In still a further aspect, the invention provides the use of activin or an analog thereof, or a ligand which binds to and activates activin type II receptors, in the preparation of a medicament for rescuing neurons otherwise destined to die as a 15 result of a prior neuronal insult. In a further aspect, the invention provides a method of treating a patient to restore the phenotype of neurons degenerating as a result of a prior neuronal insult which comprises administering to said patient activin or an analog thereof after said insult 20 in an amount effective to restore the phenotype of said neurons. In still a further aspect, the invention provides a method of treating a patient to restore the phenotype of neurons degenerating as a result of a prior neuronal insult which comprises increasing the active concentration of activin within said patient 25 after said insult such that the phenotype of said neurons is restored. In yet a further aspect, the invention provides a method of treating a patient to restore the phenotype of neurons degenerating as a result of a prior neuronal insult which comprises activating the activin type II receptors of neuronal cells of a patient 30 who has suffered a prior neuronal insult. In still yet a further aspect, the invention provides the use of activin or an analog thereof, or a ligand which binds to and activates activin type II receptors, in the preparation of a medicament for restoring the phenotype of neurons degenerating as 35 a result of a prior neuronal insult.

WO 99/15192 PCT/NZ98/00139 4 DESCRIPTION OF THE DRAWINGS While the present invention is broadly as defined above, those persons skilled in the 5 art will appreciate that it is not limited thereto and that it further includes embodiments of which the following description provides examples. In addition, the invention will be better understood through reference to the accompanying drawings in which: 10 Figure 1 is a graph showing the rescue of striatal neurons by activin following HI; Figure 2 is a graph showing the rescue of striatal cholinergic phenotype neurons by activin after HI; 15 Figure 3 shows the results of immunoreactive staining of the activin-type II receptor in neurons of the striatum (Figure 3A), the substantia nigra (Figure 3B), thalamus (Figure 3C) and paraventricular nucleus (Figure 3D); 20 Figure 4 is a graph showing the rescue of calbindin, cholinergic, NADPH diaphorase and parvalbumin phenotypic neurons by activin following an intrastriatal quinolinic acid lesion (Figures 4A to 4F); Figure 5 shows the restorative effects of rhActivinA treatment on striatal 25 interneurons immunostained for ChAT (Figures 5A to 5D); Figure 6 shows the areas of the rat brain analysed for the effect of activin and inhibin on neuronal rescue (Figure 6A: cortex and hippocampus; Figure 6B: striatum); 30 Figure 7 shows the comparative effects of activin A and inhibin A on the survival of neurons following HI (Figure 7A: Activin, Figure 7B: Inhibin); and WO 99/15192 PCT/NZ98/00139 5 Figure 8 shows the results of immunohistochemical staining of an Alzheimer's brain (Figures 8A to 8G). DESCRIPTION OF THE INVENTION 5 As broadly defined above, the present invention relates primarily to neuronal rescue. This is the maintenance of neuronal cells which would otherwise be destined to die as a result of a prior neuronal insult. The cells are therefore "rescued" from death and not merely protected prophylactically. 10 The invention also relates to phenotype restoration. The applicants have found that degenerating neuronal cells which have lost their phenotype as the result of a prior neuronal insult can be phenotypically restored. 15 The applicants have found that neuronal rescue\phenotype restoration is able to be effected using two approaches. The first approach is through a focus upon activin. The applicants have found that increasing the effective concentration of activin within a patient following neuronal insult rescues neurons and/or restores their phenotype. 20 Activin itself is critical to this approach. There are three isoforms of activin, designated activin A, activin B and activin AB. Structural analysis shows that activins are disulphide linked dimers of two subunits, which are two distinct 14 kD 0 subunits (PA and fB) (Ying, 1989, Vale et al., 1990). Activins (28 kD) are 25 homodimers of the two 0 subunits. The mature PA or PB subunit has 116 or 115 amino acids respectively including 9 cysteines with no glycosylation sites. The two P subunits share about 85% homology within each species, and are also highly homologous across species. The mature BA subunits are completely identical across porcine, bovine, human and murine species and the mature OB subunit only 30 has differences at four amino acid positions (Esch et al., (1987)). Recently, the molecular cloning of activin PC, OD, PE subunits has been reported (Fang et al., 1996). Activins made up from or including one or more of these subunits are in no way intended to be excluded. 35 WO 99/15192 PCTINZ98/00139 6 All of the above forms of activin are contemplated for use in this invention. The preferred form of activin for use in this invention is activin A. This is available from National Institute of Health, USA. 5 Most conveniently, the effective concentration of activin will be increased through direct administration using either activin itself or an activin prodrug (a form which is cleaved within the body to release activin). It is however not the applicant's intention to exclude increasing activin concentration through administration of 10 either activin agonists (substances which effect a direct increase in production or activity of activin within the body, eg. FSH, cAMP [activator of protein kinase A activator], 12-O-tetradecanoylphorbol 13-acetate [TPA, a protein kinase A activator], TGF-P3, IL- 13 and TNF-ot) or inhibitors of activin antagonists (substances which bind activin or otherwise prevent or reduce the action of activin within the body). These 15 latter compounds exert an indirect effect on effective activin concentrations through the removal of an inhibitory mechanism, and include substances such as estradiol. Follistatin is one such substance. It is a single chain glycosylated protein, which was first isolated from porcine and bovine follicular fluid. The amino acid sequence 20 of follistatin is distinct from those of the activin subunits and any other proteins in the TGF-P3 family. However, across species, follistatin amino acid sequence is highly conserved with over 98% homology. As a binding protein for activin, follistatin has been observed to have different 25 actions on the biological activities of activin. Follistatin can directly bind to activin to neutralize its function in many systems (Mathews, 1994). However, it has also been suggested to have an ability to enhance activin action through either bringing activin to its receptors or maintaining a high local concentration of activin. Thus follistatin may exert a dual effect in mediating activin activities (Macconell et al., 30 1996) as both agonist and antagonist. Inhibin can also be regarded as an activin antagonist. The mechanism by which this is achieved is not completely understood but is likely to include competitive binding to the activin receptor. Therefore, effecting a decrease in the production or 35 action of inhibin is likely to increase the effective concentration of activin.

WO 99/15192 PCT/NZ98/00139 7 Another possibility is administration of a replicable vehicle encoding activin to the patient. Such a vehicle (which may be a modified cell line or virus which expresses activin within the patient) could have application in increasing the concentration of 5 activin within the patient for a prolonged period. It is also contemplated that activin analogs can be employed in this invention. As used herein, "analog" means a protein which is a variant of activin through insertion, deletion or substitution of one or more amino acids but which retains at 10 least substantial functional equivalency. A protein is a functional equivalent of another protein for a specific function if the equivalent protein is immunologically cross-reactive with, and has at least substantially the same function as, the original protein. The equivalent can be, for 15 example, a fragment of the protein, a fusion of the protein with another protein or carrier, or a fusion of a fragment with additional amino acids. For example, it is possible to substitute amino acids in a sequence with equivalent amino acids using conventional techniques. Groups of amino acids normally held to be equivalent are: (a) Ala, Ser, Thr, Pro, Gly; 20 (b) Asn, Asp, Glu, Gin; (c) His, Arg, Lys; (d) Met, Leu, Ile, Val; and (e) Phe, Tyr, Trp. 25 Functional equivalency of activin analogs can also be readily screened for by reference to the ability of the analog to both bind to and activate the appropriate receptor. In this case, the receptor is an activin type II receptor. Similar to TGF-P receptors, there are two types of activin receptors, termed activin 30 type I receptor (ActRI) and activin type II receptor (ActRII). ActRII was the first receptor identified for activin and for other members in the TGF-3 superfamily (Mathews and Vale, 1991). The mature ActRII is comprised of 494 amino acids which includes a small 116 amino acid extracellular ligand binding domain, a single transmembrane domain and an intracellular serine/threonine kinase domain, 35 which is common in the TGF-3 superfamily. Over 90% sequence homology of ActRII WO 99/15192 PCT/NZ98/00139 8 has been observed across species, which is consistent to the high similarity of mature activin 3A sequences in various species (Mathews, 1994). A distinct but closely related activin receptor ActRIIB and its four isoforms have subsequently been characterized (Mathews, 1994; Mathews et al., 1992; Attisano et al., 1992). 5 ActRII and ActRIIB are approximately 50-60% identical in the ligand-binding domain and 60-70% identical in the kinase domain. ActRII, ActRIIB and its isoforms all bind to activin with high affinity. Collectively, ActRII, ActRIIB and their isoforms are referred to herein as "activin type 10 II receptors". Activin type II receptors are distinct from Activin type I receptors (ActRI). ActRI and its isoforms have been cloned using PCR with oligonucleotides based on the ActRII sequence. These receptors also have highly conserved serine kinases. However, 15 cells expressing ActRI but not ActRII cannot bind to activin alone (Mathews, 1994) and hence the capacity to bind to activin type II receptors is considered critical to this aspect of the invention. Activin initiates signal transduction across the membrane through both ActRI and 20 ActRII which can form a stable complex with the ligand (Willis et al., 1996; Mathews, 1994). ActRII binds activin and then associates with a type I receptor. This is followed by auto- and trans-phosphorylation between the two receptors and the initiation of intracellular signalling (Smith J, 1995; Willis et al., 1996). 25 This leads to the applicant's second approach to neuronal rescue and/or phenotypic restoration. This approach focuses upon activin type II receptors as defined above, and upon effecting neuronal rescue\phenotypic restoration through the use of ligands which both bind to and activate these receptors. 30 It will be appreciated that activin and its analogs are ligands which achieve this. Indeed, the use of activin and activin analogs represents a preferred aspect of the invention. However, it should be appreciated that this approach is not restricted to the use of activin and its analogs but also extends to any ligand which fulfils the functional requirement of both binding to and activating (stimulating) the activin WO 99/15192 PCTINZ98/00139 9 type II receptor. Implicit in this will be the ability of the ligand to effect the association with ActRI needed to initiate intracellular signalling. Such stimulatory ligands can be identified by a screening protocol employing at 5 least the ligand binding domain of an activin type II receptor. This screening method can, for example, utilize the expression of the Act II receptor in Xenopus oocytes using standard recombinant DNA methods and measurement of the Act II receptor-mediated signal transduction evolved by novel stimulatory ligands. Further classical "grind and bind" ligand-binding experiments can also be utilized. 10 Here, whole brain or specific brain regions would be homogenized and the specific binding of novel compounds to the Act II receptor characterized. This technique allows further characterization of specificity and affinity (potency) of the compound for the Act II receptor complex. 15 For the intended therapeutic application, the active compound (activin, analog or ligand) will be formulated as a medicament. The details of the formulation will ultimately depend upon the insult to be remedied and the route of administration, but will usually include combination of the active compound with a suitable carrier, vehicle or diluent. 20 Insults which can be treated in accordance with the invention include any prior neuronal insult. These include trauma, toxins, asphyxia, ischemia and disease, particularly neurodegenerative diseases such as Alzheimer's disease (including both early and late onset forms) Parkinson's disease, Huntington's disease and Lewy 25 Body disease. Also included is peripheral neuropathy. To be effective as a neural rescue\phenotype restoring agent in treating the above insults, a variety of administration routes can be used. Examples include peripherally in conditions where the blood brain barrier is disrupted (ie. ischemia), 30 intracerebroventricularly (ICV), intraventricular administration involving neurosurgical insertion of a ventricular cannula with an abdominal pump and reservoir and intraparenchymal (ie. at the site of action).

WO 99/15192 PCT/NZ98/00139 10 Dosage rates will also be formulation- and insult-dependent. However, by way of example, the recommended dosage rate of activin A formulated for injection would be in the range of Ing/ 100g to 100g/ 100g administered centrally. 5 The invention, in its various aspects, will now be illustrated by the experimental section which follows. It will however be appreciated that the experiments are non limiting. EXPERIMENTAL 10 EXPERIMENT 1 Materials and Methods Animal preparation The following experimental protocol followed guidelines approved by the University 15 of Auckland Animal Ethics Committee. HI injury was induced in weaned 21 day old Wistar rats using a modified version of the Levine rat preparation as described previously (Sirimanne et al., (1994)). Rats were maintained on a 12 hour cycle of light and dark and given free access to food and water throughout the study. Rats of both sexes weighing between 40-49g were selected, anaesthetized with halothane 20 and underwent double ligation of the right carotid artery following exposure through a midventral neck incision. After surgery rats were allowed to recover in a carefully controlled environment of 34oC with >80% relative humidity for a period of 2 hours were then exposed to 15 minute 8% PO 2 and then lightly anaesthetized again using Saffan (2mg/kg), ip). Infusions were made into the right lateral cerebral ventricle 25 with the aid of a metal cap fitted over the head of the anaesthetized animal as described by Jirikowski (1992), to allow correct placement of the infusion needle (30G X 25.4mm). A randomized block design was used to enable treatment studies to be performed in batches of litter mates of 12-16 infant rats. 30 Treatment and tissue preparation Recombinant human activin A at a dose of 1lg or vehicle was administered in a single bolus to rats two hours after hypoxia as described above (Activin A, n=23; vehicle, n=23). Activin was given as a 204l infusion at rate of 3pl/min. Only those rats where the solution flowed were accepted. (=all rats). Rats were euthanised 72 35 hours after hypoxia by sodium pentobarbital overdose and brains were collected for WO 99/15192 PCT/NZ98/00139 11 histological processing after in situ fixation by transcardial perfusion with 30ml saline (0.9%) then neutral buffered formalin (10%). Fixed brains were dehydrated through graded alcohols, defatted in chloroform and embedded in paraffin. Serial 4mm sections were then cut between at 3.5mm (for striatum) anterior to the ear 5 bars [Paxinos] and stained with acid-fuschin/thionin, or with an antibody to choline-acetyltransferase (ChAT) which identifies cholinergic neurons (Boehringer Mannheim). Double-labelling Immunohistochemistry 10 Double immunohistochemical labelling to localise ActRII to specific neuronal subtypes was performed on rat brain sections at the level of striatum (cholinergic, parvalbumin, calbindin) or substantia nigra (dopaminergic). Brains were processed as described above. Antiserum generated towards a synthetic fragment of mActRII (482-494) was a generous gift of Professor Wylie Vale (The Salk Institute, San Diego, 15 CA, USA). Sections were dewaxed, rehydrated and incubated for 30 minutes in 0.6% hydrogen peroxide to quench endogenous peroxidase activity. Incubation with ActRII antibody (1:1000) diluted in 1.5% normal goat serum and phosphate buffered saline (PBS) was carried out overnight in a humidified environment. PBS-washed sections were then incubated with biotinylated secondary antiserum for 1 hour, 20 washed, then incubated with an avidin-biotinylated horseradish peroxidase complex (Vector Labs, Burlingame, CA, USA) for a further 1 hour. Signal for ActRII was visualised using 0.05%, 3,3'diaminobenzidine tetrahydrochloride (Sigma Chemicals, St Louis, MO, USA). Sections were then reincubated with either parvalbumin (1:1000), calbindin D28K (1:1000) or ChAT (1:50, identifies cholinergic neurons) or 25 tyrosine hydroxylase (1:1000 identifies dopaminergic neurons) as described above except that immunostaining for the second antibody was visualised using benzidine dihydrochloride (BDHC, Sigma). Non-specific immunostaining was determined using normal rabbit serum, normal mouse serum and incubating sections in the absence of primary antibodies. 30 The total number of surviving neurons were counted in the CA1/2 region of the hippocampus, 3 selected areas of the cortex and 4 selected areas of the striatum or the total number of ChAT immunopositive neurons in the dorsolateral striatum were counted in the injured half brain of both treatment groups (vehicle and activin A) 35 with a light microscope (Leica, Germany) and were compared.

WO 99/15192 PCT/NZ98/00139 12 Results The above experiment provided the following results: 5 - The number of surviving striatal neurons after HI was significantly (p<0.014) increased by activin treatment. This can be seen from Figure 1. - Activin specifically rescued striatal cholinergic neurons after HI. This can be seen from Figure 2. 10 - Activin type II receptor immunoreactivity was found in cholinergic, parvalbumin and calbindin neurons of the striatum and in dopaminergic neurons of the substantia nigra. Staining was predominantly in the cell body, axons and processes of neurons but clearly did not show labelling in 15 the nucleus. ActRII immunoreactivity was observed in a number of thalamic nuclei particularly; the lateral and ventral posterior thalamic nucleic, and reticular thalamic nucleus. Staining was also seen in the caudate putamen, layer VI of the cerebral cortex, zona incerta, dorsal lateral geniculate nucleus, and to a lesser intensity in the medial septum 20 and in the hippocampus. This can be seen from Figure 3. Conclusions The above results lead to the following conclusions: 25 1. Activin A is applicable for the treatment of brain injury after HI. Since activin was significantly protective in the striatum and 95% of striatal neurons are GABAergic, activin is likely able to rescue GABAergic neurons in vivo, which has relevance for treating the loss of GABAergic neurons in the striatum seen in Huntington's disease. 30 2. Activin specifically rescues cholinergic neuronal phenotype after HI and activin type II receptor is colocalized on cholinergic neurons. This indicated an application for activin in treating the hypofunction of central cholinergic neurotransmission seen in the human neurodegenerative condition known as 35 Alzheimer's disease. This further indicated another application for activin in WO 99/15192 PCT/NZ98/00139 13 delaying or preventing the loss of cholinergic neurons in the nucleus basalis of Meynert. 3. Activin type II receptor colocalizes with tyrosine hydroxylase (dopaminergic) 5 neurons of the substantia nigra. This indicates an application for activin A in delaying or preventing the loss of dopamine neurons seen in the human neurodegenerative condition known as Parkinson's disease. EXPERIMENT 2 10 Methods The following experiments were carried out in accordance with University of Auckland Animal Ethical Committee Regulations. All efforts were made to keep the number of animals used to a minimum and to minimize animal suffering. 15 Animal Preparation and Treatment Two groups (n=4-5) of male Wistar rats weighing 360-400g (University of Auckland Animal Breeding Unit) were anaesthetized with 75 mg/kg sodium pentobarbital and positioned in a stereotaxic apparatus (Kopf Instruments, USA). Quinolinic acid (QA) 20 lesioning was performed using a modified version of that described by Alexi et al., (1997). Unilateral intrastriatal injections of 100 nmol in 2 ptl of QA (Sigma Chemicals) were made over 5 minutes using a Hamilton syringe at the coordinates 0.5 mm anterior to Bregma, 3.0 mm lateral to midline and 5.0 mm ventral to skull from the atlas of Paxinos and Watson (1986). A 22-gauge guide cannula (Plastics 25 One, USA) was permanently fixed into place 0.5 mm dorsal to the ventral coordinate (i.e., at +0.5, +3.0, -4.5). Groups of 4-5 animals received daily 1 ptl injections of either rhactivinA (0.73 tg/ptl/day, National Hormone and Pituitary Program, CA, USA) or phosphate-buffered saline (PBS, pH 7.4) vehicle beginning at the time of cannulation for 7 days following QA lesioning. 30 Immunohistochemistry At 7 days post QA lesioning rats were perfused intracardially with PBS followed by 4% paraformaldehyde in 0.1M phosphate buffer (PB, pH 7.4). Brains were post fixed overnight at 4oC in this same solution and cryoprotected serially in 10% and 35 30% sucrose in PB for 2-4 days at 4oC. Floating 30 4m coronal striatal sections WO 99/15192 PCTINZ98/00139 14 were stained by avidin-biotin-peroxidase immunocytochemistry. GABAergic neurons were stained using an antibody against feline glutamate decarboxylase-67

(GAD

67 ) made in rabbit from Chemicon International (#AB108). Sections were blocked with 10% normal goat serum and 0.25% triton X-100 in PBS for 1 hour at 5 room temperature. Sections were rinsed 3 times in PBS and incubated in primary

GAD

67 at 1:200 with 2% normal goat serum in PBS at 4 0 C for 3 days. Sections were then rinsed and incubated in the secondary antibody, biotinylated anti-rabbit IgG (Amersham RPN480, USA), at 1:1000 with 1% normal goat serum in PBS at 4 0 C overnight. Sections were rinsed and incubated in streptavidin biotinylated 10 peroxidase complex (Amersham RPN1050) at 1:100 in PBS at room temperature for 4 hours. Sections were rinsed and developed with diaminobenzidine (Sigma) in deionized water with 0.03% hydrogen peroxide for 5 minutes. Acetylcholine neurons were stained using an antibody against human choline acetyl-transferase (ChAT) made in goat from Chemicon (#AB144P). Sections were stained as for 15 GAD 67 , except that incubation in the secondary antibody was for 3 hours at room temperature and the solutions were as follows: the blocking solution was 2% rabbit serum in PBS, the primary antibody was diluted 1:100 with 2% rabbit serum in PBS, the secondary antibody was anti-goat at 1:200 in PBS with 2% rabbit serum, the peroxidase complex was AB cocktail (Vector Laboratories, USA) at 9 tl of A plus 20 9 1 il of B per ml of PBS, and the development solution was unchanged. The calcium binding protein containing neurons, calbindin, parvalbumin and calretinin, were stained as for ChAT neurons, except that incubation in the AB peroxidase cocktail was for 1 hour at room temperature and the solutions were as follows: primary antibodies against chicken calbindin (#300) and carp parvalbumin (#235) were made 25 in mouse and human calretinin (#7696) was made in rabbit (Swant Antibodies, Bellinzona, Switzerland), the blocking solution was 10% horse (calbindin and parvalbumin) or goat (calretinin) serum in PBS, the primary antibody was diluted 1:9000 in PBS with 10% respective serum and 0.25% triton X-100, the secondary antibody was anti-mouse or anti-rabbit at 1:250 in PBS, the peroxidase complex 30 and development solutions were unchanged. In order to control for staining intensity, all sections from all the animals for each antibody were processed at the same time. Negative controls consisted of omitting the primary antibody.

WO 99/15192 PCT/NZ98/00139 15 Histochemistry NADPH-diaphorase (NADPHd) cells were visualized by incubating sections in 1 mg/ml NADPH (ICN Biochemicals) and 0.1 mg/ml nitrotetrazolium blue (ICN) dissolved in PB with 0.3% triton X-100 for 2 hours at 37 0 C. 5 Quantification of Cell Counts Stained cells (ChAT, NADPHd, parvalbumin and calretinin) were counted at 100X magnification in a visual field 950 x 730 ptm. The location of the field was 465 pm below the cannula tip at the lesion center (0.5 mm anterior to Bregma), and 10 extended rostrocaudally along a parallel to sections at +360 pm rostral to the lesion center and -360 ptm caudal to the lesion center. This was done to assess the effects of radial diffusion of solutions from the cannula on the rostrocaudal plane while keeping a fixed distance along the dorsoventral and mediolateral planes. A cell was counted if it had an intact cell body and at least a neurite stump. Cells were 15 counted in a blind coded fashion. Cell counts were converted from the number of cells in the visual field to the number of cells per mm 2 by dividing by 0.6935 (950 x 730 p.m = 0.6935 mm 2 ). Stained cells (GAD 67 and calbindin) were counted at 100X magnification in a visual field 475 x 365 jtm. Cell counts were converted from the number of cells in the visual field to the number of cells per mm 2 by dividing by 20 0.1734 (475 x 365 pm = 0.1734 mm 2 ). Cell counts are presented as means of cells/mm 2 ± SEM at each of the three levels (rostral, center, caudal) in the striatum for both the contralateral (control) and ipsilateral (lesion and treatment) sides from 4-5 animals per group. To reveal the percentage of cell phenotypic survival, ipsilateral values (cells/mm 2 ) were divided by contralateral values for each animal 25 and means were calculated for each of the two groups. Comparisons between phenotypic cell survival at each coronal level were performed on percentage values to standardize for the differing sizes of the populations examined. Statistical analysis of cell counts were made using a multi-variate analysis of variance (ANOVA) for either repeated (phenotypic marker) or non-repeated (treatment) 30 measures, followed by a post hoc analysis using the Student Newman-Keuls test for statistical differences (Sigma Stat, Jandel Scientific, USA). Results 35 The results are shown in Figures 4A-4F and Figures 5A-5D.

WO 99/15192 PCT/NZ98/00139 16 Unilateral intrastriatal infusion of quinolinic acid produced a partial but significant loss by 7 days in the number of striatal neurons immunoreactive for glutamate decarboxylase (to 51.0±5.8% of unlesioned levels, see Fig 4A), calbindin (to 58.7± 5 5.1%, see Fig 4B), choline acetyl-transferase (to 68.6±6.1%, see Fig 4C), NADPH diaphorase (to 47.4±5.4%, see Fig 4D), parvalbumin (to 58.8+4.1%, see Fig 4E) and calretinin (to 60.6±8.6%, see Fig 4F) in adult rats that were administered intrastriatal phosphate buffered saline for 7 days following quinolinic acid. 10 In contrast, in rats that received intrastriatal recombinant human ActivinA once daily for 7 days following quinolinic acid, phenotypic degeneration was significantly attenuated in several populations of striatal neurons. Treatment with ActivinA had the most potent protective effect on the striatal cholinergic interneuron population almost completely preventing the lesion induced decline in choline acetyltransferase 15 expression (to 95.1±5.8% of unlesioned levels, see Fig 4C). ActivinA also conferred a significant protective effect on parvalbumin (to 87.5+7.7%, see Fig 4E) and NADPH diaphorase (to 77.5±7.5%, see Fig 4D) interneuron populations but failed to prevent the phenotypic degeneration of calretinin neurons (to 56.6±5.5%, see Fig 4F). Glutamate decarboxylase67 and calbindin-staining nerve cells represent largely 20 overlapping populations and both identify striatal GABAergic projection neurons. ActivinA significantly attenuated the loss in the numbers of neurons staining for calbindin (to 79.7±6.6%, see Fig 4B) at 7 days following quinolinic acid lesioning. The results shown in Figures 5A-5D illustrate the restorative effects of rhActivinA 25 treatment on striatal stained for ChAT. Compared with the vehicle treated control (Figure 5B) the ChAT neurons in both the contralateral striatum control (Figure 5A) and the QA-lesioned side treated with rhActivinA (Figures 5C and 5D) have healthy cell bodies and elongated neuronal 30 processes containing ChAT. Figure 5D is a high power magnification of Figure 5C. Conclusions Exogenous administration of ActivinA rescues both striatal interneurons (labelled with choline acetyltransferase, parvalbumin, NADPH-diaphorase) and striatal WO 99/15192 PCT/NZ98/00139 17 projection neurons (labelled by calbindin) from excitotoxic lesioning with QA. It also restores the phenotype of degenerating neurons. The ability of Activin A to rescue striatal GABAergic projection neurons from 5 degeneration following QA lesioning indicates an application for activin A in delaying or preventing the loss of GABAergic projection neurons which are preferentially lost in the human neurodegenerative disease, Huntington's disease. These results further indicate that activin A can restore the ChAT neuronal 10 phenotype after a QA lesion in the striatum. This indicates an application for activin A in restoring ChAT neurons in the human neurodegenerative condition Huntington's disease and also Alzheimer's disease. EXPERIMENT 3 15 The Effect of Activin A and Inhibin A on Neuronal Survival Methods ICV injection Moderate HI brain injury (15 minutes hypoxia) was first induced in 21-day-old 20 Wistar rats. Two hours after hypoxia, all rats were lightly anaesthetised with intraperitoneal injection of Saffan. An infusion needle (30G x 25.4 mm) was placed into the right lateral cerebroventricle of the rat brain with the aid of a metal skull template as described by Jirikowski (Jirikowski, 1992). Each rat was injected either with the drug (activin A or inhibin A) diluted with vehicle solution in a total volume 25 of 20 4l, or 20 ml vehicle solution. The injections into vehicle and treatment groups were performed simultaneously with a micro-infusion pump at a rate of 3 .l/minute. After infusion, the rats were left to recover in an incubator maintained at 34oC and relative humidity of 85-95%. Once awake, the rats were transferred into their holding cages and fed food and liquid ad libitum. 30 In the activin A treatment study, 46 rats were divided into two batches, weight and sex matched groups. 23 rats were injected with 1 pg rh activin A diluted with sterilized vehicle solution containing 0.15 M NaC1, 0.05 M Tris and 0.1% BSA, (pH7.4). The other 23 rats were injected with 20 il vehicle solution. 35 WO 99/15192 PCTINZ98/00139 18 In the inhibin A treatment study, 36 rats were divided into two groups as above. 18 rats were injected with 1 pg inhibin A diluted with sterilized vehicle solution containing 0.9% natural saline, 0.1% BSA, (pH 7.4) and the other 18 rats were injected with 20 41 corresponding vehicle solution. 5 Analysis of histological outcome All rats were euthanised 3 days after ICV injection with an overdose of phenobarbital. The rats were transcardially perfused with 0.9% saline followed by 4% PFA, and the brains removed and embedded in paraffin. Symmetric serial 10 coronal sections (4 mm) were cut and stained with thionin/acid-fusion for live/dead neurons (Sirimanne et al., 1994). The histological outcome of neuronal survival was examined with light microscopy (Leica) in the cortex, hippocampus and striatum in the injured half of the brain according to a reference of rat brain anatomy (Paxinos and Watson, 1982), as these areas suffer most of the neuronal loss in the moderate 15 HI brain injury model (Sirimanne et al., 1994). Only cells with a morphology like live neurons were counted, while dead neurons or cells with morphology like glial were not included. For each of the above three brain areas, one coronal section was used for each brain. 20 One coronal section between 2.8mm to 3mm from Bregma was used for the analysis of cerebral cortex. Live neurons within three rectangular areas each measuring 1000 im x 5000 pm were counted with a microscope grid (Figure 6A). The areas selected covered layer II to layer V of the parietal cortex, where selective neuronal loss occurred. 25 One coronal section between 3.3 mm to 3.5 mm posterior to Bregma was used for the hippocampus. All surviving neurons in the hippocampal CA1/2 region of the injured hemisphere were counted (Figure 6A). 30 One coronal section between 0.3 to 0.8mm from Bregma was used for the striatum. Four areas (Areas 1-4) in the upper 2/3 of the injured striatum where most damage is generally found in this model were selected with a microscope grid (Figure 6B). The size of the areas were: Area 1-3: 1000 m x 2000 4m, Area 4: 2000 m x 2000 tm. 35 WO 99/15192 PCT/NZ98/00139 19 Results The results are shown in Figure 7. In both the treatment and vehicle control group, dead cells identified by the uptake 5 of acid fusion were found in the ligated (right) hemisphere 3 days after hypoxia, particularly in regions such as the CA1/2 region of hippocampus, the upper two thirds of the striatum and layer III-IV of the cerebral cortex. The severity of cell loss ranged from non-selective to massive in these areas. No cell loss was observed in the non-injured hemisphere. 10 The neural rescue effect of rhActivin A after moderate H 1 brain injury was assessed in the hippocampus, striatum and cortex as follows: Hippocampus 15 All the live neurons in the hippocampal CA1/2 region in the injured (right) side were counted. In the activin A treated group, the mean number of surviving neurons increased to 107+39 as compared with 67+36 in the vehicle control group (mean+SEM, p<0.05, Mann-Whitney rank sum test). 20 Striatum Four representative areas in the dorsal (area 1), lateral (area 2) and central (area 4) region of the upper two thirds of the injured striatum were selected. When the number of live neurons in area 1 and area 2 were added together to represent the dorsolateral striatum, there were more surviving cells in the treatment group 25 (94+11) than in the control group (50+10) (mean+SEM, P<0.05, t-test). Cortex The total numbers of live neurons in the selected areas in the parietal cortex in the activin A treated group (310+18) and the vehicle control group (322+17) were not 30 significantly different. The effect of inhibin A on neuronal survival was also assessed in the hippocampal CA1/2 region, striatum and cerebral cortex as described in the activin A treatment 35 study. The differences of surviving neurons between the treatment and control WO 99/15192 PCT/NZ98/00139 20 group in the above three areas were not statistically significant. However, compared with the control group, there was a trend showing that inhibin A decreased the number of surviving neurons in the dorsolateral striatum from 71+10 to 70+9 and in the cortex from 311+18 to 289+19 while increased the number of surviving neurons 5 in the hippocampal CA1/2 region from 100+26 to 123+31 (mean+SEM). Conclusion When compared to the results achieved using activin A, it can be clearly seen that inhibin is not effective as a neuronal rescue agent. This is consistent with inhibin 10 binding to but not activating the activin type II receptor and therefore being a functional antagonist of activin. This in turn supports the applicant's findings regarding the critical role activation or stimulation of the activin type II receptor plays in neuronal rescue. 15 EXPERIMENT 4 Methods Follistatin expression The expression of follistatin peptide in the brain of an Alzheimer's sufferer was 20 examined immunohistochemically with a Vectastain kit (Vector labs, USA). Serial formalin fixed and paraffin embedded post mortem human brain tissue (from the medial frontal gyrus) were cut at 4 mm, which were then dewaxed and rehydrated. Non-specific staining which arises from the endogenous peroxidase activity and non-specific binding sites on the brain sections were blocked by 0.3% H 2 0 2 and 1% 25 normal goat serum respectively. The sections were incubated in three steps with a polyclonal anti-human follistatin (1:500, a gift from Hiromu Sugino, the Institute for Enzyme Research, The University of Tokushima, Japan) at room temperature overnight, biotinylated goat-anti-rabbit-IgG (1:100) at 37oC for 1 hour and horseradish-peroxidase (1:100) at 370C for 1 hour. The sections were washed with 30 0.01M PBS three times after each of the above incubation. The signals were washed with 0.01M PBS three times after each of the above incubations. The signals on the sections were detected with a solution made with DAB tablets (Sigma, USA), which showed a brown colour. After being washed with water, the sections were counterstained with thionin, dehydrated and mounted with DPX. 35 WO 99/15192 PCT/NZ98/00139 21 To identify brain cells which express follistatin, a double immunohistochemistry method was used. The brown immunostaining for follistatin was first obtained on brain sections as described above. As the follistatin positive cells have a morphology similar to glial, the above three step immunostaining procedure with 5 monoclonal anti-GFAP (1:500, Sigma) as the primary antibody was used to identify astrocytes. The signals were detected with BDHC which showed a blue colour. Cortical senile plaques which contain P-amyloid are characteristic of Alzheimer's disease. To further examine the possible spatial relationship between the follistatin 10 expressing cells and senile plaques, a triple immunohistochemistry method was used. After dewaxing and rehydration, brain sections were treated with concentrated formic acid for 5 minutes, which can enhance the intensity of the immunostaining of senile plaques. The sections were thorough washed with water and blocked for possible non-specific staining. A monoclonal anti-human P-amyloid 15 (1:1000, Dako, Denmark) was used as the primary antibody in the three step immunostaining procedure and Ni-DAB tablets (Sigma) were used as a blue chromagen. The brain sections were then immunostained for follistatin and GFAP as described above. 20 Activin PA subunit expression Activin PA peptide expression and its spatial relationship to senile plaques was examined with a similar double immunohistochemistry method as described above. Briefly, the brain sections were first immunostained for P-amyloid positive plaques with Ni-DAB as the chromagen. The sections were then incubated with polyclonal 25 anti-activin BA subunit (1:250, a gift from Professor Wylie Vale, The Salk Institute for Biological Studies, USA) as the primary antibody. DAB was used as the brown chromagen for this staining. Results 30 The results are shown in Figures 8A-8G. Compared with an age matched control brain (Figure 8A), follistatin protein is increased in an Alzheimer's disease patient brain (Figure 8B). Double immunohistochemistry labelling showed that follistatin expression was mainly 35 colocalised with glial fibrillary acidic protein (GFAP) positive astrocytes (blue WO 99/15192 PCT/NZ98/00139 22 staining) and also possibly microglia (Figures 8C and 8D). Figure 8D also shows staining for P-amyloid (red brown staining). PA activin immunoreactivity (brown staining) was also found inside P-amyloid positive plaques (black staining) in the Alzheimer's disease brains (Figure 8E = control tissue, Figure 8F = Alzheimer's brain 5 tissue). Figure 8G shows activin type II receptor staining in neurons that 'look' damaged in Alzheimer's disease brain tissue. The black staining is P-amyloid. 10 Conclusion These results indicate that of follistatin upregulation (an activin inhibitor) may have a role in the pathophysiology of Alzheimer's disease, while the upregulation of activin may indicate the induction of an endogenous neuronal rescue mechanism in the Alzheimer's disease brain. 15 INDUSTRIAL APPLICATION The invention therefore provides new approaches to neuronal rescue and neuronal phenotype restoration. These involve firstly increasing the active concentration of 20 activin in a patient following neuronal insult and secondly the activation of the activin type II receptors localized on neuronal cells, again following neuronal insult. The approaches of the invention have application in the treatment of patients who have suffered neuronal insult particularly as the result of a neurodegenerative 25 disease. Two such diseases of considerable interest are Alzheimer's disease and Parkinson's disease. Patients suffering from these diseases will benefit greatly by a treatment protocol able to rescue damaged and dying neuronal cell populations. Other applications of the present invention are in the treatment of Huntington's 30 disease and peripheral neuropathy. Still more generally, the invention has application in the rescue of neurons destined to die following insult in the form of trauma, toxin exposure, asphyxia or hypoxia ischemia. 35 WO 99/15192 PCT/NZ98/00139 23 In addition to neuronal rescue, the present methods also show the capability of restoring phenotypes in injured, degenerating and diseased neurons, particularly those of the following phenotypes: 5 ChAT, calbindin, NADPHd, parvalbumin, GABAergic and glutamatergic neurons. It will be appreciated by those persons skilled in the art that the above description is provided by way of example only and that numerous changes and variations can 10 be made while still being with the scope of the invention as defined by the appended claims.

WO 99/15192 PCT/NZ98/00139 24 REFERENCES Alexi T., Venero J.L. and Hefti F. (1997) Protective effects of neurotrophin-4/5 and transforming growth factor-a on striatal neuronal phenotypic degeneration after 5 excitotoxic lesioning with quinolinic acid. Neuroscience 78:73-86. Attisano L., Wrana J.L., Cheifet S., and Massague J. (1992). Novel activin receptors: distinct genes and alternative mRNA splicing generate a repertoire of serine/threonine kinase receptors. Cell, 68:97-108. 10 Barger S.W., Horster D., Funikawa K., Goodman Y., Kieglestein J., Mattson M.P. (1995). Proceedings of the National Academy of Sciences, USA, 92:9328-9332. Esch F.S., Shimasaki S., Mercado M., Cooksey K., Ling N., Ying S., Ueno N.X., and 15 Guillemin R. (1987). Structural characterization of follistatin: a novel follicle stimulating hormone release-inhibiting polypeptide from the gonad. Molecular Endocrinology, 1(11):849-855. Fang J.M., Yin W.S., Smiley E., Wang S.Q., and Bonadio J. (1996). Molecular 20 cloning of the mouse activin beta(E) subunit gene. Biochemical and Biophysical Research Communications, 228(3):669-674. Guan J., Williams C.E., Gunning M., Mallard E.C. and Gluckman P.D. (1993). The effects of IGF- 1 treatment after hypoxic-ischemic brain injury in adult rats. Journal 25 of Cerebral Blood Flow and Metabolism, 13:609-616. Gwag B.J., Koh J.Y., Chen M.M., Dugan L.L., Behrens M.M., Lobner D., Choi D.W. (1995). BDNF or IGF-1 potentiates free radical-mediated injury in cortical cell cultures. NeuroReport, 7(1):93-96. 30 Gunn A.J., Gluckman P.D. (1991). Flunarizine, a calcium channel antagonist, is not neuroprotective when given after hypoxia-ischemia in the infant rat. Developmental Pharmacology and Therapeutics, 17:205-209.

WO 99/15192 PCT/NZ98/00139 25 Jirikowski G.F. (1992). A non-surgical technique for accurate intracerebral injections in rat. Journal of Neuroscience Methods, 42:115-118. Macconell L.A., Barth S., and Roberts V.J. (1996). Distribution of follistatin 5 messenger ribonucleic acid in the rat brain - implications for a role in the regulation of central reproductive functions. Endocrinology, 137:2150-2 158. Matthews L.S., and Vale W.W. (1991). Expression cloning of an activin receptor, a predicted transmembrane serine kinase. Cell, 65:973-982. 10 Matthews L.S., Vale W.W., and Kinter C.R. (1992). Cloning of a second type of activin receptor and functional characterization in Xenopus embryos. Science, 255:1702-1705. 15 Mattson M.P., Barger S.W., Cheng B., Lieberburg I., Smith-Swintosky V.L., Rydel R.E. (1993). Beta-amyloid precursor protein metabolites and loss of neuronal Ca 2+ homeostasis in Alzheimer's disease. Trends in Neurosciences, 16:409-414. Paxinos G. and Watson C. (1986). The Rat Brain in Stereotaxic Coordinates. 2nd ed. 20 Sydney: Academic Press. Rabizadeh S., Bittle C.M., Butcher L.L., Bredsen D.E. (1994). Expression of the low affinity nerve growth factor receptor enhances amyloid peptide toxicity. Proceedings of the National Academy of Sciences, USA, 9(22):10703-10706. 25 Simon R.P., Chen J. Graham S.H. (1993). GM1 ganglioside treatment of focal ischemia - a dose-response and microdialysis study. Journal of Pharmacology and Therapeutics, 265(1):24-29. 30 Sirimanne E.S., Guan J., Williams C.E., and Gluckman P.D. (1994). Two models for determining the mechanisms of damage and repair after hypoxic-ischemic injury in the developing rat brain. Journal of Neuroscience Methods, 55:7-14.

WO 99/15192 PCT/NZ98/00139 26 Vale W., Hsueh A.J.W., River C. and Yu A. (1990). The inhibin/activin family of hormones and growth factors. In Peptide Growth Factors and their Receptors, Handbook of Experimental Pharmacology. M Sporn and AB Roberts, eds. (Berlin: Springer-Verlag), pp 211-248. 5 Ying S.Y. (1989). Inhibins, activins and follistatins. Journal of Steroid Biochemistry, 33:705-713.

Claims (35)

1. A method of treating a patient to rescue neurons otherwise destined to die as the result of a prior neuronal insult which comprises administering to said patient activin or an analog thereof after said insult in an amount 5 sufficient to prevent the neurons from dying.

2. A method according to claim 1 wherein activin is administered to said patient after insult.

3. A method according to claim 2 wherein the activin administered is selected from activin A, activin B and activin AB. 10

4. A method according to claim 2 wherein the activin administered is activin A.

5. A method according to claim 1 wherein an analog of activin is administered to said patient after insult.

6. A method of treating a patient to rescue neurons otherwise destined to die as the result of a prior neuronal insult which comprises increasing the 15 active concentration of activin within said patient after said insult such that the neurons are prevented from dying.

7. A method according to claim 6 wherein the active concentration of activin is increased through administration of activin to said patient after insult.

8. A method according to claim 7 wherein the activin administered is activin A, 20 activin B or activin AB.

9. A method according to claim 7 wherein the activin administered is activin A.

10. A method according to claim 6 wherein the active concentration of activin is increased through administration of an activin agonist.

11. A method of treating a patient to rescue neurons otherwise destined to die 25 as the result of prior neuronal insult which comprises activating the activin type II receptors of neuronal cells of a patient who has suffered a prior neuronal insult. WO 99/15192 PCTINZ98/00139 28

12. A method according to claim 11 wherein activin type II receptor activation is effected through administration of a ligand which binds to and activates the receptor.

13. A method according to claim 11 wherein activin type II receptor activation is 5 effected through administration of activin.

14. A method according to claim 13 wherein the activin administered is activin A, activin B or activin AB.

15. A method according to claim 13 wherein the activin administered is activin A. 10

16. A method according to claim 11 wherein activin type II receptor activation is effected through administration of an activin analog.

17. A method according to any one of the preceding claims wherein the prior neuronal insult is due to trauma, toxins, asphyxia, hypoxia-ischemia or neurodegenerative disease. 15

18. A method according to claim 17 wherein the neurodegenerative disease is Huntington's disease.

19. A method according to claim 17 wherein the neurodegenerative disease is Alzheimer's disease.

20. A method according to claim 17 wherein the neurodegenerative disease is 20 Parkinson's disease.

21. A method according to claim 17 wherein the prior neuronal insult is peripheral neuropathy.

22. The use of activin or an analog thereof in the preparation of a medicament for rescuing neurons otherwise destined to die as a result of a prior 25 neuronal insult.

23. The use of a ligand which binds to and activates activin type II receptors in the preparation of a medicament for rescuing neurons otherwise destined to die as a result of a prior neuronal insult. WO 99/15192 PCT/NZ98/00139 29

24. The use of claim 22 or claim 23 wherein the medicament is to rescue neurons otherwise destined to die as a result of neurodegenerative disease.

25. The use of claim 24 wherein the neurodegenerative disease is Huntington's disease. 5

26. The use of claim 24 wherein the neurodegenerative disease is Alzheimer's disease.

27. The use of claim 24 wherein the neurodegenerative disease is Parkinson's disease.

28. The use of claim 22 or claim 23 wherein the medicament is to rescue 10 neurons otherwise destined to die due to peripheral neuropathy.

29. The use of claim 22 or claim 23 wherein the medicament is to rescue neurons otherwise destined to die due to trauma, toxins, asphyxia or hypoxia-ischemia.

30. A method of treating a patient to restore the phenotype of neurons 15 degenerating as a result of a prior neuronal insult which comprises administering to said patient activin or an analog thereof after said insult in an amount effective to restore the phenotype of said neurons.

31. A method of treating a patient to restore the phenotype of neurons degenerating as a result of a prior neuronal insult which comprises 20 increasing the active concentration of activin within said patient after said insult such that the phenotype of said neurons is restored.

32. A method of treating a patient to restore the phenotype of neurons degenerating as a result of a prior neuronal insult which comprises activating the activin type II receptors of neuronal cells of a patient who has 25 suffered a prior neuronal insult.

33. A method according to any one of claims 30 to 32 which is to restore the phenotype of ChAT, calbindin, NADPHd, parvalbumin, GABAergic and/or glutamatergic neurons. WO 99/15192 PCT/NZ98/00139 30

34. The use of activin or an analog thereof in the preparation of a medicament for restoring the phenotype of neurons degenerating as a result of a prior neuronal insult.

35. The use of a ligand which binds to and activates activin type II receptors in 5 the preparation of a medicament for restoring the phenotype of neurons degenerating as a result of a prior neuronal insult.