EP1937820A2 - System for delivering neuronal calcium sensor-1 (ncs-1) - Google Patents

Abstract

The present invention provides a system capable of delivering neuronal calcium sensor-1 (NCS-1), or a nucleotide sequence encoding NCS-1 to a target cell, for promoting neurite outgrowth. The system may be used in the manufacture of a pharmaceutical composition for the treatment of a condition such as spinal cord injury.

Description

SYSTEM

FIELD OF THE INVENTION

The present invention relates to system, in particular a system capable of delivering a protein or a nucleotide sequence to a target cell.

The system may be capable of delivering or causing the expression of neuronal calcium sensor- 1 (NCS-I ), for the promotion of neurite outgrowth.

BACKGROUND TO THE INVENTION

The injury of actor Christopher Reeve in 1995 drew the world's attention to the tragedy of spinal cord injury. Accidents and violence cause an estimated 10,000 spinal cord injuries each year in the US, and more than 200,000 Americans live day-to-day with the disabling effects of such trauma. The incidence of spinal cord injuries peaks among people in their early 20s, with a small increase in the elderly population due to falls and degenerative diseases of the spine.

Unlike nerve cells or neurons of the peripheral nervous system (PNS) which can carry signals to and from the limbs, torso, and other parts of the body, neurons of the adult CNS have limited intrinsic capacity to regenerate, e.g. following injury.

Medical care of spinal cord injury has advanced greatly in the last 50 years. During World War II, injury to the spinal cord was usually fatal. While post-war advances in emergency care and rehabilitation allowed many patients to survive, methods for reducing the extent of injury are virtually unknown. Although techniques to reduce secondary damage, such as cord irrigation and cooling, were first tried 20 to 30 years ago, the principles underlying effective use of these strategies were not well understood. Significant advances in recent years, including an effective drug therapy for acute spinal cord injury (methylprednisolone) and better imaging techniques for diagnosing spinal damage, have improved the recovery of patients with spinal cord injuries. Current care of acute spinal cord injury involves three primary considerations. First, physicians must diagnose and relieve cord compression, gross misalignments of the spine, and other structural problems. Second, they must minimize cellular-level damage if at all possible. Finally, they must stabilize the vertebrae to prevent further injury.

The ultimate hope, of course, is not just to minimize damage, but to foster recovery. Progress has been made in devising neural prostheses that can substitute for some of the functions lost after spinal cord injury. There is also considerable interest in the possibility of inducing nerve cells to regenerate, for example in cases of avulsion injury.

The present inventors have shown that exogenous delivery of the transcription factor retinoic acid receptor β2 (RARβ2) via a lentiviral vector into adult rat dorsal root ganglion (DRG) neurons promotes axonal regeneration of injured sensory axons and restores functional recovery in the rhizotomy model of spinal cord injury (Wong et al., Paper submitted). These results suggest that RARβ2 can activate regeneration programs in injured neurons to promote axonal outgrowth.

Neuronal calcium sensor- 1 1 (NCS-I: a member of the neuronal calcium sensor protein family) has been described in EP-A-I 250 931 (and corresponding US2003/0159158) as target for therapeutic intervention in CNS disorders such as Schizophrenia, Alzheimer's disease, Parkinson's disease, major depression, bipolar disorder, anxiety disorders. Garcia N et al, (8 Aug 2005 online) J Neurosci Res, 82(1): 1-9 suggests that NCS-I is involved in the formation and function of the presynaptic nerve terminal part of the neuromuscular junction during synaptogenesis and in adult mammals. Xiao-Liang Chen et al (2001) J Physiology, 532(3): 649-659 suggests that overexpression of rat NCS-I in rodent NG108- 15 cells enhances synapse formation and transmission.

The inventors have also surprisingly shown however that NCS-I is upregulated by RARβ2 and that overexpression of NCS-I, like RARβ2, promotes neurite outgrowth in adult neurons. SUMMARY OF THE PRESENT INVENTION

The present inventors have shown that expression of NCS-I causes axonal outgrowth, making the expression of NCS-I in vivo of considerable therapeutic interest.

In a first aspect, the present invention provides a system capable of delivering neuronal calcium sensor-1 (NCS-I), or a nucleotide sequence capable of encoding NCS-I, to a target cell, for promoting neurite outgrowth.

Delivery systems include a non-viral and viral vector systems. Viral vector systems include retroviral and lentiviral systems. The system may, for example be derivable from EIAV.

The vector system of the present invention may be capable of promoting neurite outgrowth and may be used, for example, for treating spinal cord injury.

In a second aspect, the present invention relates to a pharmaceutical composition comprising a system according to the first aspect of the invention. The composition may also comprise RARβ2, or a nucleotide sequence capable of encoding RARβ2.

The system of the first aspect of the invention, the composition of the second aspect of the invention, or an agent capable of promoting the expression or activity of NCS-I may be used to promote neurite outgrowth, for example, in the treatment of spinal cord injury to activate regeneration processes in injured neurons.

In a third aspect, the present invention relates to a method of promoting neurite outgrowth by administering to a subject a system of the first aspect of the invention, a composition of the second aspect of the invention, or an agent capable of promoting the expression or activity of NCS-I.

In a fourth aspect, the present invention relates to the use of a system according the first aspect of the invention to deliver NCS-I, or a nucleotide sequence capable of encoding NCS-I, to a cell in vitro. The invention also provides a cell containing or expressing NCS- 1, produced by such a method. Such a cell may be used for implanting or transplanting into a subject, for example to promote neuron regeneration. DESCRIPTION OF THE FIGURES

Figure 1. RARβ2 enhances neurite outgrowth from DRG neurons via activation of cAMP signalling pathway

DRG neurons (500 neurons/well) are plated on reduced laminin substrate (0.1 μg/ml) and transduced with lentiviral vectors expressing (a) control β-galactosidase (EIAV-LacZ) or (b) RARβ2 (EIAV-RARβ2) or at a multiplicity of infection of 10. Vector expression of β- galactosidase (β-gal) or RARβ2 are indicated in green while neurite outgrowth is detected by βlll-tubulin in red. The average length of the longest neurite (c) and the number of neurons possessing neurites (d) in transduced cultures are measured. RARβ2 cultures are also incubated in an adenylate cyclase inhibitor 2', 5'-dideoxyadenosine DDA (100 μM; e) or a PKA inhibitor Rp-cAMP (50 μM; f). (g) cAMP levels are assessed in transduced cultures using a competitive immunoassay. Results are the mean (±SEM) of at least 8 experiments, each carried out in triplicate, (h) The average length of the longest neurite is measured in EIAV-LacZ and EIAV-RARβ2 transduced neurons in normal medium and in the presence of DDA or Rp-cAMP at various concentrations. All experiments are performed in triplicate with n=4 in each group unless otherwise stated. For measurement of neurite outgrowth 150-220 neurons are assessed in each group and data is represented as mean + SEM. Black bar; EIAV-LacZ-transduced DRGs; red/grey bar, EIAV-RARβ2- transduced DRGs. ***P<0.001, **P<0.01, *P<0.05 (Students' unpaired 2-tailed Mest). In (h) Asterisk (*) denotes significance of P<0.05 from EIAV-RARβ2 with no treatment group. Scale bar: lOOμm.

Figure 2. DRG neurite extension in spinal cord co-culture

DRGs transduced with EIAV-LacZ or EIAV-RARβ2 vectors are co-cultured with transduced or naϊve spinal cord, (a) Neurite extension as detected by GAP43 staining (red) is absent in EIAV-LacZ DRG neurons (green) that are cultured on EIAV-RARβ2 transduced spinal cord, (b) EIAV-LacZ neurons that are grown on the dorsal root (DR) projected fibers up to the DREZ but did not enter the spinal cord (SC) (indicated by white arrow). The central and peripheral regions of the DREZ are delineated by GFAP staining in blue (marked by a white dotted line), (c) EIAV-RARβ2 transduced DRG neurons (green) extended several neurites on EIAV-LacZ transduced cord and (d) neurons that are present in the DR projected extensions up to and beyond the central portion of the DREZ (indicated by white arrows). Neurons that are not transduced with EIAV-RARβ2 do not project neurites on the spinal cord or beyond the DREZ (indicated by white arrowheads). The percentage of transduced neurons bearing neurites on naive or transduced cord (e) as well as the length of the longest neurite on the spinal cord from EIAV-LacZ or EIAV- RARβ2 transduced neurons (T) are measured. All experiments are performed in triplicate with n=4 in each group. For measurement of neurite outgrowth 50-60 neurons are assessed in each group and data is represented as mean ± SEM. **P<0.01, Students' unpaired 2- tailed t-test. Scale bars: 100 μm (a, c); 400 μm (b, d).

Figure 3. EIAV-mediated gene transfer in the spinal cord and DRG

(a) β-galactosidase expression is observed in the cuneate fasciculus and the dorsal horn of the spinal cord 3 weeks following infusion of EIAV-LacZ vector into the spinal cord, (b) Retrograde transport of the vector via the dorsal root resulted in β-galactosidase expression in the ipsilateral DRG. With the EIAV-RARβ2 vector, RARβ2 mRNA is detected in the ipsilateral DRG and not in the contralateral uninjected side (c & d respectively). RARβ2 protein is observed in the ipsilateral DRG and not in the contralateral DRG (e & f respectively) and in embryonic day 14 rat DRG (g). (h) To assess for vector expression in the various subclasses of DRG neurons, β-galactosidase expression (red) in the DRG is colocalized immunohistochemically with NF200 (marker for large-diameter myelinated neurons), CGRP (marker for small-diameter peptidergic neurons) or IB₄ (marker for small- diameter non-peptidergic DRG neurons), all indicated in green. Overlapping expression is illustrated in yellow. Scale bars: 200 μm (a-d); 50 μm (e-h).

Figure 4. Axonal regeneration of injured sensory neurons across the DREZ In all micrographs, dorsal roots (DR) approach the cord from the left and the peripheral region of the DREZ is delineated by the presence of laminin (blue). Injured sensory afferents are labeled transganglionically with BDA tracer 10 days before termination of experiment, (a) At five weeks post-lesion, BDA-labeled axons (red) are present in the dorsal root up to, but not beyond, the DREZ in EIAV-LacZ animals, (b) By contrast, several BDA axons (indicated by white arrows) are observed in the DREZ after EIAV- RARβ2 expression, (c) Quantification of axonal regeneration across the DREZ is assessed by counting the number of BDA fibers at respective distances from the DREZ (distance in graph represents peripheral part of the DREZ to the central region from left to right, with the DREZ denoted at 0 μm). Results are the mean number of BDA fibers per section (± SEM). *P<0.05, two-way RM ANOVA, Tukey post-hoc), (d-f) NF200- (d), CGRP- (e) and IB₄- (f) expressing afferents (green, indicated by white arrows) are detected central to the DREZ in EIAV-RARβ2, however these afferents are undetectable in the DREZ in EIAV-LacZ animals (g-i respectively). Scale bar, 100 μm.

Figure 5. Peripheral afferent stimulation activates postsynaptic neurons in spinal cord Following noxious heat stimulation several Fos (a) and pERK (b) immunoreactive nuclei are detected in the laminae I-II of the dorsal horn (indicated by white arrows) in the spinal cord of EIAV-RARβ2 animals. By contrast fewer Fos (c) and pERK (d) nuclei are detected in the EIAV-LacZ animals compared to unlesioned control animals (e, f). The number of Fos- (g) and pERK (h) -positive nuclei is significantly higher in EIAV-RARβ2 cords compared to EIAV-LacZ cords and EIAV-RARβ2 cords with ablated dorsal roots. Results are expressed as mean number of immunoreactive nuclei per section (± SEM). *P<0.05, Students' 2-tailed unpaired ^-test). Scale bars, 50 μm.

Figure 6. RARβ2 promotes functional recovery after axonal regeneration In all graphs, closed and open squares represent injured and uninjured forelimbs of EIAV- LacZ animals respectively, while closed and open diamonds represent injured and uninjured forelimbs of EIAV-RARβ2 animals respectively. Animals are assessed in behavioral tasks before and for 4 weeks after rhizotomy. (a,b) In a tape removal task, the time taken to sense and remove the tape by the lesioned and unlesioned forelimbs are measured separately, and the injured forelimb of EIAV-RARβ2 animals (closed diamonds) demonstrated decreased latencies compared to injured forelimb of EIAV-LacZ animals (closed squares), (c) Unilateral assessment of food displacement in a staircase apparatus showed improved scores with the EIAV-RARβ2 injured forelimb (closed diamonds) compared to the EIAV-LacZ injured forelimb (closed squares). In locomotor tasks, the number of footslips made by the injured forelimb of EIAV-RARβ2 animals are lower compared to EIAV-LacZ animals (d), while no significant differences are observed between the 2 groups during the beam crossing test (e) and footprint analyses (f). Data represent mean ± SEM. *P<0.05, 2-way RM ANOVA, Tukey's post-hoc test). Figure 7. Morphology and inflammatory response in the rat spinal cord after vector injection

(a) The morphology of the adult rat spinal cord 3 weeks following injection of EIAV vectors is observed by fluorescent Nissl staining and no abnormalities are detected. The lack of innate inflammatory response to the vector is illustrated by lack of immunoreactivity against macrophages (OX42 marker; b), microglial cells (EDl marker; c) and T cells (CD8 marker; d). (e) Compared to control spinal cord, EIAV vector did not alter vascular density in the injected cord as detected by staining for glucose transporter- 1 (Glut-1). (f) Vector map illustrating the EIAV transfer genome that is used in all experiments. RARβ2 cDNA is previously isolated from a mouse brain cDNA library by PCR and cloned downstream of the human CMV promoter. The transfer vector also contains a self-inactivating (SESf) LTR, the woodchuck hepatitis virus posttranscriptional regulatory element (WPRE) and a central polypurine tract (cPPT). Control EIAV-LacZ contains the β-galactosidase gene in place of the RARβ2 gene in an identical vector. Scale bars: 200 μm (a), 100 μm (b-e).

Figure 8. Increased cAMP immunoreactivity in EIAV-RARβ2 transduced DRG neurons. cAMP immunoreactivity, indicated in green, is increased in ipsilateral DRGs of (a) EIAV- RARβ2 treated animals compared to (b) control EIAV-LacZ animals and to (c, d) contralateral untransduced DRGs of the respective groups. Scale bars, 100 μm

Figure 9. EIAV-RARβ2 treated animals with ablated dorsal roots do not show increased axonal regeneration across the DREZ or improvement in functional recovery

(a) Axonal regeneration in the DREZ of EIAV-RARβ2 with ablated dorsal roots are examined by detection of BDA-labeled axons as well as NF200, CGRP and IB₄ immunoreactivity (indicated in green). Few if any labeled fibers are observed in the DREZ of these animals, which is delineated from the peripheral region of the dorsal root by the lack of laminin staining (red). EIAV-RARβ2 treated animals with ablated roots did not show significant improvement in functional recovery compared to EIAV-LacZ rats as shown by poor performance in tape removal (b, c), paw reaching (d), ladder and beam crossing tasks (e, f). There are no significant differences in stride lengths between all groups (g). In b-g only data for the injured forelimb has been represented (mean + SEM) for EIAV-RARβ2 treated animals with ablated roots (closed triangles) and EIAV-LacZ (closed squares) and EIAV-RARβ2 (closed diamonds) groups have been included for comparison. Scale bars: 100 μm

Figure 10. Microarray and Q-PCR analyses demonstrate differential regulation of NCS-I in RARβ2-transduced spinal cord.

Figure 11. EIAV .NCS-I promotes neurite outgrowth in adult neurons compared to control EIAV.LacZ in vitro. Adult neurons from DRG (A, D), spinal cord (B, E) and cortex (C, F) are either transduced with control EIAV.LacZ (A-C) or EIAV.NCS-1 (D-F). Vector expression of either β-galactosidase (1:300, Promega; Z3781) or NCS-I (1: 200, Abeam; ab 18060) is indicated in red while neurite outgrowth is detected by GAP 43 (1:1000, Chemicon; AB5220), a regeneration marker in green. (G) The average length of the longest neurite is measured in neuronal cultures. Black bar; EIAV-LacZ transduced neurons, Blue bar; EIAV-NCS-I transduced neurons. Scale bar represents 100 μm.

DETAILED DESCRIPTION

Throughout this specification, unless the context requires otherwise, the word "comprise", or variations such as "comprises" or "comprising", will be understood to imply the inclusion of a stated element or integer or group of elements or integers but not the exclusion of any other element or integer or groups of elements or integers.

The first aspect of the present invention relates to a system capable of delivering a nucleotide sequence to a target cell.

The vector system may be a non-viral system or a viral system.

NON-VIRAL VECTOR SYSTEMS

Non-viral delivery systems include but are not limited to DNA transfection methods. Here, transfection includes a process using a non-viral vector to deliver a gene to a target mammalian cell. Typical transfection methods include electroporation, DNA biolistics, lipid-mediated transfection, compacted DNA-mediated transfection, liposomes, immunoliposomes, lipofectin, cationic agent-mediated, cationic facial amphiphiles (CFAs) (Nature Biotechnology 1996 14; 556), multivalent cations such as spermine, cationic lipids or poly lysine, 1, 2,-bis (oleoyloxy)-3-(trimethylammonio) propane (DOTAP)-cholesterol complexes (Wolff and Trubetskoy 1998 Nature Biotechnology 16: 421) and combinations thereof.

VIRAL VECTOR SYSTEMS

The vector system may be a viral vector system. Viral vector or viral delivery systems include but are not limited to adenoviral vectors, adeno-associated viral (AAV) vectors, herpes viral vectors, retroviral vectors (including lentiviral vectors) and baculoviral vectors.

Preferably the vector system is a retroviral vector system.

RETROVIRUSES

The concept of using viral vectors for gene therapy is well known (Verma and Somia (1997) Nature 389:239-242).

There are many retroviruses. For the present application, the term "retrovirus" includes: murine leukemia virus (MLV), human immunodeficiency virus (HTV), equine infectious anaemia virus (EIAV), mouse mammary tumour virus (MMTV), Rous sarcoma virus (RSV), Fujinami sarcoma virus (FuSV), Moloney murine leukemia virus (Mo-MLV), FBR murine osteosarcoma virus (FBR MSV), Moloney murine sarcoma virus (Mo-MSV), Abelson murine leukemia virus (A-MLV), Avian myelocytomatosis virus-29 (MC29), and Avian erythroblastosis virus (AEV) and all other retroviridiae including lentiviruses.

A detailed list of retroviruses may be found in Coffin et al ("Retroviruses" 1997 Cold Spring Harbour Laboratory Press Eds: JM Coffin, SM Hughes, HE Varmus pp 758-763).

In a preferred embodiment, the retroviral vector system is derivable from a lentivirus. Lentiviruses also belong to the retrovirus family, but they can infect both dividing and non- dividing cells (Lewis et al (1992) EMBO J. 3053-3058). The lentivirus group can be split into "primate" and "non-primate". Examples of primate lentiviruses include the human immunodeficiency virus (HIV), the causative agent of human acquired immunodeficiency syndrome (ADDS), and the simian immunodeficiency virus (SrV). The non-primate lentiviral group includes the prototype "slow virus" visna/maedi virus (VMV), as well as the related caprine arthritis-encephalitis virus (CAEV), equine infectious anaemia virus (EIAV) and the more recently described feline immunodeficiency virus (FIV) and bovine immunodeficiency virus (BIV). In a preferred embodiment, the retroviral vector system is derivable from EIAV.

Details on the genomic structure of some lentiviruses may be found in the art. By way of example, details on HIV and EIAV may be found from the NCBI Genbank database (i.e. Genome Accession Nos. AF033819 and AF033820 respectively). Details of HTV variants may also be found at http ://hiv-web . lanl.gov. Details of EIAV variants may be found through http://www.ncbi.nlm.nih.gov.

It is known that the separate expression of the components required to produce a retroviral vector particle on separate DNA sequences cointroduced into the same cell will yield retroviral particles carrying defective retroviral genomes that carry therapeutic genes (e.g. Reviewed by Miller 1992 Curr Top Microbiol Immunol 158:1-24). This cell is referred to as the producer cell.

There are two common procedures for generating producer cells. In one, the sequences encoding retroviral Gag, Pol and Env proteins are introduced into the cell and stably integrated into the cell genome; a stable cell line is produced which is referred to as the packaging cell line. The packaging cell line produces the proteins required for packaging retroviral RNA but it cannot bring about encapsidation due to the lack of a psi region. However, when a vector genome (having a psi region) is introduced into the packaging cell line, the helper proteins can package the ^/-positive recombinant vector RNA to produce the recombinant virus stock. This can be used to transduce a gene (i.e an NCS-I encoding gene) into recipient cells. The recombinant virus whose genome lacks all genes required to make viral proteins can infect only once and cannot propagate. Hence, the gene is introduced into the host cell genome without the generation of potentially harmful retrovirus. A summary of the available packaging lines is presented in "Retroviruses" (1997 Cold Spring Harbour Laboratory Press Eds: JM Coffin, SM Hughes, HE Varmus pp 449). The present invention also provides a packaging cell line comprising a viral vector genome which is capable of producing a vector system of the invention. For example, the packaging cell line may be transduced with a viral vector system comprising the genome or transfected with a plasmid carrying a DNA construct capable of encoding the RNA genome. The present invention also provides a kit for producing a retroviral vector system of the invention which comprises a packaging cell and a retroviral vector genome.

The second approach is to introduce the three different DNA sequences that are required to produce a retroviral vector particle i.e. the env coding sequences, the gag-pol coding sequence and the defective retroviral genome containing the NCS-I -encoding gene into the cell at the same time by transient transfection and the procedure is referred to as transient triple transfection (Landau & Liftman 1992; Pear et al 1993). The triple transfection procedure has been optimised (Soneoka et al 1995; Finer et al 1994). WO 94/29438 describes the production of producer cells in vitro using this multiple DNA transient transfection method. WO

97/27310 describes a set of DNA sequences for creating retroviral producer cells either in vivo or in vitro for re-implantation.

The components of the viral system which are required to complement the vector genome may be present on one or more "producer plasmids" for transfecting into cells.

The present invention also provides a kit for producing a retroviral vector of the invention, comprising

(i) a viral vector genome which is incapable of encoding one or more proteins which are required to produce a vector particle;

(ii) one or more producer plasmid(s) capable of encoding the protein which is not encoded by (i); and optionally

(iii) a cell suitable for conversion into a producer cell.

In a preferred embodiment, the viral vector genome is incapable of encoding the proteins gag, pol and env. Preferably the kit comprises one or more producer plasmids encoding env, gag and pol, for example, one producer plasmid encoding env and one encoding gag- pol. Preferably the gag-pol sequence is codon optimised for use in the particular producer cell (see below). The present invention also provides a producer cell expressing the vector genome and the producer plasmid(s) capable of producing a retroviral vector system of the present invention.

Preferably the retroviral vector system of the present invention is a self-inactivating (SIN) vector system (see, e.g., US Patents 6,924,123 and 7,056,699).

Preferably a recombinase assisted mechanism is used which facilitates the production of high titre regulated lentiviral vectors from the producer cells of the present invention.

As used herein, the term "recombinase assisted system" includes but is not limited to a system using the Cre recombinase / loxP recognition sites of bacteriophage Pl or the site- specific FLP recombinase of S. cerevisiae which catalyses recombination events between 34 bp FLP recognition targets (FRTs).

The site-specific FLP recombinase of S. cerevisiae which catalyses recombination events between 34 bp FLP recognition targets (FRTs) has been configured into DNA constructs in order to generate high level producer cell lines using recombinase-assisted recombination events (Karreman et al (1996) NAR 24:1616-1624). A similar system has been developed using the Cre recombinase / loxP recognition sites of bacteriophage Pl (see PCT/GBOO/03837; Vanin et al (1997) J. Virol 71:7820-7826). This was configured into a lentiviral genome such that high titre lentiviral producer cell lines were generated.

By using producer/packaging cell lines, it is possible to propagate and isolate quantities of retroviral vector particles (e.g. to prepare suitable titres of the retroviral vector particles) for subsequent transduction of, for example, a site of interest (such a DRG). Producer cell lines are usually better for large-scale production of vector particles.

Transient transfection has numerous advantages over the packaging cell method. In this regard, transient transfection avoids the longer time required to generate stable vector- producing cell lines and is used if the vector genome or retroviral packaging components are toxic to cells. If the vector genome encodes toxic genes or genes that interfere with the replication of the host cell, such as inhibitors of the cell cycle or genes that induce apoptosis, it may be difficult to generate stable vector-producing cell lines, but transient transfection can be used to produce the vector before the cells die. Also, cell lines have been developed using transient infection which produces vector titre levels that are comparable to the levels obtained from stable vector-producing cell lines (Pear et al 1993, PNAS 90:8392-8396).

Producer cells/packaging cells can be of any suitable cell type. Producer cells are generally mammalian cells but can be, for example, insect cells.

As used herein, the term "producer cell" or "vector producing cell" refers to a cell which contains all the elements necessary for production of retroviral vector particles.

Preferably the envelope protein sequences, and nucleocapsid sequences are all stably integrated in the producer and/or packaging cell. However, one or more of these sequences could also exist in episomal form and gene expression could occur from the episome.

As used herein, the term "packaging cell" refers to a cell which contains those elements necessary for production of infectious recombinant virus which are lacking in the RNA genome. Typically, such packaging cells contain one or more producer plasmids which are capable of expressing viral structural proteins (such as gag-pol and env, which may be codon optimised) but they do not contain a packaging signal.

The term "packaging signal" which is referred to interchangeably as "packaging sequence" or "psi" is used in reference to the non-coding, czs-acting sequence required for encapsidation of retroviral RNA strands during viral particle formation. In HIV-I, this sequence has been mapped to loci extending from upstream of the major splice donor site (SD) to at least the gag start codon.

Packaging cell lines may be readily prepared (see also WO 92/05266) and utilised to create producer cell lines for the production of retroviral vector particles. As already mentioned, a summary of the available packaging lines is presented in "Retroviruses" (Coffin et al., supra).

Also as discussed above, simple packaging cell lines, comprising a provirus in which the packaging signal has been deleted, have been found to lead to the rapid production of undesirable replication competent viruses through recombination. In order to improve safety, second generation cell lines have been produced wherein the 3'LTR of the provirus is deleted. In such cells, two recombinations would be necessary to produce a wild type virus. A further improvement involves the introduction of the gag-pol genes and the env gene on separate constructs so-called third generation packaging cell lines. These constructs are introduced sequentially to prevent recombination during transfection.

In these split-construct, third generation cell lines, a further reduction in recombination may be achieved by changing the codons. This technique, based on the redundancy of the genetic code, aims to reduce homology between the separate constructs, for example between the regions of overlap in the gag-pol and env open reading frames.

The packaging cell lines are useful for providing the gene products necessary to encapsidate and provide a membrane protein for a high titre vector particle production. The packaging cell may be a cell cultured in vitro such as a tissue culture cell line. Suitable cell lines include but are not limited to mammalian cells such as murine fibroblast derived cell lines or human cell lines. Preferably the packaging cell line is a human cell line, such as for example: HEK293, 293-T, TE671, HT1080.

Alternatively, the packaging cell may be a cell derived from the individual to be treated such as a monocyte, macrophage, blood cell or fibroblast. The cell may be isolated from an individual and the packaging and vector components administered ex vivo followed by re-administration of the autologous packaging cells.

It is highly desirable to use high-titre virus preparations in both experimental and practical applications. Techniques for increasing viral titre include using apsi plus packaging signal as discussed above and concentration of viral stocks.

As used herein, the term "high titre" means an effective amount of a retroviral vector or particle which is capable of transducing a target site such as a cell.

As used herein, the term "effective amount" means an amount of a regulated retroviral or lentiviral vector or vector particle which is sufficient to induce expression of NCS-I at the target site.

The presence of a sequence termed the central polypurine tract (cPPT) may improve the efficiency of gene delivery to non-dividing cells (see WO 00/31200). This cis-acting element is located, for example, in the EIAV polymerase coding region element. Preferably the genome of the vector system used in the present invention comprises a cPPT sequence.

In addition, or in the alternative, the viral genome may comprise a post-translational regulatory element and/or a translational enhancer.

MINIMAL SYSTEMS

It has been demonstrated that a primate lentivirus minimal system can be constructed which requires none of the HTV7SIV additional genes vif, vpr, vpx, vpu, tat, rev and nef for either vector production or for transduction of dividing and non-dividing cells. It has also been demonstrated that an EIAV minimal vector system can be constructed which does not require S2 for either vector production or for transduction of dividing and non-dividing cells. The deletion of additional genes is highly advantageous. Firstly, it permits vectors to be produced without the genes associated with disease in lentiviral (e.g. HTV) infections. In particular, tat is associated with disease. Secondly, the deletion of additional genes permits the vector to package more heterologous DNA. Thirdly, genes whose function is unknown, such as S2, may be omitted, thus reducing the risk of causing undesired effects. Examples of minimal lentiviral vectors are disclosed in WO-A-99/32646 and in WO-A-98/17815.

Thus, preferably, the delivery system used in the invention is devoid of at least tat and S2 (if it is an EIAV vector system), and possibly also vif, vpr, vpx, vpu and nef More preferably, the systems of the present invention are also devoid of rev. Rev was previously thought to be essential in some retroviral genomes for efficient virus production. For example, in the case of HIV, it was thought that rev and RRE sequence should be included. However, it has been found that the requirement for rev and RRE can be reduced or eliminated by codon optimisation (see below) or by replacement with other functional equivalent systems such as the MPMV system. As expression of the codon optimised gag- ol is REV independent, RRE can be removed from the gag-pol expression cassette, thus removing any potential for recombination with any RRE contained on the vector genome.

In a preferred embodiment a viral vector system according to the first aspect of the invention lacks the Rev response element (RRE). In a preferred embodiment, the system used in the present invention is based on a so-called "minimal" system in which some or all of the additional genes have be removed.

CODON OPTIMISATION

Codon optimisation has previously been described in WO99/41397. Different cells differ in their usage of particular codons. This codon bias corresponds to a bias in the relative abundance of particular tRNAs in the cell type. By altering the codons in the sequence so that they are tailored to match with the relative abundance of corresponding tRNAs, it is possible to increase expression. By the same token, it is possible to decrease expression by deliberately choosing codons for which the corresponding tRNAs are known to be rare in the particular cell type. Thus, an additional degree of translational control is available.

Many viruses, including HIV and other lentiviruses, use a large number of rare codons and by changing these to correspond to commonly used mammalian codons, increased expression of the packaging components in mammalian producer cells can be achieved. Codon usage tables are known in the art for mammalian cells, as well as for a variety of other organisms.

Codon optimisation has a number of other advantages. By virtue of alterations in their sequences, the nucleotide sequences encoding the packaging components of the viral particles required for assembly of viral particles in the producer cells/packaging cells have RNA instability sequences (INS) eliminated from them. At the same time, the amino acid sequence coding sequence for the packaging components is retained so that the viral components encoded by the sequences remain the same, or at least sufficiently similar that the function of the packaging components is not compromised. Codon optimisation also overcomes the Rev/RRE requirement for export, rendering optimised sequences Rev independent. Codon optimisation also reduces homologous recombination between different constructs within the vector system (for example between the regions of overlap in the gag-pol and env open reading frames). The overall effect of codon optimisation is therefore a notable increase in viral titre and improved safety.

In one embodiment only codons relating to INS are codon optimised. However, in a much more preferred and practical embodiment, the sequences are codon optimised in their entirety, with the exception of the sequence encompassing the frameshift site. The gag-pol gene comprises two overlapping reading frames encoding the gag-pol proteins. The expression of both proteins depends on a frameshift during translation. This frameshift occurs as a result of ribosome "slippage" during translation. This slippage is thought to be caused at least in part by ribosome-stalling RNA secondary structures. Such secondary structures exist downstream of the frameshift site in the gag-pol gene. For HTV, the region of overlap extends from nucleotide 1222 downstream of the beginning of gag (wherein nucleotide 1 is the A of the gag ATG) to the end of gag (nt 1503). Consequently, a 281 bp fragment spanning the frameshift site and the overlapping region of the two reading frames is preferably not codon optimised. Retaining this fragment will enable more efficient expression of the gag-pol proteins.

For EIAV the beginning of the overlap has been taken to be nt 1262 (where nucleotide 1 is the A of the gag ATG). The end of the overlap is at 1461 bp. In order to ensure that the frameshift site and the gag-pol overlap are preserved, the wild type sequence has been retained from nt 1156 to 1465.

Derivations from optimal codon usage may be made, for example, in order to accommodate convenient restriction sites, and conservative amino acid changes may be introduced into the gag-pol proteins.

Due to the degenerate nature of the Genetic Code, it will be appreciated that numerous gag- ol sequences can be achieved by a skilled worker. Also there are many retroviral variants described which can be used as a starting point for generating a codon optimised gag-pol sequence. Lentiviral genomes can be quite variable. For example there are many quasi- species of HIV-I which are still functional. This is also the case for EIAV. These variants may be used to enhance particular parts of the transduction process. Examples of HTV-I variants may be found at http ://hiv-web. lanl. gov. Details of EIAV clones may be found at the NCBI database: http://www.ncbi.nlm.nih.gov.

The strategy for codon optimised gag-pol sequences can be used in relation to any retrovirus. This would apply to all lentiviruses, including EIAV, FIV, BIV, CAEV, VMR, SIV, HTV-I and HTV-2. In addition this method could be used to increase expression of genes from HTLV-I, HTLV-2, HFV, HSRV and human endogenous retroviruses (HERV), MLV and other retroviruses. Codon optimisation can render gag-pol expression Rev independent. In order to enable the use of anti-rev or RRE factors in the retroviral vector, however, it would be necessary to render the viral vector generation system totally Rev/RRE independent. Thus, the genome also needs to be modified. This is achieved by optimising vector genome components. Advantageously, these modifications also lead to the production of a safer system absent of all additional proteins both in the producer and in the transduced cell.

As described above, the packaging components for a retroviral vector include expression products of gag, pol and env genes. In addition, efficient packaging depends on a short sequence of 4 stem loops followed by a partial sequence from gag and env (the "packaging signal"). Thus, inclusion of a deleted gag sequence in the retroviral vector genome (in addition to the full gag sequence on the packaging construct) will optimise vector titre. To date efficient packaging has been reported to require from 255 to 360 nucleotides of gag in vectors that still retain env sequences, or about 40 nucleotides of gag in a particular combination of splice donor mutation, gag and env deletions. It has surprisingly been found that a deletion of all but the N-termnial 360 or so nucleotides in gag leads to an increase in vector titre. Thus, preferably, the retroviral vector genome includes a gag sequence which comprises one or more deletions, more preferably the gag sequence comprises about 360 nucleotides derivable from the N-terminus.

It will be appreciated that codon optimisation can, of course, be applied to all or some parts of the system of the present invention.

PSEUDOTYPING

In the design of retroviral vector systems it is desirable to engineer particles with different target cell specificities to the native virus, to enable the delivery of genetic material to an expanded or altered range of cell types. One manner in which to achieve this is by engineering the virus envelope protein to alter its specificity. Another approach is to introduce a heterologous envelope protein into the vector particle to replace or add to the native envelope protein of the virus.

The term pseudotyping means incorporating in at least a part of, or substituting a part of, or replacing all of, an env gene of a viral genome with a heterologous env gene, for example an env gene from another virus. Pseudotyping is not a new phenomenon and examples may be found in WO 99/61639, WO-A-98/05759, WO-A-98/05754, WO-A-97/17457, WO-A- 96/09400, WO-A-91/00047 and Mebatsion et al 1997 Cell 90, 841-847.

Pseudotyping can improve retroviral vector stability and transduction efficiency. A pseudotype of murine leukemia virus packaged with lymphocytic choriomeningitis virus (LCMV) has been described (Miletic et al (1999) J. Virol. 73:6114-6116) and shown to be stable during ultracentrifugation and capable of infecting several cell lines from different species.

A viral vector system in accordance with the first aspect of the invention may be pseudotyped with any heterologous env protein.

It can be difficult to access certain neurons in the CNS. For this reason it may be advantageous to use a vector system capable of retrograde transport, so that the vector may be administered to a site distant to the target site. For example, the vector system may comprise a protein (or a mutant, variant, homologue or fragment thereof) from a virus which is capable of travelling by retrograde transport. For example, the system may comprise a protein from a rabies virus, herpes virus, adenovirus or from Ebola virus. If the vector system is a viral vector system, it may be pseudotyped with the envelope protein from such a virus. In a preferred embodiment, the vector system is pseudotyped with at least a part of a rabies G protein or a mutant, variant, homologue or fragment thereof.

It has also been shown that vector systems comprising Rabies G protein or a part thereof are capable of transducing (tyrosine hydroxylase) TH positive neurons - a subset which are difficult to transduce with conventional vectors (see WO02/36170).

NCS-I

Ca²⁺ sensing proteins (including the neuronal calcium sensor (NCS) proteins), change their conformation on Ca²⁺ binding, which allows them to modulate the interaction with their targets, and thus act as effector molecules to transduce Ca²⁺ signals into appropriate downstream events. Neuronal calcium sensor- 1 (NCS-I), the mammalian ortholog of Drosphila frequenin, is a recently cloned and characterised member of the neuronal calcium sensor protein family (Burgoyne and Weiss, 2001 Biochem J 353:1-12). This 22 kDa protein is highly conserved across species, with 100% amino acid sequence homology between rat, mouse, human, chicken and Xenopus species and only 25% divergence with C. elegans, or 28% with yeast (De Castro et al 1995 Biochem Biophys Res Commun 216: 133-140), suggesting a fundamental and highly conserved function (Burgoyne and Weiss, 2001 as above).

In situ hybridization and immunohistochemistry studies have demonstrated NCS-I is neuron specific, localizing in the cell bodies, dendrites and axons throughout the brain, spinal cord, dorsal root ganglia and peripheral nerves (Martone et ah, 1999 Cell Tissue Res 295:395-407; Olafsson et ah, 1997 Proc Natl Acad Sci U S A 92:8001-8005.; Averill et al., 2004 Neuroscience 123:419-27).

NCS-I has been shown to be present spatiotemporally in the embryonic rat spinal cord and olfactory system during development and to co-localise with GAP 43, a growth and regeneration marker (Kawasaki et ah, 2003. J Comp Neurol 460:465-475. ;Treloar et al., 2005. J Comp Neurol 482:201-216). NCS-I has also been shown to co-localise spatiotemporally with synaptophysin, a synapse marker, in embryonic spinal cord, olfactory system and retina (Kawasaki et ah, 2003 as above;Pongs et a 1993. J Comp Neurol 482:201-216,;Reynolds et ah, 2001 Neuroreport 12:725-728;Sage et ah, 2000 Hear Res 150:70-82). Overexpression of NCS-I in cell lines has been shown to enhance synaptic formation and transmission (Chen et ah, 2001 J Physiol 532:649-659). Fly mutants overexpressing frequenin showed changes in the morphology of the motor synapses (Angaut-Petit et ah, 1993 Neurosci Lett 153:227-231). Xenopus frequenin overexpressed in spinal neurons resulted in a general enhancement of synaptic efficacy of the neuromuscular junction (Olafsson et ah, 1995 Proc Natl Acad Sci U S A 92:8001-8005). Overexpression of NCS-I created acute short-term plastic changes in the synapses of hippocampal cells, unmasking somnolent synapses (Sippy et ah, 2003 Nat Neurosci 6:1031-1038). Overexpression of NCS-I has been suggested to enhance associative learning and memory in Caenorhabditis elegans (Gomez et al (2001) Neuron 30; 241-248; EP 1250931).

Thus, NCS-I has been implicated in enhancing synaptic transmission and there has been interest in its possible mechanism of action (Hilfiker (2003) Biochem Soc Trans 31:828-

832). Mammalian NCS-I is thought to interact with a phophatidylinositol 4-hydroxykinase (type III PI4Kβ) and stimulate its activity (Hilfiker as above). PI4Kβ is thus thought to be an in vivo downstream target of NCS-I, and NCS-I may act by modulating the levels of phosphoinositides .

The primary structure of all NCS protein family members is similar (Burgoyne and Weiss (2001) as above). NCS-I possesses four EF-hand motifs (EF1-EF4), each consisting of a 12-amino acid loop within which Ca2+ is coordinated with α-helices on either side.

MUTANTS AND FRAGMENTS

The term "neuronal calcium sensor-1" or "NCS-I" in the context of the present invention includes mutants and fragments of the wild-type protein as long as they retain the capacity of wild-type protein to stimulate neurite outgrowth.

The term "retinoic acid β2" or "RARβ2" in the context of the present invention includes mutants and fragments of the wild-type protein as long as they retain the capacity of wild- type protein to stimulate neurite outgrowth.

The term "wild type" is used to mean a polypeptide having a primary amino acid sequence which is identical with the native protein (i.e. NCS-I or RAR β2 from the subject species).

The term "mutant" is used to mean a polypeptide having a primary amino acid sequence which differs from the wild type sequence by one or more amino acid additions, substitutions or deletions. A mutant may arise naturally, or may be created artificially (for example by site-directed mutagenesis). Preferably the mutant has at least 90%, 95% or 98% sequence identity (or homology) with the wild type sequence. Preferably the mutant has 20 mutations or less over the whole wild-type sequence. More preferably the mutant has 10 mutations or less, most preferably 5 mutations or less over the whole wild-type sequence.

Homology comparisons can be conducted by eye, or more usually, with the aid of readily available sequence comparison programs. These commercially available computer programs can calculate % homology between two or more sequences.

% homology may be calculated over contiguous sequences, i.e. one sequence is aligned with the other sequence and each amino acid in one sequence is directly compared with the corresponding amino acid in the other sequence, one residue at a time. This is called an "ungapped" alignment. Typically, such ungapped alignments are performed only over a relatively short number of residues.

Although this is a very simple and consistent method, it fails to take into consideration that, for example, in an otherwise identical pair of sequences, one insertion or deletion will cause the following amino acid residues to be put out of alignment, thus potentially resulting in a large reduction in % homology when a global alignment is performed. Consequently, most sequence comparison methods are designed to produce optimal alignments that take into consideration possible insertions and deletions without penalising unduly the overall homology score. This is achieved by inserting "gaps" in the sequence alignment to try to maximise local homology.

However, these more complex methods assign "gap penalties" to each gap that occurs in the alignment so that, for the same number of identical amino acids, a sequence alignment with as few gaps as possible - reflecting higher relatedness between the two compared sequences - will achieve a higher score than one with many gaps. "Affine gap costs" are typically used that charge a relatively high cost for the existence of a gap and a smaller penalty for each subsequent residue in the gap. This is the most commonly used gap scoring system. High gap penalties will of course produce optimised alignments with fewer gaps. Most alignment programs allow the gap penalties to be modified. However, it is preferred to use the default values when using such software for sequence comparisons. For example when using the GCG Wisconsin Bestfit package the default gap penalty for amino acid sequences is -12 for a gap and -4 for each extension.

Calculation of maximum % homology therefore firstly requires the production of an optimal alignment, taking into consideration gap penalties. A suitable computer program for carrying out such an alignment is the GCG Wisconsin Bestfit package (University of Wisconsin, U.S.A.; Devereux et al, 1984, Nucleic Acids Research 12:387). Examples of other software than can perform sequence comparisons include, but are not limited to, the BLAST package (see Ausubel et al, 1999 ibid - Chapter 18), FASTA (Atschul et al, 1990, J. MoI. Biol., 403-410) and the GENEWORKS suite of comparison tools. Both BLAST and FASTA are available for offline and online searching (see Ausubel et al, 1999 ibid, pages 7-58 to 7-60). However, for some applications, it is preferred to use the GCG Bestfit program. A new tool, called BLAST 2 Sequences is also available for comparing protein and nucleotide sequence (see FEMS Microbiol Lett 1999 174(2): 247-50; FEMS Microbiol Lett 1999 177(1): 187-8).

Although the final % homology can be measured in terms of identity, the alignment process itself is typically not based on an all-or-nothing pair comparison. Instead, a scaled similarity score matrix is generally used that assigns scores to each pairwise comparison based on chemical similarity or evolutionary distance. An example of such a matrix commonly used is the BLOSUM62 matrix - the default matrix for the BLAST suite of programs. GCG Wisconsin programs generally use either the public default values or a custom symbol comparison table if supplied (see user manual for further details). For some applications, it is preferred to use the public default values for the GCG package, or in the case of other software, the default matrix, such as BLOSUM62.

Once the software has produced an optimal alignment, it is possible to calculate % sequence identity. The software typically does this as part of the sequence comparison and generates a numerical result.

The sequences may have deletions, insertions or substitutions of amino acid residues which produce a silent change and result in a functionally equivalent substance. Deliberate amino acid substitutions may be made on the basis of similarity in polarity, charge, solubility, hydrophobicity, hydrophilicity, and/or the amphipathic nature of the residues as long as the secondary binding activity of the substance is retained. For example, negatively charged amino acids include aspartic acid and glutamic acid; positively charged amino acids include lysine and arginine; and amino acids with uncharged polar head groups having similar hydrophilicity values include leucine, isoleucine, valine, glycine, alanine, asparagine, glutamine, serine, threonine, phenylalanine, and tyrosine.

Conservative substitutions may be made, for example according to the Table below. Amino acids in the same block in the second column and preferably in the same line in the third column may be substituted for each other:

The present invention also encompasses homologous substitution (substitution and replacement are both used herein to mean the interchange of an existing amino acid residue, with an alternative residue) may occur i.e. like-for-like substitution such as basic for basic, acidic for acidic, polar for polar etc. Non-homologous substitution may also occur i.e. from one class of residue to another or alternatively involving the inclusion of unnatural amino acids such as ornithine (hereinafter referred to as Z), diaminobutyric acid ornithine (hereinafter referred to as B), norleucine ornithine (hereinafter referred to as O), pyriylalanine, thienylalanine, naphthylalanine and phenylglycine.

Replacements may also be made by unnatural amino acids include; alpha* and alpha- disubstituted* amino acids, N-alkyl amino acids*, lactic acid*, halide derivatives of natural amino acids such as trifluorotyrosine*, p-Cl-phenylalanine*, p-Br-phenylalanine*, p-I- phenylalanine*, L-allyl-glycine*, β-alanine*, L-α-amino butyric acid*, L-γ-amino butyric acid*, L-α-amino isobutyric acid*, L-ε-amino caproic acid^#, 7-amino heptanoic acid*, L- methionine sulfone^#*, L-norleucine*, L-norvaline*, p-nitro-L-phenylalanine*, L- hydroxyproline", L-thioproline*, methyl derivatives of phenylalanine (Phe) such as 4- methyl-Phe*, pentamethyl-Phe*, L-Phe (4-amino/, L-Tyr (methyl)*, L-Phe (4-isopropyl)*, L-Tic (l,2,3,4-tetrahydroisoquinoline-3-carboxyl acid)*, L-diaminopropionic acid^# and L- Phe (4-benzyl)*. The notation * has been utilised for the purpose of the discussion above (relating to homologous or non-homologous substitution), to indicate the hydrophobic nature of the derivative whereas # has been utilised to indicate the hydrophilic nature of the derivative, #* indicates amphipathic characteristics.

Variant amino acid sequences may include suitable spacer groups that may be inserted between any two amino acid residues of the sequence including alkyl groups such as methyl, ethyl or propyl groups in addition to amino acid spacers such as glycine or β- alanine residues. A further form of variation involves the presence of one or more amino acid residues in peptoid form, which will be well understood by those skilled in the art. For the avoidance of doubt, "the peptoid form" is used to refer to variant amino acid residues wherein the α-carbon substituent group is on the residue's nitrogen atom rather than the α-carbon. Processes for preparing peptides in the peptoid form are known in the art, for example Simon RJ et al, PNAS (1992) 89(20), 9367-9371 and Horwell DC, Trends Biotechnol. (1995) 13(4), 132-134.

The term "fragment" indicates that the polypeptide comprises a fraction of the wild-type amino acid sequence. It may comprise one or more large contiguous sections of sequence or a plurality of small sections. The polypeptide may also comprise other elements of sequence, for example, it may be a fusion protein with another protein. Preferably the polypeptide comprises at least 50%, more preferably at least 65%, most preferably at least 80% of the wild-type sequence.

Human NCS-I has 190 amino acids. A system of the present invention encoding a fragment of NCS-I preferably encodes at least 150, 160, 170, 180 or 185 amino acids of the wild-type sequence. NCS-I is an N-terminally myristoylated protein that contains four EF-hand motifs, three of which are capable of binding Ca²⁺ in the submicromolar range. The fragment may contain at least three of the EF hands, the fragment may also contain the N-terminal myristoylation consensus sequence.

Human RARβ2 has two isoforms: isoform 1, having 448 amino acids; and isoform 2, which lacks an exon in its 5' region so that translation begins at a downstream, in-frame start codon. Thus, the encoded protein is shorter at the N terminus than isoform 1, having 336 amino acids. A system of the present invention encoding a fragment of RARβ2 preferably encodes at least 250, 175, 300, 325 or 330 amino acids of the wild-type sequence.

With respect to function, the mutant or fragment of NCS-I or RARβ2 should be capable promoting neurite outgrowth when delivered to or expressed in a neuron.

The system of the present invention may comprise a nucleotide sequence encoding NCS-I and/or RARβ2 (which term includes a mutant or fragment of the wild-type sequence). The nucleotide sequence may be any suitable nucleotide sequence, which need not necessarily be a complete naturally occurring DNA or RNA sequence. Thus, the NOI can be, for example, a synthetic RNA/DNA sequence, a recombinant RNA/DNA sequence (i.e. prepared by use of recombinant DNA techniques), a cDNA sequence or a partial genomic DNA sequence, including combinations thereof. Preferably, the sequence is, comprises, or is transcribed from cDNA. If the system is capable of causing the target cell to encode both NCS-I and RARβ2, it may comprise a separate construct (e.g. plasmid or vector) for each sequence or it may comprise a single construct capable of expressing both sequences.

In multi-cistronic cassettes, the two or more genes may be operably linked by one or more internal ribosome entry sequences (IRES(s)). A review on IRES is presented by Mountford and Smith (TIG May 1995 vol 11, No 5:179-184).

Although IRESs are an efficient way to co-express multiple genes from one vector, other methods are also useful, and may be used alone or in conjunction with IRESs. These include the use of multiple internal promoters in the vector (Overell et al., MoI Cell Biol. 8: 1803-8 (1988)), or the use of alternate splicing patterns leading to multiple RNA species derived from the single viral genome that expresses the different genes. This strategy has previously been used by itself for two genes (Cepko et al. Cell 37: 1053 (1984)).

The sequence(s) may be or correspond to the wild-type nucleotide sequence(s) (i.e. the endogenous sequence encoding NCS-I and/or RARβ2 in the subject species or the corresponding mRNA or cDNA). However, due to the degeneracy of the genetic code, it will be appreciated that a large number of alternative sequences will also give rise to the same amino acid sequence and these may also be used in the system of the present invention.

INDUCING NCS- 1 EXPRESSION OR ACTIVATION OF NCS-I

The present inventors have shown that overexpression of NCS-I promotes neurite outgrowth. In the Examples section, overexpression is achieved by transducing a target cell with a NCS-I -encoding gene. Overexpression of NCS-I may alternatively be achieved by delivering a factor (or a nucleotide sequence encoding a factor) which enhances the transcription or translation of NCS- 1. The present invention also provides the use of such a system to promote neurite outgrowth. The system may deliver a factor which for example is or encodes a transcription factor. Another option is to activate NCS-I rather than cause its over expression. "Activating" factors may be necessary for NCS-I to become active, or may accentuate its activity. There are many "activities" postulated for NCS-I, including calcium binding, increase of long-term potentiation of the hippocampus, facilitation of transmitter release. In the context of the present invention, all of these activities may be increased, as long as the factor causes a (possibly concomitant) increase in promotion of neurite outgrowth.

The present invention also provides the use of a system capable of delivering an NCS-I activating factor to a target cell to promote neurite outgrowth. The "activating" factor may, for example, be an anti-NCS-1 antibody, a ligand binding molecule, a calcium mimetic or a derivative of a calmodulin activator.

NEURITE OUTGROWTH

The spinal cord includes nerve cells, or neurons, and long nerve fibres called axons. Axons in the spinal cord carry signals downward from the brain (along descending pathways) and upward toward the brain (along ascending pathways). Dendrites are branched extensions of neurons that may receive signals from other nerve cells.

Neurite outgrowth is the process by which the neuron grows out axons and dendrites, in order to form functional networks with surrounding cells and other neurons. NCS-I promotes both neurite outgrowth and process production by neurons.

The target cell for delivery of an NCS-I expressing gene may be a neuron, in particular an adult neuron. An adult neuron is a substantially terminally differentiated neuron. The system may, for example deliver the NCS-I encoding gene to the dorsal root ganglion, spinal cord or to a cortical neuron. The system may activate regeneration processes in neurons, for example, injured neurons. Injured (or damaged) neurons include diseased neurons.

There are two broad ways in which stimulation of neuronal growth may improve neural function after disease or injury to the nervous system: i) to promote the regeneration of axons that have been damaged so that they reestablish functional connections and thereby restore some of the lost function; and ii) to promote the sprouting (growth) of intact neurons that have survived or been spared by the injury or disease and the establishment of functional connections by these sprouts to restore some of the lost function (promoting synaptic plasticity).

MEDICAL CONDITIONS

The system of the present invention may be used to treat and/or prevent a condition which results in, or is likely to cause, neuron damage. It may be used to treat and/or prevent a condition which is likely to be prevented or ameliorated by neuron generation or regeneration.

For example, the system may be used to treat one or more of the following conditions: spinal cord injury, avulsion injury, brachial plexus injury, traumatic brain injury, stroke, and a neurodegenerative disease.

A spinal cord injury usually begins with a sudden, traumatic blow to the spine that fractures or dislocates vertebrae. The damage begins at the moment of injury when displaced bone fragments, disc material, or ligaments bruise or tear into spinal cord tissue. Most injuries to the spinal cord don't completely sever it. Instead, an injury is more likely to cause fractures and compression of the vertebrae, which then crush and destroy the axons, extensions of nerve cells that carry signals up and down the spinal cord between the brain and the rest of the body. An injury to the spinal cord can damage a few, many, or almost all of these axons. Major damage can result in complete paralysis.

Avulsion injury is a nerve injury in which traction produces a ripping of the nerve roots. Major causes include automotive (in particular motorbike) accidents, lacerations, gunshot wounds and nerve cancer.

The brachial plexus is a network of spinal nerves that originates in the back of the neck, extends through the axilla (armpit) and gives rise to nerves to the upper limb. Injuries to the brachial plexus affect the nerves supplying the shoulder, upper arm, forearm and hand, causing numbness, tingling, pain, weakness, limited movement or even paralysis of the upper limb. Although injuries can occur at any time, many brachial plexus injuries happen during birth. The baby's shoulders may become impacted during the birth process, causing the brachial plexus nerves to stretch or tear. Brain injury can occur in many ways. Traumatic brain injuries typically result from accidents in which the head strikes an object. However, other brain injuries, such as those caused by insufficient oxygen, poisoning, or infection, can cause similar results. Traumatic brain injury (TBI) can significantly affect many cognitive, physical, and psychological skills. Physical problems can include ambulation, balance, coordination, fine motor skills, strength, and endurance.

A stroke (also referred to as a cerebrovascular accident or CVA) is the sudden death of brain cells caused by a problem with the blood supply. When blood flow to the brain is impaired, oxygen and important nutrients cannot be delivered. The result is abnormal brain function. Blood flow to the brain can be disrupted by either a blockage or rupture of an artery to the brain. There are many causes for a stroke, as shown below:

(i) Causes due to blockage of artery:

• Clogging of arteries within the brain (e.g. lacunar stroke)

• Hardening of the arteries leading to the brain (e.g. carotid artery occlusion)

• Embolism to the brain from the heart or an artery (ii) Causes due to rupture of an artery (i.e. hemorrhage) • Cerebral hemorrhage (bleeding within the brain substance)

• Subarachnoid hemorrhage (bleeding between the brain and the inside of the skull).

Neurodegenerative diseases are hereditary and sporadic conditions characterized by progressive nervous system dysfunction. These disorders are often associated with atrophy of the affected central or peripheral nervous system structures.

Neurodegenerative diseases include: Lewy Body Disease, Motor Neuron Disease, Multiple System Atrophy, Parkinsons Disease, Postpoliomyelitis Syndrome, Prion Diseases, Shy- Drager Syndrome, Cockayne Syndrome, Huntington Disease, Lafora Disease, Neurofibromatoses, Tourette Syndrome, Tuberous Sclerosis Amyotrophic Lateral Sclerosis, Creutzfeldt- Jakob Syndrome, Kuru and Scrapie Alzheimer Disease.

PHARMACEUTICAL COMPOSITION The present invention also provides the use of a system in the manufacture of a pharmaceutical composition. The pharmaceutical composition may be used to deliver NCS-I and/or a nucleotide sequence expressing NCS-I to a target cell in a subject.

The system may be a delivery system, such as a non-viral delivery system or a viral delivery system.

The pharmaceutical composition may be used for treating an individual by gene therapy, wherein the composition comprises or is capable of producing a therapeutically effective amount of a vector system according to the present invention.

The method and pharmaceutical composition of the invention may be used to treat a human or animal subject. Preferably the subject is a mammalian subject. More preferably the subject is a human. Typically, a physician will determine the actual dosage which will be most suitable for an individual subject and it will vary with the age, weight and response of the particular subject.

The composition may optionally comprise a pharmaceutically acceptable carrier, diluent, excipient or adjuvant. The choice of pharmaceutical carrier, excipient or diluent can be selected with regard to the intended route of administration and standard pharmaceutical practice. The pharmaceutical compositions may comprise as (or in addition to) the carrier, excipient or diluent, any suitable binder(s), lubricant(s), suspending agent(s), coating agent(s), solubilising agent(s), and other carrier agents that may aid or increase the viral entry into the target site (such as for example a lipid delivery system).

The vector system used in the present invention may be administered by direct injection into the subject. For example, it may be injected directly into the CNS.

Alternatively, the vector system may be administered to a site which is distant to the CNS and may then travel to the CNS by retrograde transport.

For administration of NCS-I protein, there are many available systems known in the art.

For example, US 2005/0137134 describes a method for infusion to the putamen (within the brain) using an implantable pump, and also makes reference to many other known drug delivery apparatus, catheters and combinations thereof which have been developed for dispensing medical substances to specific sites in the body. Laske et al (1997 - J. Neurosurg. 87:586-594) also describes a method for interstitial infusion to the brain.

DELIVERY TO CELLS IN VITRO

The present invention also provides a method for delivering NCS-I, or a nucleotide sequence capable of encoding NCS-I, to a cell in vitro using a system of the invention.

The cell may, for example, be a supporting cell such as an olfactory ensheathing cell or a Schwann cell. The cell may be implanted into a subject, for example to promote neuron regeneration. The cell may be implanted into the CNS of the subject. For example, the cell may be introduced into the spinal cord of the subject.

The cell to which NCS-I or its encoding sequence is delivered may be derivable from the subject (making implantation an ex vivo procedure), or it may be from an alternative source (such as from a suitable donor or a cell line etc).

The invention will now be further described by way of Examples, which are meant to serve to assist one of ordinary skill in the art in carrying out the invention and are not intended in any way to limit the scope of the invention.

EXAMPLES

Example 1 - Exogenous delivery of RARβ2 into adult rat DRG neurons promotes axonal regeneration

RARβ2 has been shown to activate neurite outgrowth in embryonic DRG neurons and adult mouse spinal cord explants (Concoran et al (2000) J. Cell Sci 113:2567-2574; (2002) J. Cell Sci. 115:3779-3786).

a)RARβ2 enhances neurite outgrowth in DRG neurons via activation of cAMP In order to transduce DRG neurons with RARβ2 at high efficiency, a minimal equine infectious anemia virus (EIAV)-based lentiviral vector (Mazarakis et al (2001) Hum MoI Genet 10:2109-21; Wong et al (2004) MoI Ther 9:101-11) expressing RARβ2 (EIAV- RARβ2) is constructed. As a control a vector expressing β-galactosidase (β-gal) (EIAV- LacZ) is also constructed. It is first confirmed that 80.6 ± 2.8% of the neurons are transduced at 3 DIV (days in vitro) following addition of the vectors to the adult rat DRG neurons (data not shown). In control EIAV-LacZ transduced cultures, mature DRG neurons when grown sparsely (500 neurons/well) on a reduced laminin substrate (0.1 μg/ml) extend either very short or no neurites at all (Fig. Ia). In contrast, neurons in EIAV-RARβ2 cultures project several long neurites on an identical substrate (Fig. Ib). These fiber outgrowths are quantified and to ensure that only transduced neurons are assessed, neurites are measured only from those neurons that are either β-gal and βlll tubulin positive (for EIAV-LacZ; Fig. Ia) or RARβ2 and βlll tubulin positive (for EIAV- RARβ2; Fig. Ib). At 3 DFV, expression of RARβ2 improves fiber growth by 9-fold compared to EIAV-LacZ transduced neurons (Fig. Ic; P<0.001, Students' 2-tailed unpaired ?-test) and increases the percentage of process-bearing neurons to 42.7 ± 6.4% from the 12.1 + 5.8% observed in EIAV-LacZ cultures (Fig. Id; P<0.05, Students' 2-tailed unpaired f-test). These results demonstrate that RARβ2 promotes outgrowth in adult DRG neurons by increasing neurite extension as well as stimulating neuritogenesis in a larger proportion of neurons. The next question is whether the increase in neurite outgrowth in RARβ2-transduced neurons is due to activation of cAMP signaling pathways. Intracellular cAMP concentrations in EIAV- RARβ2 neurons are 38-fold higher compared to EIAV-LacZ neurons (Fig. Ig; P<0.01, Students' 2-tailed unpaired £-test). In the presence of an adenylate cyclase inhibitor 2', 5'- dideoxyadenosine (DDA) EIAV-RARβ2 neurons show decreased neurite lengths (Figs. Ie & h). This blocking of RARβ2-enhanced regeneration is dose-dependent and at higher DDA concentrations (50 and 100 μM) neurite outgrowth is completely abrogated. Exposure of RARβ2-transduced neurons to a cell permeant inhibitor of cAMP-dependent protein kinase (PKA), Rp-cAMP, also decreases fiber growth by 50% (Figs. If & h).

b)RARβ2 induces axonal regeneration on the nonpermissive spinal cord Transduction of DRG neurons in vitro thus promotes neurite outgrowth on a minimal substrate. However the CNS environment contains inhibitory factors such as myelin and its derivatives that can restrict regeneration. To investigate the ability of RARβ2 to overcome this inhibition and promote outgrowth in a nonpermissive environment, the following example investigates neurite outgrowth in a co-culture system. After adult rat spinal cord and DRG neurons are transduced with either EIAV-LacZ or EIAV-RARβ2 in vivo, the transduced DRG neurons are grown on spinal cord sections which provided an inhibitory substrate for neurite outgrowth. EIAV-LacZ transduced DRG neurons grown either on naϊve spinal cord or on EIAV-RARβ2 transduced spinal cord do not extend many neurites on the spinal cord (Fig. 2a). In the occasional neuron that settles on the peripheral section of the dorsal root, neurite outgrowth is observed but these projections extended up to the DREZ and do not extend into the spinal cord (Fig. 2b). By contrast, EIAV-RARβ2 neurons grown either on naϊve or on EIAV-LacZ cord often extend projections on the spinal cord (Fig. 2c). Furthermore neurons that are grown on the dorsal root project neurites up to and beyond the DREZ into the spinal cord (Fig. 2d). Interestingly in the same cultures, neurons that are not transduced with EIAV-RARβ2 do not extend neurites beyond the DREZ or project fibers on the spinal cord (Fig. 2d). Assessment of neurite outgrowth on the spinal cord indicates that 26.3 + 2.9% of EIAV-RARβ2 neurons project neurites, compared to 3.0 ± 1.2% in EIAV-LacZ neurons (Fig. 2e; P<0.0l, Students' unpaired 2-tailed t-test). The average length of the longest neurite in EIAV-RARβ2 neurons grown on spinal cord is also significantly higher compared to that of EIAV-LacZ neurons (Fig. 2f; P<0.01, Students' unpaired 2-tailed /-test). These data suggested that expression of RARβ2 in the DRG neurons alone is sufficient to overcome the inhibitory environment of the adult spinal cord in order to promote axonal extension.

c) In vivo transduction of DRG neurons by EIAV-RARβ2

Although the above evidence indicates that RARβ2, through the induction of cAMP, stimulated neurite outgrowth of DRG neurons, it does not address regeneration in vivo. The following example is to determine if RARβ2 expression is sufficient to promote regeneration of injured sensory axons from the dorsal root into spinal cord to establish functional connectivity. Adult DRG neurons express low levels of RARβ2 mRNA and protein and in order to deliver RARβ2 to the DRG neurons, lentiviral vectors are used. High gene transfer efficiency to DRG neurons in vivo is obtained at three weeks after unilateral injection of 3 - 6 x 10⁶ transducing units of rabies-G pseudotyped EIAV-LacZ into the spinal cord. Gene transfer is observed in the cuneate fasciculus and the dorsal horn of the spinal cord, which receive afferent projections from DRG neurons (Fig. 3a). The vector is retrogradely transported to ipsilateral DRG cell bodies where extensive β-gal expression in about 35% of neurons is detected (Fig. 3b). In EIAV-RARβ2 transduced neurons RARβ2 transcripts (Fig. 3 c) and protein (Fig. 3e) are more abundant compared to control non-transduced DRGs (Figs. 3d, f), and protein expression levels are comparable to that of embryonic E 14 DRGs (Fig. 3g). In the adult rat, vector transduction is achieved in the 3 DRG subclasses - the large-diameter myelinated neurons, which express heavy neurofilament protein NF200 and subserve mechanoreceptive functions, as well as the nociceptive small-diameter peptidergic and non-peptidergic neurons that are identified by the neuropeptide calcitonin gene-related peptide (CGRP) and plant isolectin B₄ (IB₄) respectively(Averill et al (1995) Eur J NeuroSci 7:1484-94; Bennet et al (1998) J Neurosci 18:3059-72) - as detected by colocalization with the respective markers (Fig. 3h). The transduction efficiency of DRG neurons are as follows: 46 ± 2% co-expressed NF200, 33 ± 5% co-expressed CGRP and 22 ± 7% expressed IB₄. These data indicate that EIAV vectors pseudotyped with the rabies-G glycoprotein can mediate efficient gene transfer to all classes of DRG neurons after injection into the spinal cord. Injection of the EIAV vector into the spinal cord does not induce a significant inflammatory response, as is evident by normal Nissl staining and the lack of increased markers for macrophages (0X42), microglia (EDl) and T cells (CD8) (Fig. 7). Furthermore, no increased vascular density in the injected spinal cord is observed (Fig. 7e).

d) RARβ2 promotes regeneration of primary sensory afferents in the DREZ

Adult Wistar rats receive intraspinal injections (at spinal levels C5-8) of 3 - 6 x 10⁶ transducing units of EIAV-RARβ2 or control EIAV-LacZ 3 weeks prior to receiving complete crush injuries to the corresponding cervical dorsal roots. Primary afferents are labeled by transganglionic injection of the anterograde tracer biotinylated dextran amine (BDA, MW 10,000) after injury in order to visualize regeneration of injured sensory axons. In control animals, BDA-labeled axonal profiles are found to regenerate up to but not into the CNS portion of the DREZ, which is delineated from the peripheral component by the lack of laminin staining (Fig. 4a). In contrast, numerous BDA-labeled regenerating fibers are detected that cross the inhibitory DREZ and penetrate into the grey matter of the dorsal horn in EIAV-RARβ2 treated animals (Fig. 4b). At 5 weeks post-injury, the number of BDA-positive fibers crossing into the DREZ and present in the spinal cord portion of the DREZ is significantly higher in EIAV-RARβ2 compared to EIAV-LacZ treated animals (Fig. 4c; n=9 for EIAV-LacZ and n=l l for EIAV-RARβ2, P<0.05 2-way ANOVA, followed by Tukey's post-hoc test). The presence of BDA-positive fibers of different thickness is observed in the white matter of the cuneate fasciculus and in laminae I and II of the grey matter, indicating that both myelinated and unmyelinated fibers regenerated into the spinal cord. These observations are confirmed by an increase in NF200-, CGRP- and IB₄- immunoreactive fibers crossing the DREZ of EIAV-RARβ2 treated animals compared to EIAV-LacZ animals (Fig. 4d-i). These results are consistent with the observed EIAV vector expression in both large- and small-diameter DRG neurons. Furthermore, axonal regeneration in EIAV-RARβ2 treated animals is correlated with increased immunoreactivity for cAMP in the transduced DRG compared to control EIAV-LacZ animals (Fig. 8).

e^") Regenerated axons enhanced post-synaptic activity of second-order neurons in the spinal cord After entering the inhibitory DREZ, the regenerating sensory fibers need to find and activate the appropriate second-order neuronal targets in the spinal cord. Noxious heat stimuli is applied to the forelimbs of both EIAV-LacZ and EIAV-RARβ2 treated animals after injury to test for such functional activation. Noxious heat stimuli to forelimbs normally induces c-fos activation and phosphorylation of the extracellular signal-regulated kinase ERK in post-synaptic neurons in superficial laminae of the dorsal horn Hunt et al (1987) Nature 328: 632-4). Increased Fos- and phosphorylated ERK- (pERK) immunoreactivity in the dorsal horn laminae I/II is detected in EIAV-RARβ2 treated animals (Figs. 5a, b) compared to EIAV-LacZ animals (Figs. 5c, d) and expression levels are comparable to that of unlesioned controls (Figs. 5e, f). Quantification of these immunoreactive nuclei indicated significantly elevated numbers of Fos-positive nuclei in RARβ2 rats compared to control rats (11.3 + 0.8 and 6.0 + 0.4 respectively) as well as numbers of pERK-positive nuclei (3.7 ± 0.4 and 1.5 ± 0.5 respectively) (Figs. 5g, h; n=6 in each group, P<0.05 Students' 2-tailed unpaired t-test in both cases). EIAV-RARβ2 treated animals with ablated dorsal roots are not significantly different from EIAV-LacZ treated animals (Figs. 5g, h). These observed postsynaptic responses in EIAV-RARβ2 animals suggest that RARβ2 promotes C-fϊber regeneration and provide evidence for functional effects of regenerated axons on the appropriate cellular targets in the spinal cord.

This anatomical evidence for RARβ2-induced regeneration may be correlated to functional recovery. Control and treated animals are assessed for forelimb function using a range of sensory and locomotor tasks over 4 weeks post-injury, with the observer blinded to treatment (n=8 per group). Adhesive tape-removal tasks assessed sensory (awareness of the tape) and motor (ability to remove tape) function. During testing of the uninjured forelimbs both control EIAV-LacZ and EIAV-RARβ2 rats performed tasks quickly (2.0 ± 0.6s and 1.7 + 0.4s (EIAV-LacZ), and 7.5 ± 1.8s and 5.0 ± 0.6s (EIAV-RARβ2), for sense and removal tasks at 4 weeks post-rhizotomy respectively, Figs. 6a, b). The injured forelimb of EIAV-LacZ rats shows significantly increased latencies in sense and removal tasks (50.7 + 18.1s and 54.0 + 16.4s); in contrast EIAV-RARβ2 treatment produces recovery of function for sense and removal tasks (6.3 ± 4.2s and 16.5 ± 4.4s; P<0.05 2-way RM ANOVA followed by Tukey's post-hoc test, Figs. 6a, b). In a paw-reaching test, the lesioned forelimb of EIAV-LacZ treated rats is significantly impaired in its ability to reach and grasp food pellets in a staircase apparatus (Montoya et al (1991) J. Neurosci Methods 36:219-28). In comparison the lesioned forelimb of EIAV-RARβ2 demonstrates significant improvement (food displacement scores were: 0.8 + 0.5 (EIAV-LacZ injured) and 2.8 + 0.9 (EIAV-RARβ2 injured), and 3.3 ± 0.7 (EIAV-LacZ control) and 4.5 ± 1.1 (EIAV-RARβ2 control); P<0.05 2-way RM ANOVA followed by Tukey's post-hoc test, Fig. 6c). To test for locomotion the number of footslips made by the forelimbs are recorded when rats crossed a horizontal ladder or a narrow beam. No footslips are made by the control uninjured forelimb in EIAV-LacZ and EIAV-RARβ2 groups in both tests (Figs. 6d, e). In the ladder crossing test, there are fewer footslips made by the injured forelimb in the EIAV-RARβ2 group compared to the EIAV-LacZ group (3.4 ± 2.4 and 0.3 ± 0.2 respectively; P<0.05 2-way RM ANOVA followed by Tukey's post-hoc test, Fig. 6d). There are no significant differences between the 2 groups in the number of footslips made by the injured forelimb over time in the beam crossing test (Fig. 6e). Walking patterns are also assessed by analyzing footprint spacing. There are no significant differences in stride length between EIAV-LacZ and EIAV-RARβ2 treated animals (Fig. 6f). The lack of difference between the two groups in the last two tests is not surprising as this injury model involved the damage to the dorsal roots (sensory input), leaving the ventral roots (motor output) intact. Nevertheless these results indicate that EIAV-RARβ2 promoted recovery of sensory-motor functions after crush injury to the C5-C8 dorsal roots. This improvement in performance in EIAV-RARβ2 animals is not due to enhanced local plasticity in the spinal cord as EIAV-RARβ2 treated animals with ablated dorsal roots show severe impairment in all behavioral tests throughout the experiment (Fig. 9).

Experimental Procedures

Viral vector production

EIAV vector genomes are constructed from pSMART2 lentiviral vectors as previously described (Bienemann et al (2003) MoI Ther 7:588-96). The RARβ2 or LacZ gene is inserted under the control of a minimal hCMV promoter in an EIAV transfer vector containing a 5' cPPT element and a 3' WPRE enhancer (vector map illustrated in Supplementary Fig. If). Viral vector stocks pseudotyped with the rabies-G envelope glycoprotein are prepared by triple plasmid transient transfection of HEK293T cells as previously described (Mazarakis et al (2001), as above; Azzouz et al (2002) J Neurosci 22:10302-12; Bienemann et al (2003) as above; Mitrophanous et al (1999) Gene Ther 6: 1808-18) . The titer of EIAV-LacZ is determined by transduction of dog osteosarcoma Dl 7 cells (4 x 10⁸ TU ml⁴) while the titer of EIAV-RARβ2 (8 x 10^s TU ml^"1) is calculated by determining the normalized viral RNA genome copy number using real-time quantitative polymerase chain reaction analysis and comparing it to EIAV-LacZ as previously described (Martin-Rendon et al (2002) MoI Ther 5: 566-570).

DRG neuron culture

Neuronal dissociated DRG cultures are prepared as previously described (Gavazzi et al (1999) J Neurosci 11: 3405-14). Adult male Wistar rats (200-25Og) are sacrificed according to institutional and UK Home Office Regulations and the DRGs are dissected and transferred to Ham's F12 medium (Gibco, UK). DRGs are desheathed, their roots trimmed and then digested in 0.125% collagenase (Sigma, UK) at 37⁰C for 2 hr, after which they are mechanically dissected by trituration with a PlOOO Gilson pipette in ImI modified Bottenstein and Sato's culture medium (BS) in Ham's F12. The resulting cell suspension is centrifuged at 600 rpm for 8 min through a cushion of 15% bovine serum albumin (BSA, Sigma). The dissociated neurons are resuspended in lOOμl of calcium- and magnesium-free HBSS (Gibco) containing 50μg/ml DNase (Type I, Sigma) and 250μg/ml soybean trypsin inhibitor (Type II, Sigma) and diluted in modified BS culture medium to a final concentration of approximately 1600 cells/ml. Cells (500 per well) are plated in each well per eight-well plate (Labtek), which are precoated with poly-L-lysine (2 mg/ml; Sigma) and O.lμg/ml solution of EHS laminin (Sigma) for at least 2 hr at 37⁰C prior to plating. Neurite outgrowth from dissociated DRG neurons is affected by low cell density and low laminin concentrations (data not shown). Neuronal cultures are incubated at 37⁰C in a humidified atmosphere containing 5% CO₂. Viral transduction is carried out by adding the appropriate viral vectors at a multiplicity of infection (MOI) of 10 at DIVO. For cAMP inhibition experiments, cultures are incubated from DIVl in 2', 5'-dideoxyadenosine (DDA, Sigma) or Rp-adenosine-3 ',5 '-cyclic monophosphorothioate (sodium salt, Rp- cAMP, Sigma) at the concentrations indicated. After 3 days in culture, neurons are either harvested for measurement of cAMP using a competitive immunoassay, according to manufacturer's instructions (Amersham, UK), or fixed for 30 min in 4% paraformaldehyde for assessment of neurite outgrowth. Cells are permeabilized with methanol at -2O⁰C for 3 min and washed with PBS. Cells are incubated at room temperature for at least 2 hr with a combination of mouse βm tubulin (1:1000) with either rabbit β-galactosidase (1:300, Europa Biolabs) or rabbit RARβ (1:50, Santa Cruz). After further rinsing in PBS, the cultures are incubated at room temperature for 1 hr with a mixture of Alexa488 (1:1000, Molecular Probes) and Alexa546 (1:1000, Molecular Probes). Following further PBS washes, cells are mounted with FluorSave™ reagent (Calbiochem, UK) and observed under a Zeiss microscope. The length of the longest neurite for the first 150-220 neurons encountered when scanning the slide in a systematic manner is determined using an image analysis program (SigmaScan Pro 4.01), and expressed as mean length + SEM (Gavazzi et al (1999) as above).

Preparation of co-culture Animals are administered with EIAV-RARβ2 or EIAV-LacZ viral vector as described below (n=3 in each group). 3 weeks post surgery, transduced DRG neurons are prepared for culture as described above. Cryosections of the spinal cord are prepared according to the method of Golding et α/.(1999-Glia 26:309-323). Briefly, the transduced or naive cervical cord is removed and placed between two sterile glass plates on dry ice. The weight of the glass plate is sufficient to gently compress the cord so that cut longitudinal sections contained the dorsal root and the dorsal root surface of the spinal cord in the same plane. 10 μm longitudinal sections of the frozen cord are obtained on a cryostat and thaw-mounted onto sterile 13 mm diameter poly-L-lysine coated glass coverslips. The dissociated DRG neurons in modified BS culture medium are placed at 150 cells/well onto the cryosections which are individually housed in 4 well plates. Cryocultures are incubated for 3-4 days prior to fixation, after which immunostaining and quantification of neurite outgrowth on spinal cord are performed as described above. At least 50-60 transduced neurons in each experiment are analyzed. Primary antibodies used are mouse β-galactosidase (1: 600, Promega), mouse RARβ (1: 300, Chemicon), rabbit GAP43 (1: 1500, Chemicon) and goat GFAP (1 : 300, Santa Cruz). After further rinsing in PBS, the cultures are incubated at room temperature for 1 hr with a mixture of Cy3 (1: 600), AMCA (1: 300) and FITC (1:300).

Viral vector delivery and nerve labeling

All rat experiments are approved by the local veterinarian and ethical committees and carried out according to UK Home Office Regulations. In anaesthetized adult Wistar rats (ketamine/medetomidine 60 mg kg^"Vθ.25 mg kg^"1; i.p.), cervical spinal cords are exposed by laminectomy and EIAV vectors are infused unilaterally into four sites of the spinal cord (C5-C8 spinal level) at a rate of 0.2 μl min^"1 via a finely drawn glass micropipette attached to an infusion pump. 3 - 6 x 10⁶ transducing units of each vector are injected. Three weeks following vector delivery C5-C8 dorsal roots are crushed unilaterally (two times each, 10 s per crush) midway between the DRG and DREZ by an operator blinded to the treatment groups and then tested behaviorally for 4 weeks (n=8 per group). For transganglionic tracing, 1 μl of 10% BDA (MW 10,000, Molecular Probes) is injected directly into the DRG after behavioral testing is carried out. Animals are sacrificed at 10 days after BDA labeling to assess axonal regeneration (n=9 EIAV-LacZ and n=l l EIAV-RARβ2). For noxious stimulation experiments, animals are anaesthetized by urethane two hours before termination of experiment and primary afferents are stimulated by dipping injured forelimbs into a 52⁰C waterbath (3 x 20 s at 90 s intervals and is repeated at 5 min before termination, 2 h from the initial dip; n=6 each for EIAV-LacZ and EIAV-RARβ2).

In situ hybridization

Animals are irreversibly anaesthetized and perfused transcardially with heparinised 0.9% NaCl solution and 4% paraformaldehyde in 0.1 M phosphate buffer. The cervical cords with attached DRGs are dissected, frozen and sectioned (20μm) on a cryostat. For detection of EIAV-RARβ2 transcripts, in situ hybridization is carried out using non-radioactive ribroprobes specific to RARβ2 as previously described (Zelent et al (1989) Nature 339: 714-7; Rattray and Michael "In situ hybridization-a practical approach" 1998) and development is performed after 2 h hybridization.

Immunohistochemistry

Sections are analyzed by X-gal (5-bromo-4-chloro-3-indolyl-β-D-galactoside) staining and immunohistochemistry. RARβ is also detected in 20μm sections of embryonic day 14 DRGs. Primary antibodies are used as follows: mouse NF200 (1: 400, N52, Sigma), rabbit CGRP (1:8000, Sigma), Griffonia simplicifolia Isolectin B4 (10 μg/ml, Sigma), rabbit β- galactosidase (1:300, Europa Labs), rabbit RARβ (1:50, Santa Cruz), rabbit laminin (1 :200, Sigma), rabbit Fos (1:10000, Oncogene Sciences) and rabbit pERK (1:200, Sigma). For inflammatory response markers (Supplementary Fig. 1), antibodies used are NeuroTrace fluorescent Nissl (1:100, Molecular Probes), 0X42 (1:100, Chemicon), EDl (1:1000, Chemicon), CD8 (1:100, Serotec) and Glutl (1:1000, Serotec). cAMP staining (Supplementary Fig. 2) is performed with rabbit anti-cAMP (1:1000, Chemicon). Extra- avidin conjugated to fluorescein isothiocyanate (1:200, Sigma) is used to detect BDA- labeled axonal tracts. Secondary antibodies used are conjugated either to fluorescein isothiocyanate, Texas Red™ (Jackson Labs) or Alexa Fluor™ 488 and 546 (Molecular Probes) or AMCA (Jackson Labs). Quantitative analysis of axonal regeneration is carried out by counting BDA-labeled fibers at measured intervals within a 1-mm square grid graticule in the peripheral and central regions of DREZ (delineated by laminin staining) by a blinded experimenter. At least 36 random sections are counted for each animal. For transduction efficiency, immunopositive cells are counted in 6 sections per animal and expressed as a percentage of β-galactosidase positive cells. For Fos and pERK quantification, immunopositive cells are counted in 36 random sections per animal.

Behavioral analysis

Testing is carried out by experimenters blinded to the treatment for 1 week before and for 4 weeks after crush injury (n=8 each in EIAV-LacZ and EIAV-RARβ2). In a tape removal test (Thallmair et al (1998) Nat Neurosci 1: 124-131), adhesive tape (1.5 cm x 1 cm) is placed on each forepaw separately, and the times taken to sense the tape (indicated by paw shake) and remove the tape are scored. A cut-off timepoint of 2 min is used. The paw reaching test assesses the rats' ability in reaching and grasping food pellets in a staircase apparatus, and measures side bias, maximum forelimb extension and grasping skill (Montoya et al (1991) J Neurosci Methods 36:219-228). Rats are placed in the staircase apparatus for 10 min and food displacement is scored. For locomotor tasks (adapted from Kunkel-Bagden et al (1993) Exp Neurl 119: 153-164), rats are trained to cross a horizontal beam (2.5 cm X 100 cm) or ladder (18 cm X 100 cm with each rung 5 cm apart) and the number of forelimb footslips (off the beam or below the plane of the ladder) is recorded, hi footprint analysis (adapted from Kunkel-Bagden et al (1993) as above), the forepaws are covered with ink to measure walking patterns during continuous locomotion across a wooden runway, and stride lengths and widths are calculated. Behavioral responses between treated and control groups are compared using 2-way RM ANOVA followed by Tukey post hoc test.

Example 2 - NCS-I expression is differentially regulated by overexpression of RARβ2 in adult spinal cord The results of Experiment 1 suggest that RARβ2 can activate regeneration programs in injured neurons to promote axonal outgrowth. To investigate genes/proteins which may be involved in this process, microarray analyses are performed to identify genes that are differentially regulated by overexpression of RARβ2 in adult spinal cord.

Microarray and Q-PCR results

Adult rat spinal cords are transduced by EIAV-RARβ2 and a control vector EIAV- LacZ by direct in vivo injection of VSV-G pseudotyped lentiviral vectors into the spinal cord. 4 weeks after transduction, the spinal cords are harvested and RNA is extracted. RNA is utilised for microarray analyses using Affymetrix Rat Neurobiology (U34) arrays (n=3 in each group). Raw signals are normalised and analysed in GeneSpring. Raw data is filtered using a 1.5 fold cut off and statistically analysed using 2 tailed 2 sample Welch t test. NCS- 1 is identified to be upregulated in RARβ2 transduced cord by 2.44+0.4 fold compared to LacZ-transduced cord (P<0.05). Microarray data is confirmed by Q-PCR using primers specific for rat NCS-I in RARβ2-transduced rat spinal cord tissue (n=3). NCS-I is upregulated by 2.49+0.7 fold in RARβ2-transduced cord compared to LacZ-transduced cord (Figure 10).

Example 3 - Overexpression of NCS-I stimulates axonogenesis in adult neurons

In order to test if overexpression of NCS-I stimulates axonogenesis in adult neurons, lentiviral vectors encoding NCS-I are constructed. Rat NCS-I is PCR amplified from a rat brain cDNA library and cloned into EIAV transfer vectors. Lentiviral vectors encoding NCS-I are prepared via a standard 3-plasmid transient transfection method and titres are determined at 5.8 x 10⁸ integration units/ml. EIAV-NCS-I is used to transduce various types of adult neurons (dorsal root ganglion, spinal cord and cortical). In all three adult neuronal types NCS-I overexpression enhances neurite outgrowth compared to control LacZ transduced neurons transduced neurons (Figures 1 IA-F). Significantly longer neurites are observed in NCS-I transduced neurons compared to control neurons (Figure HG; PO.001, Student's t-test, from 3 independent experiments). In addition, overexpression of NCS-I in adult dorsal root ganglion (DRG) neurons increases the percentage of process bearing neurons to 53.2 ± 5.0% from 23.5 ± 4.6 % in control cultures. The data shown above demonstrate that NCS-I expression is regulated by RARβ2 overexpression in the rat spinal cord. Overexpression of NCS-I on its own, via a lentiviral vector, promotes neurite outgrowth in adult DRG, spinal cord and cortical cultures. Neurons are known to have limited intrinsic capacity to regenerate in the adult central nervous system. Overexpression of NCS-I, on its own or as part of the signalling cascade initiated by RARβ2, boosts this intrinsic growth capacity and stimulates neurite outgrowth.

Example 4 - Overexpression of NCS-I promotes axonal regeneration in vivo

a^*). Viral vector production and vector development

Concentrated viral preparations of EIAV .NCS-I and EIAV .LacZ vectors are produced using a transient transfection system.

Method: Viral vector stocks, of EIAV .NCS-I and EIAV.LacZ, are produced by transient transfection of human embryonic kidney 293 T cells plated on 10 cm dishes (3.5 x 10⁶ cells/dish). 3 DNA components; 2 μg vector plasmid, 2 μg of gag/pol plasmid (pONY3.1) and 1-2 μg of plasmid encoding envelope glycoprotein are added to a mix containing FuGENE 6™ and OptiMEM. Sodium butyrate (10 mM final concentration) is added after 16h transfection. Superaatants are harvested 24-42 hours after transfection and filtered. The supernatants are ultracentrifugated at 6000 x g at 4⁰C for at least 18 hours, followed by ultracentrifugation at 50 000 x g at 4⁰C for 90 min. After the virus has been in formulation buffer for 2-3 hours at 4⁰C, it is aliquoted and stored at -8O⁰C.

The biological titres of the viral preparations are determined by transducing canine osteosarcoma cells. After 2-3 days incubation, the cells are incubated in 5-bromo-3- indolyl-β-D-galactosidase (X-gal) solution and blue colonies are counted. In addition, RNA titres of the viral preparations are also calculated by determining the number of viral RNA genomes per ml of viral stock solution using quantitative PCR analysis. Vectors are also tested qualitatively (mycoplasma, endotoxin and sterility testing) prior to use in in vivo experiments.

by To determine if NCS-I transduced neurons in culture can induce an enhanced neurite outgrowth on inhibitory substrates. The present inventors have shown that NCS-I overexpressed in adult DRG and cortical neurons can significantly increase the length of neurites and stimulate neuritogenesis in culture. However, it is more relevant to conditions in the uninjured and injured CNS if NCS-I can reproduce similar results if grown on an inhibitory substrates such as on spinal cord cryosections (Golding et al., 1999) and on myelin substrate (Zheng et al., 2005) since these same DRG and cortical neurons in vivo will not extend axons into these inhibitory environments.

Methods: Adult Wistar rats (220 - 250 g) (n = 5) are sacrificed and multiple dorsal root ganglia (DRG) and cord dissected out for dissociation and cryoculture respectively. The

DRG are desheathed, attached roots trimmed away and placed into 0.125% collagenase for

2 h. The DRG neurons are mechanically dissociated and separated from debris by centrifugation in 15% bovine serum albumin solution. The DRG neurons are in Bottenstein and Sato's medium until ready to plate onto either spinal cord cryosections or onto myelin substrate at a density of 300 neurons per well.

Another set of adult male Wistar rats (220 - 250 g) (n = 5) are sacrificed and their cortices dissected out for dissociation and the spinal cord and cortex dissected out for cryoculture. The adult cortical neurons are similarly prepared as follows. Briefly, the cortices are cut into 0.5 mm longitudinally sections using a Mcllwain tissue chopper, white matter trimmed away then dissociated in 2 mg/ml papain. The cortical neurons are mechanically dissociated and separated from debris by centrifugation in four 1-ml steps of Optiprep in B27/HibernateA medium. Fractions containing neurons are collected, washed and resuspended in B27/NeurobasalA medium, ready for plating at a density of approximately 2000 neurons per well.

Cover slips are either covered with spinal cords cryosections (10 μm) or with 25 μl of 4 μg/ml myelin solution and left overnight to dry before plating with neurons. To prepare the CNS myelin substrate, CNS myelin extract are prepared in advance and stored at -8O⁰C until required.

After 18 h for the neurons to settle onto the spinal cord cryosection or myelin substrate, EIAV .NCS-I pseudotyped with rabies virus (ERA) envelope is added to the media at MOI 100. It has been demonstrated that EIAV vectors pseudotyped with ERA mediate strong transduction of all types of DRG neurons and CNS neurons (Mazarakis et al., 2001; Yip et al., 2004, as above). The neurons are studied for up to a further 3 days in vitro before they are prepared for histology by fixation in 4% paraformaldehyde and permeabilised with cold methanol.

Outcome measures involve cultures double immunostained for GAP43 (growth and regeneration marker) and for PGP9.5 (pan neuronal marker) or for markers of specific subpopulations (NF200, P2X3, CGRP). For each time point and myelin concentration at least 100 neurons are assessed for: the length of the longest neurite, percentage of cells with neurites. Survival is also checked under different conditions.

c^*). To determine if overexpression of NCS-I can restore sensory and motor function in a spinal cord injury model.

Broadly there are two different ways in which animals can recover from spinal cord injury.

These are either regeneration of damaged axons or anatomical plasticity of un-lesioned axons, which sprout to restore the lost functions. Both these processes in adult CNS under normal conditions have a very limited capacity for functional and anatomical repair after lesion. The following example is to assess both types of putative anatomical growth:

i) To demonstrate axonal regeneration, a dorsal column crush model is used (Wong et al., 2005 as above).

Method: Adult male Wistar rats (n = 6 per group) are anaesthetised using a combination of ketamine and medetomidine and fixed in a stereotaxic frame. The skull is exposed and 6 holes per side are made at the sensorimotor cortex region using coordinates ranging from, in reference to bregma, -1.5 to 2 mm anteriorposteriorly, 1.5 to 3.5 mm laterally and -2 mm dorsoventrally. A viral vector (EIAV.NCS-1 pseudoryped with vesicular stomatitis virusenvelope (VSV-G) are directly injected at 1 μl per site via a finely pulled glass micropipette at a slow rate using a microinfusion pump. EIAV vector pseudotyped with a VSV-G envelope produce strong anterograde transgene expression (Mazarakis et al., 2001; Yip et al., 2004, as above). Four weeks after the viral injection to allow maximal transgene expression, SCI is induced by performing a bilateral dorsal column crush at the level of C4 using fine forceps. Following injury, regeneration of the descending motor system are assessed behaviourally, electrophysically and anatomically using established protocols. These include grid and beam walking, rearing, forelimb pellet retrieval (staircase test), forelimb grip strength, and footprint analysis, which are determined once per week (Kunkel-Bagden et al., 1993 as above). Assessment of all parameters are made weekly for 6 weeks post lesion and compared with the appropriate groups such as a control vector (EIAV.LacZ) treated group as well as a sham operated group. At the end of the behavioural assessment, the animals are sacrificed and perfused transcardially with 4% paraformaldehyde and tissue collected for histology. Regeneration of the descending corticospinal axons is assessed by injecting biotinylated dextran amine (BDA) bilaterally into the sensorimotor cortex 4 weeks before perfusion. Behavioural outcomes will then be correlated with anatomical data in order to determine the efficacy of NCS-I. Electrophysiology: cord dorsum potentials above and below the lesion site are measured following electrical activation of one sensorimotor cortex in terminal experiments performed 6 weeks after SCI under anaesthesia, as previously described (Bradbury et al., 2002 as above).

ii) To assess whether NCS-I promotes sprouting of intact axons in adult animals, a rat pyramidotomy model is used, unilaterally lesioning one CST.

Method: Adult male Wistar rats (n = 6 per group) are injected with VSV-G EIAV.NCS-1 or EIAV.LacZ (as a control) into the right sensorimotor cortex as described above. After 4 weeks post viral injection the animals are re-anaesthetized and the occipital bone are exposed using a ventral approach. The left medullary pyramid is exposed and the left pyramidal tract is transected rostral to the decussation with the basilar artery serving as a landmark for the midline.

Following injury, plasticity of the left descending CST are assessed behaviourally, electrophysically and anatomically as described in 3 a, above. The anatomy and electrophysiology are assessed of the 'spared' intact CST.

d). Delayed treatment with NCS-I.

The former experiments using pre-treatment with EIAV.NCS-1 are useful in elucidating the efficacy of NCS-I in promoting regeneration. However, in the clinical scenario, treatment will follow injury. Clinical trials of SCI treatment are likely to be applied to stable patients, i.e. with established injuries. With time, the inflammatory response and scar formation following injury may alter the nature of the environment for axonal regeneration. In order to address this issue, the EIAV.NCS-1 in applied two weeks post lesion in this example. The methods described in experiments ci and ii, above, are used. Method: Adult male Wistar rats (n = 6 per group) are subjected to spinal cord injury. The animals are tested in the first 2 weeks post lesion using behavioural tasks (see above) to demonstrate the emergence of functional deficit. On the second week VSV-G EIAV.NCS-1 or EIAV.LacZ is injected into the sensorimotor cortex, and the animals are subjected to weekly behavioural tasks, BDA tracing, electrophysiology and then sacrificed for histology as described in ci and ii, above.

AU publications mentioned in the above specification are herein incorporated by reference. Various modifications and variations of the described methods and system of the invention are apparent to those skilled in the art without departing from the scope and spirit of the invention. Although the invention has been described in connection with specific preferred embodiments, it should be understood that the invention as claimed should not be unduly limited to such specific embodiments. Indeed, various modifications of the described modes for carrying out the invention which are obvious to those skilled in cellular studies using flow cytometry or related fields are intended to be within the scope of the following claims.

Claims

1. A system capable of capable of delivering neuronal calcium sensor- 1 (NCS-I), or a nucleotide sequence capable of encoding neuronal calcium sensor-1 (NCS-I), to a target cell, for promoting neurite outgrowth.

2. A system according to claim 1, which is a viral vector.

3. A system according to claim 2, which is a lentiviral vector.

4. A system according to claim 3 which is derivable from equine infectious anaemia virus (EIAV).

5. A system according to any preceding claim, wherein the target cell is a neuron.

6. A system according to claim 5, wherein the target cell is an adult neuron.

7. A system according to claim 5 or 6, wherein the target cell is an injured neuron.

8. A system according to any preceding claim for treating a condition selected from the following: spinal cord injury, avulsion injury, traumatic brain injury, brachial plexus injury, stroke and a neurodegenerative disease.

9. A pharmaceutical composition comprising a system according to any preceding claim.

10. A pharmaceutical composition according to claim 8 which also comprises RARβ2, or a nucleotide sequence capable of encoding RARβ2.

11. A method for promoting neurite outgrowth which comprises the step of administering to a subject an effective amount of a system according to any of claims 1 to 8 or composition according to claim 9 or 10, or an agent capable of promoting the expression or activity of NCS-I.

12. A method for treating spinal cord injury which comprises the step of administering to a subject an effective amount of a system according to any of claims 1 to 8 or composition according to claim 9 or 10, or an agent capable of promoting the expression or activity of NCS-I.

13. The use of a system according to any of claims 1 to 8 or composition according to claim 9 or 10, or an agent capable of promoting the expression or activity of NCS-I in the manufacture of a medicament for the promotion of neurite outgrowth.

14. The use according to claim 13, to promote regeneration processes in damaged axons so that they re-establish functional connections.

15. The use according to claim 13 or 14, to promote synaptic plasticity.

16. The use according to any of claims 13 to 15, for the treatment of a condition selected from the following: spinal cord injury, avulsion injury, stroke, a neurodegenerative disease, traumatic brain injury and brachial plexus injury.

17. A method for delivering NCS-I, or a nucleotide sequence capable of encoding NCS-I, to a cell in vitro using a system according to any of claims 1 to 8.

18. A cell containing or expressing NCS-I, produced by a method according to claim 17.

19. A cell according to claim 18, which is a supporting cell.

20. The use of a cell according to claim 18 or 19 in the manufacture of a medicament for use in implantation or transplantation to a subject.

21. A system substantially as described herein, with reference to the accompanying examples.

22. A viral vector comprising a nucleotide sequence encoding neuronal calcium sensor-1 (NCS-I).

23. The viral vector of claim 22, which is a lentiviral vector.

24. The lentiviral vector of claim 23, which is an equine infectious anaemia virus (EIAV) vector.

25. A vector comprising a nucleotide sequence encoding NCS- 1 and RARβ2.

26. The vector of claim 25, which is a viral vector.

27. The viral vector of claim 26, which is a lentiviral vector.

28. The lentiviral vector of claim 27, which is an EIAV vector.

29. A pharmaceutical composition comprising the viral vector according to claim 22 or the vector according to claim 25, and a pharmaceutically acceptable carrier.

30. A cell ex vivo comprising the viral vector according to claim 22 or the vector according to claim 25.

31. The cell according to claim 30, which is a neuron.

32. The neuron of claim 31, which is an adult neuron.

33. The neuron of claim 31 , which is an injured neuron.

34. A method for promoting neurite outgrowth comprising delivering to a neuron the viral vector according to claim 22 or the vector according to claim 25, wherein the NCS-I is expressed in the cell thereby promoting neurite outgrowth.

35. The method according to claim 34, wherein the neuron is an adult neuron.

36. The method according to claim 34, wherein the neuron is an injured neuron.

37. A method for treating neuronal injury comprising administering to a subject in need of such treatment the viral vector according to claim 22 or the vector according to claim 25, wherein the NCS-I is expressed in a neuron in the subject, thereby promoting neurite outgrowth and treating neuronal injury in the subject.

38. The method according to claim 37, wherein expression of NCS-I promotes regeneration processes or synaptic plasticity.

39. The method according to claim 37, wherein the neuronal injury is spinal cord injury, avulsion injury, stroke, neurodegenerative disease, traumatic brain injury, or brachial plexus injury.