IE20020467A1 - An expression vector for the central nervous system - Google Patents

Abstract

A Semiliki Forest based vector system for delivery to the central nervous system is described, as is its use as an expression for delivery to the central nervous system and pharmaceutical compositions for use in drug delivery to the CNS.

Description

Field of the Invention The present invention relates to a vector system based on Semliki Forest Virus, and to its use as an expression system for delivery to the central nervous system (CNS) and to pharmaceutical compositions for use in drug delivery to the CNS.

Background to the Invention Work on viral pathogenesis has traditionally been focused on molecular determinants of viral pathogenicity and host defence against infection. In the past few years, it has become apparent that some characteristics of viruses, which normally contribute to pathogenesis, may be manipulated to treat disease rather than cause it.

Thus viruses are being developed as vectors for vaccine construction, as gene therapy agents, and as cancer therapy agents.

The main viruses that have been exploited as vectors in this work so far are DNA viruses such as adeno and adeno-associated viruses, and RNA viruses that synthesise a DNA copy of their genome such as retroviruses. Herpes viruses, which are large complex DNA viruses, are just beginning to be exploited as central nervous system (CNS) vectors, based on their ability to induce latency in neurons. The cytomegalovirus (CMV) immediate early promoter has also been used as the basis for the construction of prototype DNA vaccines (1). Vaccinia virus, particularly the highly attenuated modified vaccinia Ankara (MVA) strain, has been used as a vector for the construction of prototype vaccines, but this is a large DNA poxvirus which poses biosafety problems as a vaccine vector. An alternative poxvirus, canarypox, has also been used as a vaccine vector because its replication is restricted in mammalian cells.

Recently, RNA vectors have been developed, based on the alphavirus replicon, that have several advantages over DNA-based vectors. Such vectors include those based on Sindbis virus and Venezuelan equine encephalitis virus (VEE). These vectors OPUUO PU3UC iNSPfiCTtetf } IMOER SECTION 25 ΑΝ0 RUKfe JNL ; it Ο 2 046 7 cannot be used to permanently replace the function of defective genes, since their expression is transient and they do not integrate a copy of their genome into that of the host cell. However, for some applications these properties can be advantageous safety features. The expression of cloned genes induced by alphavirus vectors is transient but high level. The present inventors have also shown that persistence of administered Semliki Forest virus (SFV) RNA vectors in host tissue is short-term (less than 7 days compared to months for DNA vectors, [16]). In addition, high levels of transgene expression was observed in organotypic hippocampal slices in vitro and in vivo in rodent brain following intracerebral inoculation using SFV vectors expressing reporter molecules (Lundstrom et al, 2001) [27].

Thus alphavirus vectors are ideal for such applications as vaccine construction, tumour therapy and treatment of acute disease episodes.

SFV is an enveloped positive-stranded RNA virus of the genus Alphavirus of the family Togaviridae. It is a relatively simple virus encoding only nine functional proteins; four non-structural proteins are concerned with viral RNA synthesis, while the structural proteins form the capsid (the C protein) and the envelope (the El, E2 and E3 proteins). It is able to infect most animal cells, including mammalian, avian and insect cells [1]. SFV has been widely exploited as a model for viral neuropathogenesis [2,3], and the SFV4 strain of SFV, derived from an infectious clone, is lethal when given intranasally (i.n.) to adult mice. Avirulent strains such as A7 do not kill the mice but induce immune-mediated central nervous system (CNS) demyelination [4], Recombinant alphavirus RNA molecules and the expression of heterologous proteins from them is known from US Patent No. 5,739,026. The recombinant RNA comprises an SFV RNA genome and an exogenous RNA molecule, with at least one stop or deletion codon mutation to prevent at least one SFV structural protein being made by a host cell, and the exogenous RNA being inserted into a region of the SFV genome which is non-essential to replication.

PCT Publication No. WO 95/27069 describes the use of naked (that is, unencapsulated by viral proteins) alphavirus RNA with heterologous protein gene(s) in place of structural alphavirus genes, as a vaccine composition. The RNA is optionally stabilised with lipid.

IE 02 046 1 PCT Publication No. WO 95/27044 describes alphavirus cDNA vectors comprised of recombinant cDNA consisting of cDNA derived from an alphavirus and heterologous cDNA coding for a desired substance.

Another SFV vector system consists of a self-replicating RNA vector (rSFV) in 5 which a heterologous gene replaces the SFV structural proteins, while the structural proteins are supplied in trans by helper RNA(s) to form encapsidated particles [5,6,7], The helper RNA(s) have the packaging sequence deleted to prevent encapsidation. To improve the biosafety of this vector, a split-helper system was developed. Two separate RNAs supply the capsid and spike proteins, which prevents the production of replication-proficient viruses. A further mutation, an S219A substitution, in the capsid sequence encoded by the helper abolishes the self-cleavage ability of the capsid protein [8], This prevents the production of infectious virus in the unlikely event that a complete coding sequence is generated by recombination. The recombinant particles produced by this vector system are capable of expressing cloned genes transiently during a single round of virus multiplication, but progeny particles will not be produced because the packaged vector RNA lacks the viral structural protein genes.

The development of the SFV vector system has opened a wide spectrum of opportunities for vaccine development [9,10], SFV has the potential to infect a broad range of cells, and most humans and animals outside Africa have no pre-existing immunity to the vector. As an RNA vector, replication is cytoplasmic so there is no risk of integration into the chromosome, and the virus genome does not persist in host tissue [1,6,11,12], As vaccines, SFV vectors have shown the ability to induce strong humoral and cell-mediated immunity against several disease agents [9,11,13-15], Following intramuscular inoculation with rSFV particles, the vector is detected at the injection site only, with mild lesions [16]. Compared to DNA immunisation encoding the same antigens, rSFV particles induce stronger immune responses [9].

The SFV vector system has also been developed for use as an anti-cancer agent. It has been shown that direct infection of tumours with rSFV particles can inhibit tumour growth in mice and induce regression in the p53-deleted human non-small cell lung carcinoma cell line H358a. Pathological analysis of rSFV treated tumours indicated that tumour growth inhibition was consistent with an initial induction of IE 0 2 04 6 7 apoptosis followed by a large depletion of tumour cells by oncotic necrosis [17], The apoptotic function is encoded by the non-structural region of the SFV genome [18], which is maintained in the vector. This allows for insertion of genes in the multicloning site of the vector, located in place of the structural region, which could modify or enhance pro-apoptotic or anti-tumour properties. One such study involved the insertion of the Bax gene to enhance the cytopathic and anti-tumour potential of the vector [19].

The present inventors have surprisingly demonstrated the potential of SFV as a CNS vector. Previous work by them has shown that i.n. infection of mice and rats with wild-type SFV leads to infection of the CNS by infection of neurons in the olfactory mucosa, followed by axonal transport to the olfactory bulb [20,21], The virus then multiplies in neurons and oligodendrocytes, and is transported to other areas of the CNS [4], The present inventors have shown that rSFV recombinant particles expressing enhanced green fluorescent protein (EGFP) infect the olfactory mucosa, but infecting RNA does not enter the olfactory bulb. However, transport of the EGFP protein to the olfactory bulb does occur and protein expression was detected in this area of the CNS.

This is surprising since most previous studies of SFV infection of the CNS have involved peripheral inoculation. The present study utilises intranasal inoculation, and although the olfactory mucosa is not part of the CNS, results in CNS infection and is a more rapid and direct route to the CNS. If wild-type virus is inoculated there is extensive CNS damage, and for virulent strains, the animals die. However, by using the vector to infect the olfactory mucosa, there is expression of protein from a cloned gene in the CNS and no CNS damage can be detected.

The need for an efficient drug delivery system for the CNS is apparent from the number of CNS diseases for which there is no adequate treatment. For example, Multiple Sclerosis (MS) is a chronic disease of the CNS which involves demyelination of nerve fibres. Myelin is the insulating substance which forms a sheath around nerve fibres (axons) and breakdown of myelin in discrete areas leads to loss of nerve function. One partially effective treatment for MS is the use of interferon-beta, which is one of a family of immunoregulators called cytokines. Many other cytokines are known which have a variety of effects on the immune system and, for example, interleukin-10 (IL-10) IE U 2 04 6 1 and transforming growth factor-beta (TGF-beta) have been postulated to have potential in MS treatment [22].

Two major problems in using such cytokines administered as protein molecules, as in the currently available treatment with interferon-beta, are their instability in the body and the probability that they need to act within the target organ, namely, the CNS. Delivery to the CNS is currently problematic since it involves injection directly into the CNS which is both hazardous and traumatic for the patient.

Previously known vectors for use in the CNS, such as herpes and adenovirus vectors, have to be administered by intracerebral injection. This is a highly invasive procedure which could cause additional damage to the CNS. With the present invention, the SFV vector can be administered intranasally. This in itself is surprising, since wild-type virus given by this route causes extensive CNS damage and with the SFV4 strain, infected mice usually die. With the SFV vector of the invention, there is no damage caused by extensive expression of protein in the olfactory bulb.

Object of the Invention It is an object of the invention to utilise a vector system that will penetrate the CNS and express cloned genes which it carries to the CNS thus providing non-invasive but efficient treatments for CNS disease.

It is a further object to use a vector system to produce a medicament for treatment of CNS disease, such as multiple sclerosis and other inflammatory and degenerative conditions or infections of the CNS, such as Alzheimer’s disease, Parkinson’s disease, spinal cord injury and the like. Another object is to provide an improved method of and improved compositions for the treatment of CNS disease.

Summary of the Invention According to the present invention there is provided use of an RNA vector system, comprising the Semliki Forest Virus genome and an exogenous RNA sequence in the preparation of a medicament for the treatment of CNS disease, the vector being IE 0 2 04 6 7 incapable of expressing at least one structural SFV protein. The vector is a suicide particle, and thus causes no damage to the CNS as it cannot enter the CNS.

The CNS diseases which may be treated by the invention include Multiple Sclerosis, herpes simplex virus infection, cytomegalovirus infections and any other disease treatable with cytokines and/or other therapeutic agents such as therapeutic genes, antisense sequences, ribozymes and the like.

Preferably, the SFV genome is derived from the SF4 strain of virus.

At least one SFV structural protein coding sequence of the vector may comprise a stop codon or deletion mutation such that the structural protein is not expressed.

Alternatively, the exogenous RNA sequence may replace a structural gene sequence. In an alternative embodiment, the exogenous gene may replace all of the structural genes and the structural gene(s) may be provided in trans by helper RNA vector(s). In a particularly preferred embodiment, the helper vector(s) has the packaging signal deleted to prevent encapsidation. Most preferably, two helper RNA vectors are provided, one encoding the capsid protein and one encoding the envelope proteins.

The cloned gene may be selected from cytokine-encoding genes, such as genes encoding interferon-beta (IFNbeta),, interleukin-10 (IL-10), interleukin-4 (IL-4) and transforming growth factor-beta (TGF-beta) and other pro- and anti-inflammatory cytokines and their inhibitors.

The invention also provides a pharmaceutical composition comprising a vector as defined above together with a pharmaceutically acceptable carrier or excipient. Preferably, the pharmaceutical composition is adapted for intra-nasal administration.

In another aspect, the invention provides a method of medical treatment comprising administering to a patient a pharmaceutically effective amount of a vector system as described above.

The invention will now be described in greater detail with reference to the following drawings in which:Figure 1. EFGP expression in the olfactory bulb of mice infected i.n. with rSFV-EGFP particles. Particles were prepared and titrated as previously described [8,17], (A) Section taken from the olfactory bulb at 5 dpi, visualised using both DAPI and GFP IE 0 2 04 6 1 filters. x200.

(B) Section labelled with anti-MAPI antibody, axons labelled in red, x400.

Figure 2.

Detection and distribution of rSFV-EGFP RNA in mice. Agarose gel electrophoresis was performed using RT-PCR products from total RNA extracted from the nasal passages of infected Balb/c mice. Lane 1. 16 hpi. Lane 2. 1 dpi. Lane 3. 3 dpi. Lane 4. 5 dpi. Lane 5. 7 dpi. Lane 6. 10 dpi. (A) Representative gel of the 482 bp region of the EGFP DNA amplified with specific primers. (B) Representative gel of the 520 bp portion of the nsP3 region of SFV4 amplified using specific primers. In each case the outermost lanes show molecular weight markers.

Figure 3.

Histological staining of olfactory bulb sections from i.n. infected mice. (A) rSFV-EGFP, at 7 dpi, x200. No abnormalities are apparent. (B) Wild-type SFV4 virus at 5 dpi. Note necrosis of neurons and oedema around glomeruli area, x200.

Figure 4.

Immunocytochemistry using the Vectastain Elite ABC Kit on BHK-21 cells counterstained with methyl green.

(A) Infected cells showing positive for IL-10 (brown). .(B) Mock-infected cells showing no presence of IL-10.

Figure 5.

(A) The structure of the viral RNA, which has the 5’ terminal cap and 3’ poly A tract typical of mammalian mRNA. The genome is divided into non-structural and structural coding regions, separated by a short untranslated region. Untranslated sequences are also found at the ends of the genome. The structural protein sequence is amplified by the subgenomic 26 S mRNA species. p62 is the precursor to the E2 and E3 envelope proteins which is cleaved during virus maturation.

Ic Ο 2 04 6 7 (B) The SFV infectious clone pSP6-SFV4 showing some of the restriction sites which can be used for genetic manipulation. The SFV genome (11445 nt) has been cloned under the control of the SP6 promoter; to produce infectious virus, the plasmid is linearised with Spel, then transcribed with SP6 polymerase to produce infectious RNA, which is then transfected into BHK cells by electroporation.

Figure 6.

The SFV1 vector system. The organisation of the wild type SFV genome is shown at the top. The right-handed arrow denotes the promoter that is recognised by the viral io replicase for transcription of the subgenomic RNA encoding the structural proteins.

MCS indicates a multiple cloning site where heterologous sequences can be inserted.

The helper encodes all structural proteins of SFV but has nearly the complete replicase region deleted, including the packaging signal (hatched/black box in the replicase region). Black boxes at the ends denote sequences used by the replicase for amplification of the RNA.

Detailed Description of the Invention Materials and Methods Production of recombinant SFV particles expressing EGFP.

The SFV vector system, including a vector expressing EGFP, was obtained from Professor P Liljestrom, Microbiology and Tumorbiology Center, Karolinska Institute, Stockholm, Sweden. The production of rSFV RNA and generation of infectious vector particles have been previously described [8,17]. Briefly, BHK-21 cells were cotransfected with rSFV RNA (with the reporter gene EGFP) and two helper RNAs, one coding for the capsid protein, the other for the envelope protein. RNAs were transcribed in vitro from appropriate plasmids. After 24 h or 36h incubation, infectious virus particles containing the recombinant gene were harvested. Virus particles were concentrated by ultracentrifugation through a 20% (w/v) sucrose cushion (. 100,000g for 1.5 h at 4°C) and resuspended in TNE Buffer (50 mM Tris-HCL, ph 7.4,100 mM NaCl, 0.1 mM EDTA). Titration was performed by direct (rSFV-EGFP) epifluorescence.

IE Ο 2 04 6 Ί Animals and virus A working stock of the virulent SFV4 strain of SFV was prepared from a plaquepurified seed stock, purified through sucrose gradients and titrated on BHK-21 cells as previously described [4], Balb/c mice aged 6-8 weeks were used for animal experiments. For virus infection experiments, mice were infected i.n. with 107 plaqueforming units of virus in 20 μΐ of phosphate buffered saline (PBS), placed on the end of each nostril so that the droplet was inhaled. For particle experiments, mice were inoculated i.n. with 107 infectious units (IU) of rSFV-EGFP in 20 μΐ of TNE buffer.

EGFP Expression At designated times after i.n. infection with rSFV-EGFP particles, mice were anaesthetised with halothane and perfused via the left ventricle with 4% paraformaldehyde (PFA) for 5 min. The separated olfactory bulbs and remaining brains were removed intact and snap-frozen in liquid nitrogen-cooled isopentane [22], Frozen tissues were sectioned to a thickness of 7μιη using a cryostat. Sections were then fixed with 4% PFA, counterstained using the nuclear stain DAPI (Sigma, UK), and visualised by fluorescent microscopy using filters at 488 and 400 nm for EGFP and DAPI detection, respectively.

Cloning of pSFVl-cytokine constructs TGF-beta Total RNA was isolated from the brains of SFV-infected Balb/c mice 3 to 10 days following ip inoculation and used to synthesize cDNA for TGF-beta by RT-PCR, as described previously (16), using primers that recognise the mature form of murine TGF-beta. The resulting PCR product was purified and ‘polished’ in order to remove Aoverhangs using cloned Pfu polymerase before been ligated into the pPCR-Script vector according to the manufacturers instructions (Stratagene). Recombinant clones were screened and putative positives confirmed by sequence analysis. DNA from positive IE U2 046 Ί clones was linearised with Kpn I, ‘cleaned’ using the QIAquick nucleotide removal kit (Qiagen) and 100 to 200ng of DNA template was amplified in a standard proofreading PCR reaction containing 3.75U Pfu Turbo Polymerase (Stratagene), lx cloned Pfu Polymerase buffer, 25 mM each of dATP, dCTP, dGTP, and dTTP and 0.25 μΜ each of the following forward and reverse primers. A primer designed to incorporate the native SFV ribosome binding sequence (underlined) and the ATG start codon (bold) in TGF beta clones was as follows: TGFbetaF: 5 '-TGTGGATCCTATAGCACCATGGCC CTGGATACCAACTAT-3'. TGFbetaR: 5'- CTGGGATCCGTCTCAGCTGCACTTGCAGGA-3'.

The reaction was performed in an automated DNA thermal cycler using the following temperature profile: 2min at 95°C, 3min at 80°C followed by 30 cycles of incubations at 95°C for 30sec, 51°C for 30sec and 72°C for 5min with a final extension of 15min at 72°C. The resulting PCR products were analysed by agarose gel electrophoresis and gel extracted using the Qiagen MinElute Gel extraction kit (Qiagen) before been ligated into the Smal site of the pSFV 1 expression vector. Recombinant clones in the correct orientation were confirmed by restriction digest (Ban II and Ava I) and sequence analysis.

IL-10, IL-4 and IFN-beta Total RNA was extracted from the spleens of SFV-infected mice using Genosys RNA isolator as described previously (16) 5 to 7 days following ip infection and cDNA for IL-10, IL-4 and IFN beta amplified by RT-PCR as described above. The primers used corresponded to the ends of the murine cytokine genes and included appropriate sequences to facilitate cloning into the pSFVl vector using standard cloning techniques.

Positive clones were confirmed by restriction digest using the appropriate enzymes and sequence verification confirmed by sequence analysis.

Production and in vitro expression of cytokine encoding recombinant SFV particles. pSFVl-capsid and spike DNA and pSFVl-TGF beta, pSFVl-IL-10, pSFVl-IL-4 DNA was linearised using SpeL and cleaned using QIAquick nucleotide kit (Qiagen) before being transcribed as described above. After 24h (TGF beta) or 33h (IL-10 and IL-4) incubation, infectious virus particles expressing the individual cytokine genes IE 0 2 04 6 / were harvested. Virus particles were concentrated by ultracentrifugation as described above, titrated on BHK cells and cell staining visualised by immunocytochemistry using rat anti-mouse TGF-βΙ, rat anti-mouse IL-10 and rat anti-mouse IL-4 respectively (5mg/ml, PharMingen).

Expression of the individual cytokines was confirmed in vitro by infection of BHK-21 cells with rSFV-TGF beta, rSFV-ILlO or rSFV-IL4. Cells were incubated overnight at 37°C, 5% CO2, fixed in 5% formalin, 15min, and post-fixed in acetic acid'.methanol at -20°C for 5 mins. Endogenous peroxidase activity was blocked with 3% H2O2 in PBS for lOmin. Cells were incubated sequentially with primary antibody (PharMingen), 37°C, for lh, biotinylated secondary antibody, 30min, and avidin-biotinperoxidase complex (Elite ABC standard kit, Vector). Between each reagent, cells were washed twice with PBS containing 0.05% Tween 20 and washed twice in PBS.

The immunostaining was visualized using diaminobenzidine chromogen (Sigma) and the cells counterstained with methyl green.

Immunocytochemistry At 4 days post infection (dpi) with rSFV-EGFP, mice were anaesthetised with halothane and perfused via the left ventricle with 4% paraformaldehyde (PFA) for 5 min. The olfactory bulb area of the brain was removed intact and snap-frozen in liquid nitrogen-cooled isopentane [22]. Frozen tissues were sectioned to a thickness of 7pm using a cryostat. Sections were then fixed with 4% PFA. Oligodendrocytes were labelled with an anti-CNPase antibody (Sigma C5922), [23] and an anti-MAPI antibody (Sigma M4278) [24] was used for selective labelling of neurons with stronger staining of axons than dendrites. Briefly, sections were washed in PBS, blocked with 5% normal rabbit serum (Sigma) in PBS for 1.5 hours at 37°C. Sections were then incubated overnight at 4°C with primary antibody diluted in PBS (anti CNPase 1:10, and anti MAPI 1:25). Following incubation, sections were washed in PBS Tween (0.05% Tween 80, Merck) followed by PBS. Sections were incubated with rabbit anti-mouse immunoglobulins TRITC conjugate (Dako R0270), diluted 1:20 for 1 hour at 37°C, washed in PBS Tween followed by PBS. Sections were visualised by fluorescent microscopy using filters at 488 and 546 nm for EGFP and TRITC detection, respectively.

IE 0 2 04 6 Ί RT-PCR assays Total RNA was extracted from the nasal passages, olfactory bulb, and remaining areas of the brain of mice following rSFV-EGFP particle inoculation. First strand cDNA was synthesised from lOOpg mouse RNA tissue sample using oligo (dT) primers and AMV reverse transcriptase (RT) in a 20μ1 reaction as described previoulsy [16]. Negative control samples contained cDNA from RNA extracted from PBS mockinfected mouse tissues, whereas positive control cDNA for nsP3 and EGFP amplification was prepared from tissue of mice injected i.n. with SFV4 (wild-type virus), and i.m. with rSFV-EGFP respectively. Internal controls for the RT-PCR procedure included samples with either no RNA or no RT added. Oligonucleotide primers were designed from a non-structural region of SFV (nsP3) and the coding region of the EGFP reporter molecule (Table 1). Oligonucleotide primers that produce 540 and 508 bp products of the mouse β-actin and GAPDH proteins, respectively, (Table 1) were used to demonstrate the integrity of the DNA. The amplified products were analysed on 1 % agarose gels stained with ethidium bromide.

Histological Analysis Brains were sampled at 3, 5, 7, and 14 dpi, following rSFV-EGFP particle administration. SFV4 -infected mice were sampled at 3 dpi and 5 dpi. Control PBS mock-infected mice were sampled at 5 dpi and 7 dpi. Mice were anaesthetised with halo thane and perfused via the left ventricle with 10% formol saline for 5 min. Perfused mice were left overnight in fixative at 4°C . Tissues were routinely embedded in paraffin and 4μηι sections were stained with haematoxylin and eosin (H&E).

Results Location and persistence of EGFP expression in mice The duration and location of EGFP expression was determined by fluorescent microscopy. Cryosections of brain samples following inoculation with rSFV-EGFP particles exhibited bright green staining indicative of protein expression (Fig 1). Expression was first detected at 16 hours post infection, and remained visible until day IE Ο 2 04 6 7 14. No protein could be seen at day 21. EGFP expression was limited to the olfactory bulb area of the brain (Table 2). By overlaying images of protein expression using the GFP filter with that produced using the DAPI filter, we were able to determine the location of expression in the olfactory bulbs, as well as the type of cells expressing EGFP (Fig 1 A,B). Cells surrounding the glomeruli areas of the olfactory bulb expressed EGFP, and expression was confined to this area. This area contains the incoming axons from receptor cells in the olfactory epithelium.

Immunocytochemical analyses To confirm that protein expression was restricted to incoming axons in the olfactory bulbs, brain sections were labelled for axons, as well as oligodendrocytes. Within the olfactory bulb area, no oligodendrocytes were found to be expressing EGFP. Axons of cells surrounding the glomeruli areas of the olfactory bulb that were expressing EGFP were also positively labelled with the MAPI antibody (Fig. IB).

Persistence and location of rSFV-EGFP RNA in mice RNA from the nasal passages, olfactory bulb, and brains from animals infected with rSFV-EGFP was isolated at specified times after infection. RNA was also isolated from control animals infected with SFV4 (wild-type SFV) and mock-infected with PBS.

No bands were visible at any time for RNA extracted from the olfactory bulbs or brains of rSFV-EGFP infected mice, with either set of primers. Both RNAs were detected in the nasal passages of infected mice as early as 16 hours post infection (hpi). Table 3 shows the results obtained for EGFP primers; similar results were obtained for nsP3 primers (data not shown). Maximum expression was found at 24 hpi. A week band could also be seen at 7 dpi, while no bands were detected at 10 dpi (Fig 2). RT-PCR performed on tissue taken from PBS mock-infected mice was consistently negative, while those from SFV4 infected mice were positive for both brain and nasal passages only when the nsP3 primer was used (data not shown).

IE 02 04 6 7 Pathological changes in mice following administration of particles and SFV4 In order to detect possible pathological changes associated with particle administration, mice were either inoculated with rSFV-EGFP, infected with SFV4 or mock-infected with PBS. No brain lesions were detected in any rSFV-EGFP infected mice. Brain sections of control mice, infected with SFV4 and sampled at day 3, displayed perivascular cuffing and mild neuronal degeneration. Mice sampled at day 5 showed massive neuronal necrosis (Fig 3). Mock-infected mice showed no damage.

Cytokine Expression Cytokine expression was visualized in vitro in BHK-21 cells. Cells positive for IL-10 expression are demonstrated in Figure 4. Similar results were obtained when using IL-4 and TGF-β (data not shown).

Discussion Semliki Forest virus is a neurotropic virus that has been used extensively as a model to study viral neuropathogenesis [2,6]. It infects man [26] as well as small rodents, so the vector derived from it may be useful for the treatment and prevention of human disease. So far, the SFV vector has been mainly developed for recombinant vaccine construction and tumour treatment [6,10,16,17,19]. In the current study, we tested the feasibility of utilising the SFV vector system for expression in the central nervous system.

We were able to demonstrate that recombinant particles administered intranasally, a non-invasive method, infect the olfactory mucosa only and the infection does not proceed beyond this point. This is different to the situation with the wild-type virus, where intranasal infection leads to widespread infection of the CNS. Unlike the wild-type virus, which causes extensive CNS damage [4,20,21], histopathological analysis showed that recombinant SFV particles caused no CNS damage.

Protein encoded by the foreign gene inserted into the vector (in this case the EGFP reporter gene) was expressed in the incoming axons from the receptor cells in the olfactory epithelium that terminate in the glomeruli area of the olfactory bulbs. Within IE 02 04 6 7 this area, synaptic contacts between the incoming axons, and dendrites from cells within the olfactory bulb, are established. Axons from these cells extend to the anterior olfactory nucleus area of the brain, and onwards to other areas of the CNS. Therefore it is possible that administration of SFV recombinant particles could introduce a therapeutic protein into the brain and by-pass the blood-brain barrier. This vector system could therefore be useful for the administration of proteins, or genes carried by vectors, which cannot traverse the blood-brain barrier when administered peripherally.

It may be an alternative to invasive intracerebral administration.

Protein expression levels in the olfactory bulb were high, especially during the first 48 hours following inoculation. Protein expression was detected in the same locations within the bulb throughout the time course of the experiment. This finding, together with the fact that RNA transcription took place only in the nasal passage of infected animals, suggests that duration of expression is dependent on the half-life of the protein used, and the viral vector is not travelling any further into the brain. The findings also suggest that RNA replication of the viral vector is taking place in the cells of the olfactory mucosa, which have the capability to regenerate themselves if any damage is caused, as is the case in any viral infection of the upper respiratory tract.

Viral RNA persisted only for 7 days in infected mice following inoculation, which was consistent with previous findings by Morris-Downes et al [16], as opposed to dispersing in the host and persisting for many weeks, which is the case for DNA vectors [25].

It is important to note the EGFP expression is localized and confined to cells which are actively expressing the protein. This is very different from the type of expression achieved with cytokines, which are very small, secreted molecules that unlike EGFP would have the ability to migrate among the tissues and initiate a cascade of events.

While not suitable for the permanent replacement of defective genes in its current form, this system may be suitable for situations where short-term expression of a therapeutic protein is required, for example treatment of disease episodes such as occur in infections of the CNS or relapses of multiple sclerosis. The use of this vector system represents a clear advantage over other methods. The particles produced are stable, safe and capable of reaching titres above 1010 infectious units per ml. Every mouse given the particles remained healthy throughout the period of experimentation. This system may IE Ο 2 04 6 Ί therefore have the potential to treat diseases of the CNS by administration of a vector encoding a therapeutic gene, by the non-invasive and convenient intranasal route.

IE02 046 Z References 1. Atkins GJ, SheahanBJ, Liljestrom P. (1996) Manipulation ofthe Semliki Forest virus genome and its potential for vaccine construction. Molecular biotechnology :33-38 2. Atkins GJ, Sheahan BJ, Dimmock NJ. (1985) Semliki Forest virus infection of mice: a model for genetic and molecular analysis of viral pathogenicity. Journal of General Virology 66:395-408 3. Atkins GJ, Balluz IM, Glasgow GM, Mabruk MJ, Natale VA, Smyth JM, Sheahan BJ. (1994) Analysis of the molecular basis of neuropathogenesis of RNA viruses in experimental animals: relevance for human disease? Neuropathology and applied neurobiology 20:91-102 4. Balluz IM, Glasgow GM, Killen HM, Mabruk MJ, Sheahan BJ, Atkins GJ. (1993) Virulent and avirulent strains of Semliki Forest virus show similar cell tropism for the murine central nervous system but differ in the severity and rate of induction of cytolytic damage. Neuropathology and Applied Neurobiology 19:233-239 . Berglund P, Sjoberg M, Garoff H, Atkins GJ, Sheahan BJ, Liljestrom P. (1993) Semliki Forest virus expression system: production of conditionally infectious recombinant particles. Biotechnology 11:916-920 6. Atkins GJ, Sheahan BJ, Liljestrom P. (1999) The molecular pathogenesis of Semliki Forest virus: a model virus made useful? Journal of General Virology 80:2287-2297 7. Liljestrom P, Garoff H. (1991) A new generation of animal cell expression vectors based on the Semliki Forest virus replicons. Bio/technology 9:1356-61 8. Smerdou C, Liljestrom P. (1999) Two-helper RNA system for production of recombinant Semliki Forest Virus particles. Journal of Virology 73:1092-1098 9. Fleeton MN, Liljestrom P, Sheahan BJ, Atkins GJ. (2000) Recombinant Semliki Forest virus particles expressing louping ill virus antigens induce a better protective response than plasmid-based DNA vaccines or an inactivated whole particle vaccine. Journal of General Virology 81:749-758 IE 0 2 04 6 7 . Berglund P, Fleeton MN, Smerdou C, Tubulekas I, Sheahan BJ, Atkins GJ, Liljestrom P. (1998) Vaccination with recombinant suicidal DNA/RNA. In Developments and New Applications in Animal Cell Technology, pp. 583-590. Edited by O. Merten, P. Perrin & B. Griffiths. Dordrecht: Kluwer Academy Publishers. 11. Zhou X, Berglund P, Rhodes G, Parker SE, Jondal M, Liljestrom P. (1994) Selfreplicating Semliki Forest virus RNA as recombinant vaccine. Vaccine 12:15101514. 12. Tubulekas I, Berglund P, Fleeton M, Liljestrom P. (1997) Alphavirus expression vectors and their use as recombinant vaccines: a minireview. Gene 190:191-195. 13. Berglund P, Quesada-Rolander M, Putkonen P, Biberfeld G, Thorstensson R, Liljestrom P. (1997) Outcome of immunization of cynomolgus monkeys with recombinant Semliki Forest virus encoding human immunodeficiency virus type 1 envelope protein and challenge with a high dose of SHIV-4 virus. AIDS research and human retroviruses 13:1487-95. 14. Berglund P, Fleeton MN, Smerdou C, Liljestrom P. (1999) Immunization with recombinant Semliki Forest virus induces protection against influenza challenge in mice. Vaccine 17:497-507.

. Fleeton MN, Sheahan BJ, Gould EA, Atkins GJ, Liljestrom P. (1999) Recombinant Semliki Forest virus particles encoding the prME or NS1 proteins of louping ill virus protect mice from lethal challenge Journal of General Virology 80:1189-98. 16. Morris-Downes MM, Phenix KV, Smyth J, Sheahan BJ, Lileqvist S, Mooney DA, Liljestrom P, Todd D, Atkins GJ. (2001) Semliki Forest virus-based vaccines: persistence, distribution and pathological analysis in two animal systems. Vaccine 19:1978-1988. 17. Murphy AM, Morris-Downes MM, Sheahan BJ, Atkins GJ. (2000) Inhibition of human lung carcinoma cell growth by apoptosis induction using Semliki Forest virus recombinant particles. Gene Therapy 7:1477-1482. 18. Glasgow GM, McGee MM, Tarbatt CJ, Mooney DA, Sheahan BJ, Atkins GJ. (1998) The Semliki Forest virus vector induces /?53-independent apoptosis. Journal IE Ο 2 04 6 7 of General Virology 79: 2405-10. 19. Murphy AM, Sheahan BJ, Atkins GJ. (2001) Induction of apoptosis in Bcl-2expressing rat prostate cancer cells using the Semliki Forest virus vector.

International Journal of Cancer 94:572-578 . Sammin DJ, Butler D, Atkins GJ, Sheahan BJ. (1999) Cell death mechanisms in the olfactory bulb of rats infected intranasally with Semliki Forest virus. Neuropathology and Applied Neurobiology 25:236-243. 21. Sheahan BJ, Ibrahim MA, Atkins GJ. (1996) Demyelination of olfactory pathways in mice following intranasal infection with the avirulent A7 strain of Semliki Forest virus. European Journal of Veterinary Pathology 2:117-125. 22. Morris MM, Dyson H, Baker D, Harbige LS, Fazakerley JK, Amor S. (1997) Characterization of the cellular and cytokine response in the central nervous system following Semliki Forest virus infection. Journal of Neuroimmunology 74:185-97. 23. Sprinkle TJ. (1989) 2’,3’-cyclic nucleotide 3'-phosphodiesterase, an oligodendrocyte-Schwann cell and myelin-associated enzyme of the nervous system. Critical reviews in neurobiology 43:235-301 24. Huber G, Matus A. (1984) Differences in the cellular distributions of two microtubule-associated proteins, MAPI and MAP2, in rat brain. The Journal of neuroscience 4:151-160.

. Donnelly JJ, Ulmer JB, Shiver JW, Liu MA. (1997) DNA vaccines. Annual review of immunology\5'. 617-648. 26. Mathiot CC, Grimaud G, Garry P, Bouquety JC, Mada A, Daguisy AM, Georges AJ. (1990). An outbreak of human Semliki Forest virus infection in Central African Republic. American Journal of Tropical Medicine and Hygiene 42:38625 393. 27. Lundstrom K, Schweitzer C, Rotmann D, Hermann D, Schneider EM, Ehrengruber MU. (2001) Semliki Forest virus vectors: efficient vehicles for in vitro and in vivo gene delivery. FEBS Lett 2001 Aug 31;504(3):99-103.

IE0 204 6 7 Table 1 Primer sequences for RT-PCR Protein Sequence Product size (bp) Mouse β-actin Mouse GAPDH SFV-nsp3 EGFP ’-GTGGGCCGCTCTAGGCACCAA-3' 540 ’-CTCTTTGATGTCACGCACGCTTTC-3' '-ACCACCATGGAGAAGGCTG-3' 508 '-CTCAGTGTAGCCCAGGATGC-3' ’-GCGGAATTCCTCATCTTTTCCCCTCCCGA-3' 520 '-CGCGAATTCATCGACCTCGTGCTCGTCAA-3' -CTGGACGGCGACGTAAACGGCCAC-3' 482 -AGCTGCACGCTGCCGTCCTCGATG-3' IE Ο 2 04 6 7 Table 2 Duration and location of EGFP expression following rSFV-EGFP infection of Balb/c mice Olfactory Time Bulbs Brain 16 hpi 0/4a 0/4 1dpi 4/4 0/4 2 dpi 3/5 0/5 3 dpi 4/4 0/4 4 dpi 4/5 0/5 10 5 dpi 4/5 0/5 6 dpi 5/5 0/5 8 dpi 2/5 0/5 10 dpi 2/3 0/3 14 dpi 1/3 0/3 15 21 dpi 0/3 0/3 Number positive/number tested IE Ο 2 04 6 7 Table 3 Duration and location of RNA expression following rSFV-EGFP infection in Balb/c mice8 Sample 16 hrs Time post inoculation 10 days 1 day 3 days 5 days 7 days Nasal Passage 4/4b 4/4 4/4 4/4 4/4 0/4 Olfactory Bulb 0/4 0/4 0/4 0/3 0/3 0/3 Brain 0/4 0/4 0/4 0/3 0/3 0/3 ’Detection of RNA using the EGFP primer pair. b Number positive/number tested IE 02 04 6 7

Claims (12)

Claims

1. Use of an RNA vector system comprising the Semliki Forest Virus genome and an exogenous RNA sequence in the preparation of a medicament for the treatment of CNS disease, the vector being incapable of expressing at least one structural SFV protein.

2. Use as claimed in claim 1 wherein the SFV genome is derived from the SFV4 strain of virus.

3. Use as claimed in any preceding claim wherein at least one SFV structural protein encoding sequence of the vector comprises a stop codon or deletion mutation such that at least one structural SFV protein is not expressed.

4. Use as claimed in claim 1 or 2 wherein the exogenous RNA sequence replaces a structural gene sequence.

5. Use as claimed in claim 1 or 2 wherein the exogenous gene replaces all of the structural genes and the structural gene(s) is provided in trans by helper RNA vector(s).

6. Use as claimed in claim 5 wherein the helper vector(s) has the packaging signal deleted to prevent encapsidation.

7. Use as claimed in claim 5 or 6 wherein two helper RNA vectors are provided, one encoding the capsid protein and one encoding the envelope proteins.

8. Use as claimed in any preceding claim wherein the cloned gene is selected from cytokine-encoding genes, such as genes encoding interferon-beta, interleukin-10, interleukin-4, TGF-beta, other pro- and anti-inflammatory cytokines and inhibitors thereof. IE02 04 6 7

9. Use as claimed in any preceding claim wherein the medicament is formulated for intranasal administration.

10. A pharmaceutical composition comprising a vector system comprising the Semliki Forest Virus genome and an exogenous RNA sequence the vector being incapable of expressing at least one structural SFV protein, together with a pharmaceutically acceptable carrier or excipient.

11. A pharmaceutical composition as claimed in claim 10 adapted for intranasal administration.

12. A method of medical treatment comprising administering to the patient a pharmaceutically effective amount of a vector system comprising the Semliki Forest Virus genome and an exogenous RNA sequence, the vector being incapable of expressing at least one structural SFV protein.