WO2006074963A2

WO2006074963A2 - Lentiviral gene transfert system for ribozyme mediated rna-repair

Info

Publication number: WO2006074963A2
Application number: PCT/EP2006/000358
Authority: WO
Inventors: Michael Foley
Original assignee: GENEERING GmbH
Current assignee: GENEERING GmbH
Priority date: 2005-01-17
Filing date: 2006-01-17
Publication date: 2006-07-20
Anticipated expiration: 2007-07-17
Also published as: DE102005003207A1; WO2006074963A3

Abstract

The invention relates to a safer and efficient recombinant gene transfer system for ribozyme mediated repair of a target RNAin non-dividing and dividing cells, based on the Visna lentivirus (VLV) comprising: (a) a first plasmid comprising a sequence encoding the VLV Gag, (b) a second plasmid as transfer plasmid comprising a ribozyme cassette th the catalytic domain of a self-splizing ribozyme (class I intron) and a sequence encoding for the accurate counterpart of the target RNA to be repaired, (c) a third plasmid comprising a sequence encoding VLV Pol, (d) a fourth plasmid comprising sequences encoding VLV Vif, Rev and Tat, (e) a fifth plasmid comprising a sequence encoding an envelope protein Env, which is not derived from VLV, wherein the plasmids a and c to e do not encode for other VLV proteins and do not contain a competent packaging signal.

Description

Lentiviral Gene Transfer system for ribozyme mediated RNA-repair

The invention relates to viruses as vectors useful in gene delivery and gene therapy, and more specifically to lentiviral vectors useful in ribozyme mediated RNA-repair in non-dividing and dividing cells.

The capacities to introduce a particular foreign or native gene sequence into a mammalian cell and to control the expression of that gene are of substantial value in the fields of medical and biological research. Such capacities provide a means for studying gene regulation, and for designing a therapeutic basis for the treatment of disease.

Gene therapy typically aims to complement a mutant allele of a gene with a functional one. Although the technology is still in its infancy, it has been used with some success. However, most gene therapy approaches known only add a functional allele and do not replace the defective mutant allele of the gene. Thus, the mutant allele remains intact, which is not a problem in case the mutant allele is inactive. However, in case of the mutant allele having a changed activity those approaches will fail, because they just add a desired activity alongside a changed activity. Another disadvantage of the conventional gene therapy approaches is that the level of expression of the inserted gene depends heavily on the site the gene is inserted into the genome of the cell. Thus, expression level normally divers from the natural expression level of the gene.

Among the viral vectors used for gene-transfer retrovirus-based vectors are particularly favoured because they result in a stable integration of transferred genes into the genome of mammalian cells. Lentiviruses are a subgroup of retroviruses that are capable of infecting non-dividing cells. Several lentiviral vector system based on the human immunodeficiency virus (HIV) or the related simian immunodeficiency virus (SIV) are reported (US 5,665,577, US 5,747,324, US5861282, US 5,919,458, US 5,981 ,276, US 6,013,516, US 6,555,342, US 6,627,442, US 6,669,936, US 6,712,612, US 6,790,641, US 6,790,657).

However, the use of a HIV- or SIV-based system for gene transfer in humans raises serious safety concerns, due to the possibility of recombination by the vector into a virulent and disease-causing form. This holds less true for vectors, which are based on related viruses as the bovine immunodeficiency virus (BIV - WO 03/066810), equine infectious anaemia virus (EIAV - US 6,277,633, US 6,312,683) or feline immunodeficiency virus (FIV - US 6,555,107). The extensive homology between the accessory and structural protein genes and the transduction vector used in those approaches increases the likelihood of recombinations. US 6,479,281 relates to a lentiviral virus vectors (derived from HIV, SIV, FIV, visna virus or EIAV) that possess a truncated matrix protein (claim 17b) replaced with a heterologous myristylation anchor and a truncated envelope protein (claim 17c). As the vector is a replication-competent retrovirus the risk of spread and recombination is significant. No examples of the transduction of a heterologous gene were shown.

Visna virus (Virus Code. 61.0.6.4.002) is a lentivirus, which infects sheep and is not pathogen to humans. The wild type visna virus has a dimeric RNA genome (single-stranded, positive polarity) that is packaged into a spherical enveloped virion containing a nucleoprotein core. Replication of the wild type visna virus genome occurs via reverse transcription and integration into the host cell genome. The virus genome consist of one RNA (9202 bp, Genbank Accession Number: NC_001452) and contains three major genes encoding the Gag, Pol and Env polyproteins and long terminal repeats (LTR) at each end of the integrated viral genome. The LTR is about 600nt long, of which the U3 region is 450, the R sequence 100 and the U5 region some 70 nt long. The gag gene of wild type visna virus encodes the polyprotein Gag, which is cleaved by the viral protease into the structural protein of the matrix p17, the major capsid protein p24 and the nucleocapsid proteins p7 to p11. The pol gene encodes the polyprotein Pol, which is cleaved by the protease into the following enzymes: the protease itself, the reverse transcriptase (RT), the RNase H (RH), the dUTPase and the integrase.

The natural visna virus comprises additional genes encoding the accessory proteins Vif, Tat, and Rev which are involved in regulation of synthesis and processing virus RNA and other replicative functions. The Viral infectivity factor (Vif ) forms part of the pre-integration complex with the integrase that packages the DNA product formed from the viral RNA by the Pol- derived RT in a form that allows it to pass the nuclear membrane and thus allows the stable integration of the viral- into the host cell-genome in the presence of an intact nuclear membrane as in a resting or terminally differentiated cell. The transactivator of transcription (Tat) protein is a pleiotropic factor that induces a broad range of biological effects in numerous cell types. Tat is a powerful transactivator of gene expression, which acts by both inducing chromatin remodelling and by recruiting elongation-competent transcriptional complexes onto the viral LTR promoter. The visna virus Tat protein is required for efficient viral transcription from the visna virus long terminal repeat (LTR). AP-1 sites within the visna virus LTR, which can be bound by the cellular transcription factors Fos and Jun, are also necessary for Tat-mediated transcriptional activation. The wild type genome of visna virus also contains several cis-acting sequences, promoter elements such as the Tat response-element (TRE) the that control transcriptional initiation of the integrated provirus; a packaging sequence (Ψ); a Rev-response element (RRE) responsible for the export of the virion mRNA from the host cell nucleus.

The disadvantages of most retroviruses, including the wild type visna virus, are a limited cell tropism, low viral titer and the risk of the generation of replication-competent virus during packaging.

Another problem with the traditional approaches to gene therapy is that when a healthy gene is inserted to replace a defective version, it is inserted at random into the hosts genome. This can cause problems because the gene loses much of its surrounding DNA, which contains regulatory information, and the replacement gene can also interfere with its new neighbours.

A different approach to gene therapy is antisense therapy, which aims to silence a mutant allele of a gene by inserting of complementary nucleic acids sequences.

Ribozymes are RNAs that have enzymatic activity. Different ribozymes are known, as the hammerhead ribozyme, the hairpin ribozyme, the Tetrahymena group I intron, RNase P, and the hepatitis delta virus ribozyme. They catalyse self-cleavage (intramolecular or "in-cis" catalysis) or cleavage of external substrates (intermolecular or "in-trans" catalysis). Ribozymes like the hammerhead ribozyme and the hairpin ribozyme are already used in lentiviral vectors for gene silencing approaches (e.g. WO9710334, WO0032765). In the gene silencing approaches ribozymes are used and applied using many of the same principles as antisense nucleic acids. Also ribozymes rely on complementary binding to nucleic acids for target recognition. In contract to antisense nucleic acids they not only bind to the target messenger RNA, but also silence the gene by cleaving the RNA.

Group I intron ribozyme mediated mechanism of ribozyme-mediated RNA has long been stated to be a promising tool for gene therapy (reviewed by Long MB et al. 2003 J. Clin. Invest. 112: 312 - 318). However, the use of class I intron ribozyme-mediated RNA repair for gene therapy has up to now been hampered by the limited efficiency of the transfer of the gene encoding the ribozyme to human cells, the limited intracellular activity of the ribozyme in the cytoplasm and the short effective life of the ribozyme when transiently transfected.

To date there is no safe and efficient gene transfer system for ribozyme mediated RNA-repair known, which can be used for a broad range of dividing and non-dividing cells. As the safety of gene transfer by means of vector packaging constructs is of paramount concern, there is stiil a need for a safer and efficient lentiviral vector systems capable of mediating gene transfer into a broad range of dividing and non-dividing cells.

One object of the present invention is to provide a safer and efficient gene transfer system for ribozyme mediated RNA-repair, which can be used in a broad range of dividing and non- dividing cells.

This object is solved by the present invention, which provides a recombinant lentiviral gene transfer system for ribozyme-mediated repair of a mutated target RNA comprising: a. a first plasmid comprising an nucleic acid sequence encoding the polyprotein Visna lentivirus (VLV) Gag, wherein said plasmid does not encode for other VLV proteins and does not contain a competent packaging signal, b. a second plasmid as transfer plasmid comprising:

(i) VLV nucleic acid sequence comprising cis-acting sequence elements required for reverse transcription of the plasmid genome, (ii) a competent packaging signal and

(iii) a ribozyme cassette comprising the catalytic domain of a self-splizing intron ribozyme (class I intron) sequence and

(iv) a heterologous gene or a part of a gene encoding for the accurate (non- mutated) counterpart of the target RNA to be repaired or a cloning site wherein the heterologous gene or a part of a gene encoding for the accurate counterpart of the target RNA to be repaired may be inserted, whereas the second plasmid does not encode for any functional protein of VLV; c. a third plasmid comprising an nucleic acid sequence encoding the VLV polyprotein VLV Pol, wherein said plasmid does not encode for other VLV proteins and does not contain a competent packaging signal, d. a fourth plasmid comprising nucleic acid sequences encoding the VLV proteins Vif, Rev and Tat, wherein said plasmid does not encode for other VLV proteins of VLV and does not contain a competent packaging signal, e. a fifth plasmid comprising a nucleic acid sequence encoding a viral envelope protein Env, which is not derived from visna virus, and wherein said plasmid does not encode for VLV proteins of VLV and does not contain a competent packaging signal.

"Target RNA to be repaired" means any target RNA to be modified by splicing. "Accurate counterpart of the target RNA to be repaired" means any target RNA that is spliced onto the target RNA to be repaired. For Gene Therapy this normally means that a mutated part of a gene is replaced with the functional one. However the recombinant lentiviral gene transfer system of the invention can also used to link a marker gene to a functional gene or to replace a part of a functional gene with a marker gene. To generate humanised animal cells or transgenic animals the recombinant lentiviral gene transfer system of the invention is also used to splice the human counterpart onto an RNA encoding an animal protein.

The recombinant lentiviral gene transfer system of the invention can be used to produce replication-defective virion vector particles, which are a suitable tool for ribozyme mediated gene therapy. The safety of the recombinant lentiviral gene transfer system is enhanced as the risk of recombination and formation of a functional virus is limited to a minimum by a combination of several strategies:

1. separating of the genes encoding the functional VLV proteins Gag, Pol, Vif, Tat and Rev onto three different plasmids,

2. separating the genes encoding the functional VLV proteins from the packaging signal and the cis-acting sequences required for reverse transcription,

3. separating the genes encoding the functional VLV proteins from the gene encoding for the envelope protein, which is not derived from visna virus.

The risk of recombination is further reduced and the safety enhanced as the genes encoding the functional proteins of the visna virus are preferably composed of synthetic non-naturally occurring nucleotide sequences.

The proteins Pol, Vif, Rev and Tat are required for the proper function of the recombinant lentiviral gene transfer system.

To reduce the risk of recombination events all the plasmids of the invention are build in a way that they have minimal homology and minimal overlap between each other on the nucleic acid level.

The Pol, accessory (Vif, Rev and Tat) and structural protein coding sequences (Gag) are separated from the transfer plasmid on 3 separate plasmids.

By separating the genome of the construct over a number of plasmids the risk of the production of a replication competent retrovirus from the inventive gene transfer construct is reduced, as the number of recombination events, required is increased. In contrast to the natural visna virus, wherein the gag and pol genes overlap, in the present invention the genes encoding for the nucleocapsid protein - Gag - and the genes encoding for the Pol protein are separated onto two different plasmids- further referred to as "Pol plasmid" (c) and "Gag plasmid" (a).

In the wild-type virus the Gag and Pol polyproteins are encoded on overlapping open reading frames in a manner that ensures that Pol is produced at a lower rate than Gag. In the gene transfer construct of the present invention Gag and Pol proteins are coded for by synthetic nucleotide sequences. These synthetic nucleotide sequences code for the same amino acid residue sequence as the wild type using codons that are as different as possible from the wild type sequence. This alteration in codon usage abolishes the possibility of a open reading frame overlap for the Gag and Pol coding sequences.

To ensure a high titre of gene transfer constructs both Gag and Pol expression should be under the control of strong promoters. However, the Gag and Pol coding sequences need to be under the control of separate promoters. In the wild-type visna virus the Pol poly protein is produced at a lower rate (approximately 20-fold lower) than the Gag polyprotein. As the Pol coding sequence and Gag coding sequence are separated in the invention on different plasmids their products should also be expressed at different rates, in a ratio that corresponds to the natural ratio. To achieve this Pol expression is according to the invention always under the control of a weaker, preferably ten to twenty-fold weaker, promoter than Gag expression. Preferably, the Gag coding sequence is under the control of the CMV promoter and the Pol coding sequence under the control of the SV40 promoter. Other promoter pairs, which ensure that the expression of Gag significantly exceeds that of Pol, are also suitable.

Genes encoding the accessory proteins of the visna virus, namely Vif, Rev and Tat, which are necessary for virion production are located preferably on a third plasmid further referred to as "VTR plasmid" (d).

In the wild type virus the ORF (open reading frame) vif overlaps the ORF pol by 43 bp. In the present invention alteration in codon usage of vif and pol necessitates that the overlap between these ORF is abolished and that these are under the control of separate promoters. This necessity, is met by separating the accessory protein coding sequences for Vif, Tat and Rev onto a separate plasmid. Again, the separation of the sequences vif, tat and reu from the sequence pol introduces a safety improvement. The gene transfer system further comprises a transfer plasmid (b) with a cassette for class I intron ribozyme-mediated RNA repair, which contains a class I intron ribozyme sequence and a cloning site, wherein a heterologous gene or a part of a heterologous gene may be inserted.

When the transfer plasmid is transferred into a eukaryotic cell the class I intron ribozyme sequence is stably integrated as DNA into the target cell genome where the ribozyme RNA is expressed under the control of a viral promoter. The ribozyme and the heterologous gene sequence are then transcribed to a RNA which anneals with the mRNA transcribed by a target gene (target mRNA). This ribozyme then splices a predetermined sequence 3' to a cut it makes in the target mRNA. By the trans-splicing activity of the class I intron ribozyme a part of the target mRNA is than replaced by the RNA provided by the heterologous gene sequence.

Compared to conventional gene therapy approaches this class I intron ribozyme-mediated RNA repair has the advantage that the target is transcribed under its natural promoter and then repaired on the messenger RNA-level. By the trans-splicing activity of the intron I ribozyme the mutant part of the target mRNA is replaced with the correct counterpart. In this way the target mRNA is reassembled into an mRNA encoding a functional protein with desired activity.

The ribozyme cassette containing the group I intron catalytic ribozyme sequence has four major functional domains:

1. the promoter driving expression of the ribozyme;

2. the guide sequence that is responsible for the recognition of the target RNA and the cleavage and splicing site within the target RNA;

3. the catalytic RNA and;

4. the sequence to be spliced 3' onto the cleaved 5' portion of the target RNA or a cloning site, wherein the sequence to be spliced 3' onto the cleaved 5' portion of the target RNA can be inserted.

Thus, the ribozyme cassette consists of the promoter, that drives the expression of the ribozyme, immediately followed by the guide sequence, that recognises and binds to the target RNA molecule, followed by the ribozyme catalytic unit, followed by the sequence to be spliced 3' to the 5' fragment of the cleaved target RNA molecule. The expression of the ribozyme is preferably placed under the control of the SV40 promoter. The ribozyme catalytic unit is preferably from a Tetrahymena thermophylia 26S rDNA intron, preferably T. hyperangularis (GenBank X03106) T. sonneborni (GeπBank X03108), T. cosmopolitanis; (GenBank X03107), T. pigmentosa (GenBank V01412) most preferably T. thermophylia (GenBank V01416).

The guide sequence and the sequence to be spliced are chosen dependent on the target mRNA molecule. The guide sequence precedes the ribozyme catalytic unit immediately 3' and recognises the defined splice site in the target RNA by Watson-Crick base paring.

The sense oligonucleotide of the DNA fragment posses the sense portion of the guide sequence followed by the first 5-nt of the ribozyme catalytic sequence. The reverse complement oligonucleotide of the DNA fragment posses the 5-nt complement to the overhang at the end of the ribozyme promoter sequence followed by the antisense complement to the desired guide sequence.

The sequence to be spliced 3' onto the cleaved 5' portion of the target RNA depends on the location of the mutation, which has to be repaired. As the trans-splicing activity of the ribozyme replaces the 3' part of the target mRNA the ribozyme cassette comprises the nucleotides, which correspond to the residues where the mutation has occurred and the part of the gene encoding the target mRNA which is 3' to the residues where the mutation has occurred until the end of the gene in question.

The ribozyme cassette is preferably constructed as one unit. The ribozyme cassette as one unit makes it suitable for further development as a therapeutic agent to correct mutant RNA.

A preferred ribozyme cassette to repair mutant amyloid precursor protein (βAPP) mRNA is part of the transfer plasmid with the nucleic acid sequence according to SEQJD No. 6 and the plasmid map shown in Fig. 4. This ribozyme cassette is designed to repair a series of mutations in the mRNA encoding the amyloid β-A4 precursor protein (βAPP).

Preferably the L1 loop of the ribozyme catalytic unit is altered so that it posses a reverse complement (here the 6-nucleotide sequence GGACAT),"to the 5' part, preferably the 2^πd-7^th nt, of the sequence to be spliced (here atgtcc). This approach can be applied to the exchange of any mutant sequence encoded by the nuclear genome. The reverse complement is part of the guide sequence "GS". A preferred guide sequence (GS) used for the ribozyme that recognises the amyloid precursor protein (APP) that is expressed both by NGF-stimulated PC12 cells and HEK293 cells is GGTGGCGCTCCTCTGGGG. This GS serves as a recognition site and splice site for the ribozyme. The GS-DNA fragment is for example created from two oligonucleotides GCTCGGTGGCGCTCCTCTGGGG and TAATCCCCAGAGGAGCGCCACC. Annealing these two oligonucleotides produces a DNA fragment encoding the anti APP ribozymal GS that was compatible to the ribozyme 5' overhang and the CMV promoter 3' overhang. Changing the sequence of the GS allows a ribozyme to be designed for potentially any target RNA.

In an another embodiment of the invention the ribozyme cassette contains a stuffer sequence in place of the guide sequence. The guide sequence or the stuffer has a preferred length of 5 to 20 nucleotides length, more preferred 10 to 16 nt length.

The stuffer enables the end user to insert a GS of their choice into the ribozyme cassette. The target RNA, and the splice site within that target RNA, for the ribozyme cassette is determined by the GS. The transfer plasmid possessing the stuffer in place of the guide sequence is thus a versatile construct that can be adapted to revise the target RNA of choice. The stuffer is designed to facilitate its replacement by the guide sequence. The stuffer possess unique restriction sites. The stuffer replacing the guide sequence, e. g. gcgcgccgaacctcgagtacgccggcg, possesses preferably a Psrl restriction enzyme recognition- site. The Psrl restriction site ([7/12]GAACNNNNNNTAC[12/7]) cuts both up stream and down stream of the recognition site leaving 5-nt overhangs.

The end-user replaces the stuffer to produce a transfer plasmid with specificity towards the target RNA of choice which is to be repaired or edited The stuffer is replaced by a DNA fragment containing the guide sequence constructed from two oligonucleotide of preferably 10 to 20 nucleotides length, more preferred 12 to 16 nt in length.

In a further embodiment of the invention the ribozyme cassette is constructed with a marker protein coding sequence to be spliced 3' onto the cleaved 5' portion of the target RNA. Preferred marker proteins are enhanced green fluorescent protein (EGFP), placental alkaline phosphatase (PLAP) or human codon optimised versions of the nucleotide sequences coding for E. coli β-galactosidase or E. coli alkaline phosphatase (ECAP). These ribozyme cassette sequences are part of the transfer plasmids pBR322dtVLV-ECAP, pBR322dtVLV-PLAP, pBR322dtVLVt-βGal and pBR322dtVLV-GFP, (SEQJD No. 7 to 10, Fig. 2 and Fig. 9 to 11 ).

The ribozyme cassette supplied with a stuffer in place of the guide sequence and a marker protein coding sequence to be spliced into the target RNA can be used to create series of transfer plasmids that hybridise a marker protein to an endogenous protein normally produced by the host cell for transport and localisation studies.

An example of a ribozyme cassette with a stuffer in place of the guide sequence and a coding sequence coding for the marker protein GFP is part of the transfer plasmid pBr322dtVLV- GFP with the nucleic acid sequence according to SEQJD No. 10 and the plasmid map shown in Fig. 11.

In a preferred embodiment of the invention the ribozyme cassette is constructed with a stuffer in place of the guide sequence and a cloning site in place of the sequence to be spliced 3' onto the cleaved 5' portion of the target RNA.

When the sequence to be spliced onto the cleaved target RNA is replaced by a cloning site, this cloning site (e. g. aggtcctagtactegagtacctgcaggg) posses preferably a Bael recognition site ([10/1S]ACNNNNGTAYC[IiZ/?]). Digestion with Bael leaves a 5-nt overhang at the 3' end of the ribozyme catalytic sequence and a 5'-overhang at the 5'-end of the U3 sequence. A marker-protein coding-sequence can be inserted between these sites.

A ribozyme cassette, provided with stuffer in place of the guide sequence and a cloning site in place of the sequence to be spliced onto the target RNA, is a versatile construct that can, in principle, be adapted to specifically cleave and splice any target RNA of interest.

An example of a ribozyme cassette with a stuffer in place of the guide sequence and a cloning site in place of the sequence to be spliced onto the target RNA is given in the transfer plasmid pBR322dtVLV-PsrBae with the nucleic acid sequence according to SEQ ID No. 5 (Fig. 12).

The transfer plasmid further contains cis-acting elements required for reverse transcription of the plasmid genome and a competent packaging-signal. Cis acting elements are regions of RNA that have a functional role as RNA in the processing of the plasmid or cassette that they are part of.

The transfer plasmid preferably contains the major cis-acting elements responsible for the packaging of the viral genome into the virion, which, in the wild type visna virus and in the transfer plasmid, are located 3' to gag and in the p16 region of gag. In the transfer plasmid of the invention gag has been inactivated by the removal of the start codon (SEQJD No. 16) and truncated after the homologue to the p16-coding region to obtain D-p16 (SEQJD No. 17). This treatment inactivates gag and leaves the cis-acting elements intact.

While the major portions of the RNA elements responsible for interaction with the packaging proteins of the virion are located in the gag leader sequence Gag-L, and in D-p16, they are not exclusively restricted to these positions. Thus, the transfer plasmid preferably further comprises further cis acting elements chosen from:

1. the Rev response element (RRE₁ SEQJD No. 21 ) of the wild type visna virus,

2. the primer-binding site of the wild type visna virus,

2. the LTRs of the wild type visna virus,

3. the poly purine tract and its complement of the wild type visna virus,

4. the central termination sequence of the wild type visna virus,

5. inactivated sequences of VLV tat (D-tat, SEQ_ ID No. 18), VLV rev (D-rev, SEQ_ ID No. 19) and VLV env (D-env, SEQ_ ID No. 20), which doe not encode for functional proteins as the start codons are removed.

The cis acting elements are preferably chosen from the group of nucleotide sequences according to SEQ_ ID No. 16 to 21.

The competent packaging signal in the transfer plasmid comprises RNA elements that are sufficient to induce the packaging, of the RNA molecule that they are part of, to be packed into a virion. The competent packaging signal is contained in the gag-leader (Gag-L) and in the inactivated p16 (D-p16) sequence of the transfer plasmid.

All the other plasmids (other than the transfer plasmid) do not have a competent packaging signal. This is preferably accomplished by omitting the packaging signal from the artificial sequences constructed.

The transfer plasmid containing cis-acting elements do not encode for functional proteins of VLV. The lack of functional proteins in the transfer plasmid and its lack of homology to the plasmids containing the functional proteins abolish the risk of the emergence of recombinant replication competent viruses from the gene transfer construct.

Preferred transfer plasmids have the sequences according to SEDJD No. 5 to 10. To remove cis-acting elements from the Gag plasmid of the invention the region 3' to gag has been deleted. In the Gag plasmid of the invention the cis-acting nucleotide regions in gag have been disrupted by maximally altering the codon usage. The expression of Gag in the Gag plasmid of the invention is preferably under the control of the CMV promoter. The Gag plasmid thus does not contain a functional packing signal and is not able to recombine with the inactivated gag sequences on the transfer plasmid.

In the recombinant lentiviral gene transfer system of the invention the gene encoding the visna virus envelope protein has been inactivated in the transfer plasmid by omission of the start codon to obtain D-env. The wild-type env ORF contains the rev1 and rev∑ ORFs and the RRE. An altered form of rev is present in the VTR plasmid. To extend the cell-tropism, of the gene-transfer construct, in the invention the visna-virus envelope protein gene is replaced by a gene encoding for an envelope protein of a different virus. The process of encapsulating the gene transfer constructs with an envelope protein differing from that of the wild-type virus is termed pseudotyping. Preferred envelope proteins are: Ebola envelope (GenBank U31033), gp64 envelope glycoprotein from baculovirus (GenBank AF190124), influenza A virus Haemagglutinin (GenBank, Z46395), Jaagsiekte sheep retrovirus (JSRV) envelope (GenBank Y18303), Lymphocytic choriomeningitis virus (LCMV) glycoprotein (LCMV-GP, GenBank AF186080), Mokola virus envelope glycoprotein (MK-G, GenBank S59448), amphotropic murine leukaemia virus (MuLV) amphotropic envelope (4070A-Env, Genbank M33469), rabies-virus envelope glycoprotein (GenBank AB110669), RD114 retrovirus envelope (RD114-Env ; GenBank X87829).

A most preferred envelope protein is vesicular stomatitis virus envelope glycoprotein (VSVG; GenBank V01214).

The envelope protein (Env) encoding sequence is preferably located on a further plasmid (e), which does not encode for VLV proteins and does not contain a competent packing signal. This plasmid is further referred to as "Env plasmid". A preferred Env plasmid encoding the vesicular stomatitis virus envelope glycoprotein (VSVG) pBR322dtVSVG has the nucleic acid sequence according to SEQ ID No. 4 (see Fig. 8 for the plasmid map).

All the plasmids, beside the transfer plasmid, preferably contain a known polyadenylation site, preferably from bovine growth hormone.

To further minimize the risk of heterologous recombination with natural occurring viruses or with the transfer plasmid the genes encoding the functional proteins (Gag, Pol, Vif, Rev and Tat) of the visna virus used in the present invention are altered with respect to the wild type sequences naturally occurring in the visna virus. The genes are preferably composed of synthetic non-naturally occurring nucleotide sequences. Compared to the natural occurring genes the genes are altered at the nucleic acid level so that they differ as much possible on the nucleic acid level, whilst still encoding the same proteins. At the same time the DNA sequences are optimised for translation in human cells.

This alteration is achieved by using different codons than those used by the wild type viral genes. The codons are substituted for codons that represent the same amino acid residue but differ in as many bases as possible. Of the alternative codons possible, the codon most frequently used by human genes that differed from the wild type-codon is used. This alteration results preferably in a homology between the synthetic genes and the wild type gene, which is lower than 65 %, preferably 61%. The inventive alteration of on average every third nucleotide leads advantageously to the destruction of c/s-acting inhibitory sequences and also the destruction of the packaging signal in the Gag-coding region.

^■ The homology between the transfer plasmid and the VTR, Gag and Pol plasmids is further reduced since most of the Gag-coding region and the complete Pol-coding region was eliminated from the transfer plasmid. This strategy reduces the frequency of recombinations advantageously over 100 fold.

The altered genes are preferably constructed from synthetic oligonucleotides with length of 48 to 80 nucleotides, which have been assembled by PCR and successive ligations and digestions. This procedure has three advantages: Firstly, the construction from smaller fragments enables a greater variety of restriction sites to be used to construct the derivatives. Secondly, the omission of a particular fragment from the completed plasmid enables a simple approach to reducing the viral-derived content of the final transfer plasmid. Thirdly, the design of the virus was based on a consensus sequence instead of a characterised wild-type virus and thus the construction strategy provides greater control of sequence fidelity than does RT- PCR of a wild-type virus.

The gene encoding the Gag polyprotein has preferably a nucleic acid sequence according to SEQJD No. 11.

The gene encoding the Pol polyprotein has preferably a nucleic acid sequence according to SEQJD No. 12.

The genes encoding Vif, Rev and Tat preferably have the nucleic acid sequence according to SEQ ID No. 13 to 15. in a preferred embodiment of the invention the recombinant lentiviral gene transfer system comprises: a.) a Gag plasmid comprising an artificial nucleic acid sequence encoding VLV Gag, preferably according to SEQJD No. 11 and preferably under control of the CMV promoter, and a polyadenylation site, preferably from bovine growth hormone, wherein said plasmid does not contain a competent packaging signal; b.) a transfer plasmid - comprising an VLV nucleic acid sequence containing cis-acting sequence elements required for reverse transcription of the plasmid genome wherein said vector contains:

(i) the Gag-Leader (Gag-L), and inactivated sequences of gag p16 (D-p16), containing a competent packaging signal and the cis acting elements D-tat, D-rev and D-env, which do not encode for functional proteins as the start codons are removed, and a competent Rev-response element (RRE);

(ii) a ribozyme cassette comprising either a guide sequence of 10 to 20 nucleotides length or a stuffer flanked by restriction sites, preferably Psrl restriction sites, the catalytic domain of a group I intron, preferably a Tetrahymena 26S rDNA intron, either a nucleic acid sequence encoding for a the accurate counterpart of the target RNA to be repaired to be spliced onto the cleaved target RNA or a cloning site flanked by restriction sites, preferably Bael restriction sites, whereas the ribozyme cassette is preferably under the control of the SV40 promoter; c.) a Pol plasmid comprising an artificial nucleic acid sequence encoding VLV Pol, preferably according to SEQJD No. 12 and preferably under control of the SV40 promoter, and a polyadenylation site, preferably from bovine growth hormone, wherein said plasmid does not contain a competent packaging signal; d.) a VTR plasmid comprising artificial nucleic acid sequences encoding VLV Vif, Rev and Tat, preferably according to SEQJD No. 13 to 15 and preferably under control of the SV40 promoter, and a polyadenylation site, preferably from bovine growth hormone, wherein said plasmid does not contain a competent packaging signal; e.) a Env plasmid comprising nucleic acid sequence encoding (i) a viral envelope protein, which is not from visna virus, preferably vesicular stomatitis virus envelope glycoprotein (VSVG) and a polyadenylation site, preferably from bovine growth hormone, wherein said piasmid does not contain a competent packaging signal.

A most preferred Gag plasmid is pBr322dtVLVgag, with the nucleic acid sequence according to SEQJD No. 1. This plasmid includes the Gag encoding sequence according to SEQ ID No. 1 1. A most preferred Pol plasmid is pBr322dtVLVpol, with the nucleic acid sequence according to SEQJD No. 2. This plasmid includes the Pol encoding sequence according to SEQ ID No. 12.

A most preferred VTR plasmid is pBr322dtVLV-VTR, with the nucleic acid sequence according to SEQJD No. 3. This plasmid includes the Vif, Rev and Tat encoding sequences according to SEQ ID No. 13 to 15.

In a most preferred embodiment of the invention the recombinant lentiviral gene transfer system comprises: a.) the Gag plasmid pBr322dtVLVgag with the nucleic acid sequence according to

SEQJD No. 1 , b.) a transfer plasmid with a nucleic acid sequence chosen from pBR322dtVLVt-βGal, pBR322dtVLVdAPP, pBR322dtVLV-ECAP, pBR322dtVLV-PLAP, pBR322dtVLV-GFP, pBR322dtVLV-PsrBae with the nucleic acid sequences according to SEDJD No 5 to

10, c.) the Pol plasmid pBr322dtVLVpol with the nucleic acid sequence according to SEQJD

No. 2, d.) the VTR plasmid pBr322dtVLV-VTR with the nucleic acid sequence according to

SEQJD No. 3, e.) the Env plasmid pBr322dtVSVG with the nucleic acid sequence according to SEQJD

No. 4.

After transduction of the recombinant lentiviral gene transfer system into cells the cell will produce a replication-defective lentivirus particle.

A second aspect of the present invention is a method for producing replication-defective visna virus particles, comprising transfecting producer cells with a gene transfer system of the invention as described above.

The transfection of producer cells is performed by standard molecular biological methods, e. g. the calcium phosphate co-precipitation.

Preferred cells for virus production are the cell lines 293 (ATCC # CRL-1573) or 293T (ATCC # CRL-11268) or 293ts/A1609 (DuBridge et al., 1987, MoI Cell Biol. 7: 379-387). To produce a lentiviral stock the transfected producer cell is grown under cell culture conditions sufficient to allow production of replication-defective lentivirus particles in the cell and the replication-defective lentivirus particles are collected from the producer cell.

The visna virus particle obtained by one of the method of the invention and its use for ribozyme mediated repair of a mutated target RNA in eukaryotic cells are also objects of the present invention.

A further aspect of the present invention is a method of ribozyme mediated repair of a mutated target RNA or modification of a non-mutated RNA in eukaryotic cells, comprising the steps: a.) cloning a heterologous gene or a part of a gene encoding for the accurate counterpart to be repaired (e. g. the wild-type cDNA sequence of the mutated target RNA) corresponding to the part of the mutated target RNA, which has to be replaced, into the cloning site of the transfer plasmid, b.) transfecting producer cells with the so produced transfer plasmid and the other plasmids of the inventive gene transfer system; c.) growing the producer cells under cell culture conditions sufficient to allow production of replication-defective visna virus particles in the cell and collecting the replication- defective visna virus particles from the producer cell; d.) infection of eukaryotic cells with the replication-defective visna virus particles.

Before infection of eukaryotic cells the replication-defective visna virus particles are sterilized and cellular debris is removed, preferably by filtration through 0,22 μm pore size filters. Preferably the replication-defective visna virus particles are subsequently concentrated and macromolecules are removed, preferably on 0,1 μm filters.

For the infection the replication-defective visna virus particles are preferably resuspended in phosphate buffered saline.

By the infection of eukaryotic cells either in culture or administration of the replication- defective visna virus particles by in vivo injection, the ribozyme is brought into the cells and ribozyme-mediated repair is initiated.

The effectiveness of the infection and ribozyme mediated repair can be controlled by inserting a maker gene (as GFP) on the transfer plasmid. However, in many cases the presence of a maker gene is not desired, especially in therapeutic approaches of the ribozyme mediated RNA repair. Thus, in those cases, the effectiveness of the infection and ribozyme mediated repair is determined by specific measure of the amount of repaired RNA. This is preferably achieved by extraction of mRNA from the infected tissue or cell culture and performing a RT- PCR with primers specific for the 5' target RNA and the heterologous nucleotide sequence spliced 3' onto the cleaved target RNA. The production of a RT-PCR DNA product demonstrates the success of the ribozyme mediated mRNA repair.

Preferred uses of the recombinant lentiviral gene transfer system are:

• Screening Assays,

• High-Throughput In Vitro Metabolism Studies,

• Pharmacological profiling,

• In Vitro Pharmacological Characterisation,

• In Vivo Pharmacological Characterisation.

• Gene therapies.

For Screening Assays, In Vitro Metabolism Studies, Pharmacological profiling (Panning) and the In Vitro Pharmacological Characterisation the recombinant lentiviral gene transfer system of the invention is preferably applied to edit a gene message (mRNA) in animal cells so that they code for the human protein rather than the animals own version. Humanising proteins (e. g. enzymes or receptors as drug candidates) in animal cells in this way has several advantages:

Firstly, the test of the action of a candidate substance can be carried out in a cell that is similar to the human cell as possible. This produces a reproducible test system, which is more reliably than using human cells, as primary human cells in culture tend to change over time giving varying results. Also primary human cells are usually derived from patients and usually do not represent cells functioning normally.

Secondly, the humanised protein is expressed under control of the natural promoter and gene environment, which controls normally the expression of its animal counterpart. This ensures that the protein is expressed in natural levels and functions normally.

To humanise the protein a ribozyme cassette is constructed which contains at least a part of the human gene sequence to be spliced. The homologous animal gene sequence provides the information needed to design the recognition guide sequence. Analysis of the animal target-coding sequence provides putative splice sites for the ribozymes. Ribozymes using the recognition sequences and transferring easily detectable marker-sequences are constructed and inserted into the transfer vector. The recombinant lentiviral gene transfer system with the transfer vector possessing the ribozyme cassette infects primary animal cells. These cells express the human homologue in a direct assay for its presence. An assay for the physiological function confirms that the target exercises the presumed role in those cells that is missing in cells where the target has been replaced with a marker peptide.

Conventional Assays for Screening, In Vitro Metabolism Studies, Pharmacological profiling (Panning) and the In Vitro Pharmacological Characterisation can be adapted by replacing the conventional cells used for screening by the animals cells expressing the humanized protein, which is obtained by transfection with the lentiviral gene transfer system of the invention.

Another aspect of the invention is a kit, which contains the tools the applicant needs for inserting the gene of interest and a specific guide sequence into the ribozyme cassette of the transfer vector. With this kit the applicant can advantageously adapt the recombinant lentiviral gene transfer system of the invention for almost all his needs, e. g. to create a specific transfer plasmid to humanize a specific protein in a given animal cells. This kit comprises:

a.) The recombinant lentiviral gene transfer system of the invention with the five plasmids a) to e), whereas the ribozyme cassette of the transfer plasmid contains a cloning site, wherein the heterologous gene or a part of a gene encoding for the accurate counterpart of the target RNA to be repaired or modified can be inserted and a stuffer sequence flanked by restriction sites, wherein the specific guide sequence can be inserted, b.) a laboratory protocol describing the use of the kit for the end user.

The kit preferably contains additionally packaging cells, buffers for transfection, restriction enzymes, oligonucleotides for the construction of the guide sequence and materials and reagents for expansion of the plasmids from other commercial sources described in the laboratory protocol. In an alternative embodiment of the kit all or a part of these materials are supplied by the end user.

The kit preferably contains the plasmids pBR322dtVLVgag as plasmid a) and pBR322dtVLVpol, pBR322dtVLV-VTR and pBR322dtVSVG as plasmid c) to e).

A kit for tagging an E.Coli β-galactosidease sequence to the gene selected by the end user preferably contains the plasmid pBR322dtVLV-ECAP as transfer plasmid (plasmid b)). A kit for tagging an humanised β-galactosidease sequence to the gene selected by the end user preferably contains the plasmid pBR322dtVLVt-βGal as transfer plasmid (plasmid b)). A kit for tagging an placental alkaline phosphatase sequence to the gene selected by the end user preferably contains the plasmid pBR322dtVLV-PI_AP as transfer plasmid (plasmid b)). A kit for tagging an enhanced green fluorescent protein sequence to the gene selected by the end user preferably contains the plasmid pBR322dtVL V-GFP as transfer plasmid (plasmid b)).

A kit for tagging an reporter gene to be selected by end user sequence to the gene selected by the end user preferably contain the plasmid pBR322dtVLV-PsrBae as transfer plasmid (plasmid b)).

Another use of the recombinant lentiviral gene transfer system of the invention is to generate transgenic animals expressing a humanised protein. These transgenic animals can be used for In Vivo Pharmacological Characterisation of drug candidates.

These transgenic animals are preferably obtained by injecting replication-defective virion particles into the animals, preferably intravenously.

Where the humanised protein should be expressed in the central nervous system the the replication-defective virion particles are injected intra cranial.

For injection the replication-defective virion vector particles are obtained from the conditioned medium of producer cells as described above by transfecting producer cells with the recombinant lentiviral gene transfer system of the invention.

In a preferred embodiment the conditioned medium used for injections is initially filtered through 0.22μm syringe filters. The filtered conditioned medium is then concentrated by centrifugation (9 00Og for 1h at 4°C) through centrifuge filters with a cut-off of 1000 kDa. The concentrated conditioned medium is resuspended in isotonic saline buffered with 20 mm sodium phosphate at pH 7.2. The resuspended concentrated conditioned medium is used directly for injections.

Another preferred use of the recombinant lentiviral gene transfer system of the invention is Gene Therapy. When applied for Gene Therapy the recombinant lentiviral gene transfer system of the invention revises a gene message that causes a disease to the normal gene message and thereby prevents or cures the disease. The following Examples are provided to enable one skilled in the art to practice the invention and are merely illustrative of the invention. They should not be read as limiting the scope of the invention as defined in the claims.

Fig. 1 shows a plasmid map of the plasmid pBR322Lnk1 , a parent plasmid for the transfer plasmids of the lentiviral gene transfer construct expression system.

pBR322Lnk1 has been constructed from the cloning plasmid pBR322 by removing the tetracycline resistance gene (tet) between the Styl and Hindlll restriction sites. A synthetic linker (Lnk1 ) has been cloned in the place of the removed sequence. This linker Lnk1 posses a Swal blunt-end restriction site into which a PCR product can be cloned. Two Aarl recognition sites flank the Swal recognition and restriction site. The Aarl recognition sites are positioned so that the associated restriction sites fall at the ends of the DNA fragment that is cloned into the Swal site. A subsequent digestion of the derivative formed by ligation into the Swal site releases a DNA-fragment, derived from the inserted sequence with the four bp at the ends converted to four-base 5' overhangs. This sticky-end DNA-fragment is used to construct the plasmids of the present invention.

The plasmid pBR322Lnk1 contains further the sequences of the original pBR322, which are necessary for the replication, transcription and translation of the plasmid: two promoters (P1 P, P3 P), Shine-Dalgarno sequences (SD SEQ) acting as the ribosome binding site (RBS), the Ampicillin resistance gene (β-lactamase) AP^r, an L-strand and H-Strand-Y-effector site (L Y EFF and H Y EFF), an origin of replication (ORI) and a gene coding for the ROP protein.

To obtain pBR322Lnk1 the cloning plasmid pBR322 (ATCC 37017) is digested with the restriction enzymes Hindlll (New England Biolabs) and Styl (New England Biolabs). To perform the double digestion 1 μg pBR322 DNA, 1.5 μl of Styl and 1 μl of Hindlll and 1 μl BSA are added to 50 of NEB buffer 3 in a 500μl reaction tube. The mixture is incubated in a Thermomixer (Eppendorf, Wesseling-Berzdorf, Germany) at 37⁰C and 300 rpm for 1 hour. This digestion excises the tetracycline resistance gene. The digestion yields two fragments: the Styl-Hindlll fragment (1370-29) of 3021 bp length and the Hindlll-Styl fragment (30-1369) of 1340 bp length. The fragments are separated by electrophoresis in Agarose and the larger DNA fragment band is excised from the electrophoresis gel and the DNA is extracted and split into three aliquots.

To obtain the Linker Lnk1 the oligonucleotide Lnkis (agcttcacctgcatttaaatgcaggtgc) was annealed to another oligonucleotide (Lnki rc; caaggcacctgcatttaaatgcaggtga) and phosphorylated (5' with T4-polynucleotide kinase; NEB, M0201). Lnk1 is ligated into the pBR322 Styl-Hindlll fragment. The resulting plasmid is termed pBR322Lnk1. £. CoIi are transformed with the pBR322Lnk1 and the plasmid expanded overnight and then extracted.

Fig. 2 shows a plasmid map of pBR322dtVLVt-βGal, an example of the inventive transfer plasmid based on VLV, that splices the transcript of the reporter gene E. coli β-galactosidase gene 3' into a target mRNA. pBR322dtVLVt-βGal contains a ribozyme cassette designated "ribozyme" and a stuffer sequence designated "sPsrl". The stuffer "sPsrl" can be substituted for a ribozyme guide- sequence recognizing a specified target mRNA. The coding sequence to be spliced 3' by the ribozyme is here the reporter-gene "LacZ", encoding for the E. CoIi β-galactosidase (βGal ), wherein the codon-usage has been altered to the optimal human form. The ribozyme cassette is under the control of the SV40 promoter. The ribozyme corresponds to the catalytic domain of the 26S rDNA intron from T. thermophylia (GenBank V01416).

To enable reverse transcription, packaging and integration of the genome the transfer plasmid pBR322dtVLVt-βGal contains cis-acting elements of VLV₁ contained in nucleotide sequences designated "Gag-L", "pbs", "D-env", "D-rev, "D-tat", "D-p16", which do not encode for functional proteins, as the start codons (encoding for the first methionine) are removed. The gag-leader "Gag-L" and in the inactivated p16 sequence "D-p16" do also contain a competent packaging signal.

The transfer plasmid pBR322dtVLVt-βGal contains further a Rev-response element (RRE) responsible for the export of the virion mRNA from the host cell nucleus, a Shine-Dalgamo sequence (SD SE) acting as the ribosome binding site (RBS).

The RRE is required for the transport of viral RNA out of the nucleus of the producer cell and acts by an interaction with Rev. The REV2-L is the nucleotide sequence that contains the RRE. The sequence rev 2 is empirically determined to be necessary for the function of the GTC. U5 and R5 are necessary elements of the 5' LTR while, U3 is the untranslated leader sequence for the 3' LTR. The nucleotide sequence of pBR322dtVLVt-βGal is listed in SEQJD No. 9.

To obtain pBR322dtVLVt-βGal the plasmid pBR322Lnk1 is digested with the restriction-enzyme Swal (New England Biolabs). This digestion results in a blunt-end linear plasmid. The digested plasmid is dephosphorylated with Calf Intestinal Alkaline Phosphatase (CIP, New England Biolabs) and purified on an agarose gel. The DNA encoding the ribozyme cassette and the cis-acting elements are constructed based on synthetic oligonucleotides which are assembled to generate four DNA fragments U5-p16, tat-RRE, R2L-LacZ and U3-R3. A scheme for the construction of the plasmid pBR322dtVLVt- βGal is shown in Fig. 3. To summarize, pBR322dtVLVt-βGal is obtained by assembling the fragments U5-p16, tat-RRE, R2L-LacZ and U3-R3 to generate the fragment "VLV-LacZ" and cloning "VLV-LacZ" into the digested pBR322Lnk1.

The U5-p16-fragment contains the primer binding site (pbs), the long terminal repeats R5 and U5 of VLV, the Gag-leader (Gag-L, SEQ ID NO. 16) and the inactivated p16 region of gag ("D-p16", SEQ ID NO. 17). The tat-RRE-fragment contains the inactivated sequences "D-env" (SEQ ID NO. 20), "D-rev" (SEQ ID NO. 19), "D-tat" (SEQ ID NO. 18) and the RRE (SEQ ID NO. 21), Rev2-L, rev2 sequences. The R2L-LacZ-fragment contains the ribozyme cassette, including the PsRI-Stuffer (sPsRI), the SV40 promoter and the gene encoding for βgal (LacZ) The U3-R3-fragment contains the U3 and R3 region of VLV comprising the 3' LTR.

These DNA fragments U5-p16, tat-RRE, R2L-LacZ and U3-R3 were assembled from DNA fragments of 44 to 73 bp generated by PCR of chemically synthesized single stranded DNA oligonucleotides pairs (Sigma-Genosys Ltd., Cambridge CB2 4EF, UK). In the PCR the oligonucleotides acted both as primers and templates. The 3' oligonucleotide was equivalent to half the fragment plus 12 bases in length. The 5' oligonucleotide was equivalent to four bases of the fragment following 3' plus half the fragment in question plus 10 bases. The PCR extends the oligonucleotides, both sense and antisense, to the full length of the DNA fragment. Each DNA fragment constructed in this way overlaps the following DNA fragment by four bases. The overlap was designed to enable the subsequent ligation of the fragments to construct the full plasmid.

The PCR is performed in a sterile, nuclease-free 0,5 ml PCR-tube in a thermal cycler (MultiCycler PTC 200, Biozym Diagnostik GmbH, Hessisch Oldendorf, Germany) that has been preheated to 95⁰C under the following conditions:

PCR reaction mixture:

PCR thermal cycling conditions:

The PCR reaction produces 67-123 bp fragments (depending on the length of the oligonucleotides used in the PCR) that are separated from the reaction mixture by agarose gel electrophoresis on a 2% gel (61-3112 Certified Low-Melt agarose, Bio-Rad Laboratories GmbH, Mϋnchen, Germany) on a Sub-Cell Model 192 Cell exactly according to the manufacture's instructions (Sub-Cell^® Model 96 and Model 192 Agarose Gel Electrophoresis Systems Instruction Manual, Bio-rad).

The band containing the DNA fragment is cut out and the DNA extracted from the gel according to the manufactures instructions (QIAquick PCR purification/gel extraction kit, Qiagen, Hilden, Germany). The DNA fragment is blunt-end ligated into pBR322Lnk1 at the Swal site. The resulting plasmid is used to transform competent E.coli (One Shot^® TOP10 Chemically Competent E. coli, Invitrogen GmbH, Karlsruhe, Germany) exactly according to the manufacture's instructions. The plasmid is expanded overnight and extracted with a mini prep kit (Qiagen).

All other PCR, plasmid expansions and extractions are performed in exactly the same way unless otherwise stated. The plasmid DNA is digested with BfuAI (New England Biolabs GmbH, Frankfurt am Main, Germany) for 2h at 37⁰C. The fragment corresponding to the ligated PCR product was purified from an Agarose gel after electrophoresis.

The DNA fragments U5-p16, tat-RRE, R2L-LacZ and U3-R3 are assembled in this way by successive rounds of cleavage by the restriction enzyme BfuAI and re-ligation.

The SV40 early promoter is obtained from the plasmid pcD N A3.1 /CT-G FP-TO PO (Invitrogen) by a PCR amplification of the promoter coding sequence with the primers CTGTGGAATGTGTGTCAGTTAGGGTGT and

CTATTGGTTTAAAGACTAGCTACCAGGTGCAT under the PCR conditions as described above.

The DNA fragments U5-p16, tat-RRE, R2L-LacZ and U3-R3, the SV40 early promoter are ligated into pBR322Lnk1 at the Swal site to obtain pBR322dtVLVt-βGal. Fig. 4 shows a plasmid map of pBR322dtVLVdAPP a second example of the inventive transfer plasmid based on VLV. Plasmid pBR322dtVLVdAPP is a transfer plasmid designed to repair mutated βAPP.

The ribozyme cassette contains here a guide sequence "GS", which is complementary to the target sequence in βAPP mRNA and the βAPP pre-mRNA, the catalytic domain of the 26S rDNA intron Tetrahymena thermophylia ribozyme, and a sequence encoding for the homologous non mutated sequence of the 3'-part of APP (nt 1168 to 2266.).

The guide sequence "GS" recognizes the targets βAPP mRNA and the βAPP pre-mRNA, the ribozyme is capable of splicing the sequence encoded by the 3'-portion of the ribozyme cassette (encoding the homologous non mutated sequence) onto the cleaved target RNA.

The nucleotide sequence of pBR322dtVLVdAPP is listed in SEQJD No. 6.

The construction of pBR322dtVLVdAPP is identical to the construction of pBR322dtVLVt- βGal (s. Fig. 2) except that the sequence encoding the 3'-part of APP is introduced in place of LacZ and the guide sequence "GS" is introduced in place of the stuffer sPSRI.

The guide sequence (GS) used for the ribozyme that recognises the amyloid precursor protein (APP) that is expressed both by NGF-stimulated PC12 cells and HEK293 cells is GGTCCTCGGTCGGCAGCA . This GS serve as a recognition site and splice site for the ribozyme. The GS DNA fragment is created from two oligonucleotides GCTCGGTCCTCGGTCGGCAGCA and TAATTGCTGCCGACCGAGGACT. Annealing these two oligonucleotides produces a DNA fragment encoding the anti-APP ribozymal guide- sequence that was compatible to the ribozyme 5' overhang and the CMV promoter 3' overhang. Changing the sequence of the GS allows a ribozyme to be designed for potentially any target RNA.

Fig. 5 shows a plasmid map of pBR322dtVLVgag, an example for a Gag plasmid, encoding the polyprotein Visna lentivirus (VLV) Gag.

The plasmid pBR322dtVLVgag includes an expression cassette for the production of the polyprotein VLV-Gag (comprising the proteins p16, p25 and p14). The VLV-Gag coding sequence is completely artificial. The codon-usage of the VLV-Gag coding sequence has been maximally altered from the wild-type VLV gag coding sequence to minimize the risk of homologous recombination with wild type. Further the codon usage is optimized for translation in human cells and the VLV-Gag coding sequence is supplied with a bovine growth hormone poly-adenylation signal (BGH pA). The expression of the VLV-Gag coding- sequence is under the control of the Cytomegolo virus (CMV) promoter.

The nucleotide sequence of pBR322dtVLVgag is listed in SEQJD No. 1.

pBR322dtVLVgag is constructed based on pBR322Lnk1 and by assembling the VLV-Gag encoding sequence starting from synthetic oligonucleotides and subsequent ligations and digestions similar to the manner described under Fig. 2 for the transfer plasmid .

The CMV (cytomegalovirus immediate-early gene) promoter is obtained from the plasmid pcDNA3.1(+) (Invitrogen, Cat # V790-20) by a PCR amplification of the promoter coding sequence with the primers:

GTTGACATTGATTATTGACTAGTTATTAATAGTAATCAATTACGGGGTCATTAG and

GAGCTCTGCTTATATAGACCTCCCACCG under the PCR conditions as described above.

Fig. 6 shows a plasmid map of of pBR322dtVLVpol, an example for a Pol plasmid, encoding the polyprotein Visna lentivirus (VLV) Pol.

The plasmid pBR322dtVLVpol includes an expression cassette for the production of the polyprotein VLV-PoI. The VLV-PoI coding sequence is completely artificial. The codon-usage of the VLV-PoI coding sequence has been maximally altered from the wild-type VLV pol gene to minimize the risk of homologous recombination with wild type. Further the codon usage is optimized for translation in human cells and the VLV-PoI coding sequence is supplied with a bovine growth hormone poly-adenylation signal (BGH pA). The expression of the VLV-PoI coding-sequence is under the control of the Simian Virus 40 (SV40)-promoter.

The nucleotide sequence of pBR322dtVLVpol is listed in SEQJD No. 2.

The plasmid pBR322dtVLVpol is constructed based on pBR322Lnk1 and by assembling the VLV-PoI encoding sequence starting from synthetic oligonucleotides and subsequent ligations and digestions similar to the manner described under Fig. 2 for the transfer plasmid.

Fig. 7 shows a plasmid map of pBR322dtVL V-VTR an example for the VTR plasmid, comprising nucleic acid sequences encoding the accessory proteins VLV Vif, Rev and Tat. The plasmid pBR322dtVL V-VTR includes an expression cassette for production of the VLV- proteins Vif, Tat and Rev. The codon-usage of the Vif, Tat and Rev coding-sequence has been maximally altered from the wild-type to minimize the risk of homologous recombination with wild type. Further the codon usage is optimized for translation in human cells and the the VLV -Vif, Tat and Rev coding-sequence is supplied with a bovine growth hormone poly- adenylation signal (BGH pA).. The expression of the Vif, Tat and Rev coding-sequence is under the control of the Simian Virus 40 (SV40)-promoter.

The nucleotide sequence of pBR322dtVLV-VTR is listed in SEQJD No. 3.

ρBR322dtVL V-VTR is constructed based on pBR322Lnk1 and by assembling the Vif, Tat and Rev coding-sequence starting from synthetic oligonucleotides and subsequent ligations and digestions similar to the manner described under Fig. 2 for the transfer plasmid.

Fig. 8 shows a plasmid map of pBR322dtVSVG an example for the Env plasmid, encoding a viral envelope protein Env, which is not derived from visna virus.

Plasmid pBR322dtVSVG includes an expression cassette for the vesicular stomatitis virus envelope glycoprotein (VSVG). The VSVG coding sequence has been optimized for human codon usage and is supplied with a bovine growth hormone poly-adenylation signal (BGH pA). The expression of VSVG is under the control of the CMV promoter.

The plasmid pBR322dtVSVG is constructed based on pBR322Lnk1 and by assembling oliogonucleotides.

The nucleotide sequence of pBR322dtVSVG is listed in SEQJD No. 4.

Fig. 9 shows a Plasmid map of pBR322dtVLV-ECAP a third example for the inventive transfer plasmid, that splices the transcript of the reporter gene E. coli alkaline phosphatase 3¹ into a target mRNA

pBR322dtVLV-ECAP includes a ribozyme cassette similar to the plasmid pBR322dtVLVt- βGal (shown in Fig. 2), wherein the sPsrl stuffer can be substituted for a ribozyme guide- sequence for a specified target mRNA and the coding sequence to be spliced 3' by the ribozyme is the reporter-gene E. coli alkaline phosphatase coding-sequence wherein the codon-usage has been altered to the optimal human form.

The nucleotide sequence of pBR322dtVLV-ECAP is listed in SEQJD No. 7. The construction of pBR322dtVLV-ECAP is identical to the construction of pBR322dtVLVt- ^βGal (s. Rg. 2) except that the DNA-Fragment LacZ is replaced by the E. coli alkaline phosphatase coding-sequence optimized for human codon usage and is constructed from synthetic oligonucleotides as described above.

Fig. 10 shows a Plasmid map of pBR322dtVLV-PLAP a fourth example for the inventive transfer plasmid, that splices the transcript of the reporter gene placental alkaline phosphatase (PLAP) 3' into a target mRNA

The plasmid pBR322dtVLV-PLAP includes a ribozyme cassette similar to the plasmid pBR322dtVLVt-βGal (shown in Fig. 2), wherein the sPsrl stuffer can be substituted for a ribozyme guide-sequence for a specified target mRNA and the coding sequence to be spliced 3' by the ribozyme is the reporter-gene human) placental alkaline phosphatase (PLAP) coding-sequence wherein the codon-usage has been altered to the optimal human form.

The nucleotide sequence of pBR322dtVL V-PLAP is listed in SEQJD No. 8.

The construction of pBR322dtVLV-PLAP is identical to the construction of pBR322dtVLVt- βGal (s. Fig. 2) except that the DNA-Fragment LacZ is replaced by the placental alkaline phosphatase (PLAP) coding-sequence, which is obtained from as a synthetic sequence that has been optimized for human codon usage and is constructed from synthetic oligonucleotides as described above.

Fig. 11 shows a Plasmid map of pBR322dtVL V-GFP, a fifth example for the inventive transfer plasmid, that splices the transcript of the reporter gene enhanced green fluorescent protein (eGFP) 3' into a target mRNA

The plasmid pBR322dtVLV-GFP includes a ribozyme cassette similar to the plasmid pBR322dtVLVt-βGal (shown in Fig. 2), wherein the sPsrl stuffer can be substituted for a ribozyme guide-sequence for a specified target mRNA and the coding sequence to be spliced 3' by the ribozyme is the reporter-gene enhanced green fluorescent protein (eGFP) coding- sequence wherein the codon-usage has been altered to the optimal human form.

The nucleotide sequence of pBR322dtVLV-GFP is listed in SEQJD No. 10.

The construction of pBR322dtVLV-GFP is identical to the construction of pBR322dtVLVt-βGal (s. Fig. 2) except that the DNA-Fragment LacZ is replaced by the the reporter-gene enhanced green fluorescent protein (eGFP) coding-sequence, which is obtained from the plasmid pEGFP (BD Biosciences Clontech, Heidelberg, Germany) by a PCR amplification of the promoter coding sequence with the primers gtgagcaagggcgaggagctg and ttacttgtacagctcgtccatgccg. Fig. 12 shows a Plasmid map of pBR322dtVLV-PsrBae, a sixth example for the inventive transfer plasmid, that contains a cloning site, wherein a sequence to be spliced 3' by the ribozyme can be inserted into.

pBR322dtVLV-PsrBae includes a ribozyme cassette similar to the plasmid pBR322dtVLVt- βGal (shown in Fig. 2), wherein the sPsrl stuffer can be substituted for a ribozyme guide- sequence for a specified target mRNA. pBR322dtVLV-PsrBae further contains a cloning site (csBael) flanked by Bael restrictions sites, wherein a sequence to be spliced 3' by the ribozyme can be inserted into after a Bael-digest of the plasmid.

The nucleotide sequence of pBR322dtVLV-PsrBae is listed in SEQJD No. 5.

The construction of pBR322dtVLV-PsrBae is identical to the construction of pBR322dtVLVt- βGal (s. Fig. 2) except that the DNA-Fragment LacZ is replaced by the cloning site csBae constructed by insertion of a linker in place of the βGal coding nucleotide sequence.

Production of replication-defective virion vector particles:

For the production of replication-defective virion vector particles cells were transfected with an example of the recombinant lentiviral gene transfer system of the invention: the Gag plasmid pBR322dtVLVgag, the transfer plasmid pBR322dtVLV-GFP, the Pol plasmid pBR322dtVLVpol, the VTR plasmid pBR322dtVLV-VTR, encoding the accessory proteins Vif, Rev and Tat, and the Env plasmid pBR322dtVSVG, encoding the VSV-envelope protein.

To target the ribozyme cassette to the β-amyloid precursor protein (APP)-coding mRNA the stuffer sPsrl in the transfer plasmid pBR322dtVLV-GFP is here replaced by the guide sequence GGTGGCGCTCCTCTGGGG recognizing the APP-coding mRNA (s. Fig. 4). The ribozyme cassette will splice the EGFP coding sequence onto APP-mRNA.

The cell line 293T (ATCC CRL-11268) is maintained in Dulbecco's modified Eagle's medium supplemented with 10% foetal calf serum (FCS), penicillin, streptomycin, and glutamine (Invitrogen). The 293T cells were transfected by the calcium phosphate co-precipitation method as described (Soneoka et al., 1995, Nucleic Acids Res. 23: 628-633). The supernatant of the transfected 293T cells containing the replication-defective virion vector particles is passed through a 0,45 μm pore size filter and stored in 500 μl aliquots at -80⁰C. The frozen aliquots containing the replication-defective virion vector particles form a lentiviral stock, which is subsequently used for the infection of cells. Infection of cells with virion vector particles:

The infection of cells with virions vector particles is performed in the absence and presence of aphidicolin to determine if non-dividing cells can be transfected.

Virion vector particles are produced as described above.

Cultured rat pheochromocytoma cells PC12 (ATCC CRL-1721 ) and HEK293 cells are seeded in 96-well plates at a density of 2 x 10⁴ cells/well. PC12 cells are cultured in Ham's F12K medium with 2 mmol/l L-glutamine adjusted to contain 1.5 g/l sodium bicarbonate, 82.5%; horse serum, 15%; foetal bovine serum, 2.5% (Invitrogen) and 50ng/ml 2.6s nerve growth factor (Sigma-Aldrich Chemie, Deisenhofen, Germany). HEK293 cells are cultured in minimum essential medium (Eagle) with 2 mmol/l L-glutamine and Earle's BSS adjusted to contain 1 ,5 g/l sodium bicarbonate, 0,1 mmol/l non-essential amino acids, and 1 ,0 mmol/l sodium pyruvate, 90%; heat-inactivated horse serum, 10% (Invitrogen).

One day later, the medium is removed and cells are incubated for 2 to 4 hr with serial dilutions (1 :30, 1 :100, 1 :300, 1 :1000, 1 :3000, 1 :10000) of the virion vector particles preparations in a total volume of 150 μ//well of a 24-well plate or 30 to 50 μ//well of a 96-well plate. Fresh medium is added and the number of GFP-positive cells is determined 2 days after infection.

To arrest PC12 and HEK293 cells in the G-i/S phase of the cell cycle, cells are seeded in a 96-well plate at a density of 4 x 10⁴ to 5 x 10⁴ cells/well at an aphidicolin concentration of 5 μg/ml (Sigma) being present during the entire culture period.

The number of transfected cells is determined by measuring the GFP fluorescence in the lysates of three independent transfections in the presence of aphidicolin at 5μg/ml (+) or its absence (-) (see Fig. 13).

The transfected PC12 and HEK293 cells are suspended in luciferase cell culture lysis reagent (Promega, Mannheim, Germany) and the GFP fluorescence intensity of the lysates was determined in a 1420 multi-label counter (Victor; Wallac, Turku, Finland) by the method of Schnell et al. (2000, Development of a self-inactivating, minimal lentivirus vector based on simian immunodeficiency virus, Human Gene Therapy 11: 439- 447). In Fig. 13 the means and standard errors of the GFP fluorescence measurement are shown (GFU, Green Florescence-forming Units).

In presence and absence of Aphidicholin a high number of PC12 and 293 cells are infected with the virion vector particles. Although titres were lower in case of Aphidicholin treatment the results significantly show that non-dividing cells can be efficiently transfected with the replication deficient virion particles according to the invention.

Repairing a point mutation by ribozvme mediated RNA repair:

The repair of point mutations in the messenger RNA encoding for the β-amyloid-precursor protein is described as an example of ribozyme-mediated repair with the recombinant lentiviral gene transfer system of the invention.

A number of point mutations in the β-amyloid-precursor protein (APP, GenBank D87675) coding gene, that are amenable to correction by substitution of a 3' sub-sequence, give rise to clinically significant pathologies (Online Mendelian Inheritance in Man *104760, NCBI, Bethesda, USA):

0001 Glu693Gln Dutch Type Cerebroarterial Amylodosis

0002 Val717lle Familial Alzheimer Disease

0003 Val717Phe Familial Alzheimer Disease

0004 Val717Gly Familial Alzheimer Disease

0005 Ala692Gly Pre-Senile Dementia and Cerebral Amyloidosis

0006 Aia713Val Schizophrenia

0008 Lys670Asn and Met671 l_eu Familial Alzheimer Disease

0009 Ala713Thr Familial Alzheimer Disease

0010 Glu665Asp Late Onset Familial Alzheimer Disease

0011 He716Val Early Onset Familial Alzheimer Disease

0012 Val715Met Early Onset Familial Alzheimer Disease

0015 Thr714lle Early Onset Familial Alzheimer Disease

0017 Thr714Ala Familial Alzheimer Disease

The transfer plasmid used to transfer the self-splicing ribozyme that corrects these point mutations is pBR322dtVLVdAPP, which is represented as a plasmid map in Fig. 4.

For the production of replication-defective virion vector particles cells were transfected as described above with the transfer plasmid pBR322dtVLVdAPP, the Gag plasmid pBR322dtVLVgag, the Pol plasmid pBR322dtVLVpol, the VTR plasmid pBR322dtVL V-VTR, encoding the accessory proteins Vif, Rev and Tat and the Env plasmid pBR322dtVSVG, encoding the VSV-envelope protein. PC12 cells and HEK293 cells were infected with virion vector particles as described above. Before infection PC12 cells were stimulated with 50 ng/ml nerve growth factor (NGF) (N6009, Sigma-Aldrich Chemie GmbH, Germany) for seven days.

Alternatively primary cells can be infected in culture or the virion vector particles can be administered by in vivo injection.

To control the efficiency of infection mRNA is extracted from the infected cell culture (or tissue) by means of the Oligotex Direct mRNA Micro Kit (Qiagen, Hilden, Germany) according to the manufactures instructions.

An RT-PCR is performed with primers specific for the 5' substrate RNA and the heterologous nucleotide sequence spliced 3' onto the cleaved substrate RNA The RT-PCR is performed by the method of Bangsow, Huch, Male and Mϋller (2002; chapter 5.2.2.1 Schrimpf (Ed) in Gentechnische Methoden, Spektrum Akademischer Verlag GmbH, Heidelberg, Germany) with the PCR performed under the conditions specified above. The antisense primer used is GACGATCACTGTCGCTATGACAACAC while the sense primer is TGCCGACCGAGGACTAATGTCCCAGGTCATGAGAGA.

The DNA fragments produced by the RT-PCR are separated by Agarose gel electrophoresis. Agarose gel electrophoresis is carried out as described above. The antisense primer is specific for wildtype βAPP. The presence of a band of RT-PCR product (length 952bp) is indicative of effective trans-splicing of the transduced ribozyme. The production of a RT-PCR DNA product demonstrates the success of the mRNA repair.

Thθ following abbreviations are used in the description of the invention and in the figures:

3¹APP ^{part of tne} β-Amyloid-precursor protein (APP) encoding sequence to be spliced 3' to repair a mutant APP message

Aarl Recognition site for the restriction enzyme Aarl

Ap^r Ampicillin resistance gene (β-lactamase)

ATCC American Type Culture Collection

Bael Bael restriction site csBael cloning site flanked by Bael restriction sites for the introduction of a target sequence

BGH Bovine growth hormone

CMV Cytomegolovirus

D-env Inactivated VLV envelope nucleotide sequence

D-p16 Inactivated VLV p16 nucleotide sequence

D-rev Inactivated VLV rev nucleotide sequence

D-tat Inactivated VLV tat nucleotide sequence

ECAP Humanised E. coli alkaline phosphatase coding-sequence

EGFP Enhanced green fluorescent protein

EGFP-PA EGFP polyadenylation signal

Gag Humanised Gag-coding sequence

Gag-L Gag leader sequence

GFP Green fluorescent protein

GFU, Green Florescence-forming Units.

GS Ribozyme guide-sequence

H Y EFF H-strand Y-effector site

HEK293 Human embryonic kidney fibroblast cell-line

Hindi!! Hindlll restriction enzyme recognition site

L Y EFF L-strand Y-effector site

LacZ Sequence coding for a humanised form of the E. coli β-galactosidase

Lnk1 Linker 1

LTR3 3' Long terminal repeat

LTR5 5' Long terminal repeat

ORI Bacterial origin of replication p Plasmid

P10' Loop P10 of the ribozyme catalytic sequence

P1 P Bacterial promoter P1

P3 P Promoter P3

PA Polyadenylation signal pbs Primer binding site

PC12 Rat pheochromocytoma cell line

PLAP Optimised human placental alkaline phosphatase coding sequence

Pol Humanised sequence encoding for the VLV Pol-polyprotein p - Stuffer flanked by Psrl restriction sites for the introduction of a ribozyme guide sequence

Psrl Restriction enzyme Psrl recognition site

R5 5' repeat

Rev 2 Second rev exon

Rev2-L Rev 2 leader

Ribozyme Ribozyme catalytic sequence

ROP Rop protein coding gene

RRE Rev response element

SD SEQ Shine-Dalgarno sequence

Styl Styl restriction enzyme recognition site

SV40 simian virus 40 early promoter

Swal Swal restriction enzyme recognition site

Tat Humanised transactivator of transcription-coding sequence

U3 Untranslated 3¹ sequence

U5 Untranslated 5' region of VLV vif Humanised Viral-infectivity factor coding-sequence

VLV Visna lentivirus

VSVG Vesicular stomatitis virus envelope glycoprotein coding-sequence

Claims

Claims:

1. A recombinant lentiviral gene transfer system for ribozyme mediated repair of a target RNA comprising:

(a) a first plasmid comprising an nucleic acid sequence encoding the polyprotein Visna lentivirus (VLV) Gag, wherein said plasmid does not encode for other VLV proteins and does not contain a competent packaging signal,

(b) a second plasmid as transfer plasmid comprising:

(iii) a ribozyme cassette comprising the catalytic domain of a self-splizing intron ribozyme (class I intron) sequence and a heterologous gene or a part of a gene encoding for the accurate counterpart of the target RNA to be repaired or a cloning site wherein the heterologous gene or a part of a gene encoding for the accurate counterpart of the target RNA to be repaired may be inserted;

(c) a third plasmid comprising an nucleic acid sequence encoding the polyprotein VLV Pol, wherein said plasmid does not encode for other VLV proteins and does not contain a competent packaging signal,

(d) a fourth plasmid comprising nucleic acid sequences encoding the proteins Vif, Rev and Tat, wherein said plasmid does not encode for other VLV proteins and does not contain a competent packaging signal,

(e) a fifth plasmid comprising a nucleic acid sequence encoding a viral envelope protein Env, which is not derived from visna virus, and wherein said plasmid does not encode for VLV proteins and does not contain a competent packaging signal.

2. A recombinant lentiviral gene transfer system according to claim 1 , characterized in that the sequences encoding VLV Gag, Pol, Vif, Rev and Tat are composed of synthetic non-naturally occurring nucleotide-sequences.

3. A recombinant lentiviral gene transfer system according to claim 2, characterized in that the sequences encoding the VLV proteins Gag, Pol, Vif, Rev and Tat comprise the sequences according to SEQJD . No. 11 to 15.

4. A recombinant lentiviral gene transfer system according to one of the claims 1 to 3, characterized in that, the virai envelope protein Env encoded by the fifth plasmid is chosen from the group vesicular stomatitis virus envelope glycoprotein (VSVG), amphotropic MuLV envelope (4070A-Env), Ebola envelope, gp64 envelope glycoprotein from baculovirus, influenza A virus Haemagglutinin (HA), Jaagsiekte sheep retrovirus (JSRV) envelope, Lymphocytic choriomeningitis virus (LCMV) glycoprotein (LCMV-GP), Mokola virus envelope glycoprotein (MK-G), murine leukemia virus envelope, rabies-virus envelope glycoprotein and RD114 retrovirus envelope (RD114-Env).

5. A recombinant lentiviral gene transfer system according to one of the claims 1 to 4, characterized in that, the self-splicing self-splizing intron ribozyme (class I intron) is a Tetrahymena 26S rDNA intron.

6. A recombinant lentiviral gene transfer system according to claim 5, characterized in that, the Tetrahymena 26S rDNA intron is from the organisms chosen from Tetrahymena hyperangularis, Tetrahymena sonneborni, Tetrahymena cosmopolitans, Tetrahymena pigmentosa and Tetrahymena thermophylia.

7. A recombinant lentiviral gene transfer system according to one of the claims 1 to 6, characterized in that, the ribozyme cassette further comprises a guide sequence complementary to a part of the target RNA or a stuffer sequence flanked by restriction sites.

8. A recombinant lentiviral gene transfer system according to one of the claims 1 to 7, characterized in that, the nucleic acid sequence encoding the VLV protein Gag is under the control of the Cytomegolo virus (CMV) promoter and the nucleic acid sequences encoding the VLV proteins Pol, Vif, Rev and Tat Visna pol, vif, rev, tat and the envelope protein Env are under control of the Simian Virus 40 (SV40) promoter.

9. A recombinant lentiviral gene transfer system according to one of the claims 1 to 8, comprising the plasmids with the sequences according to SEQJD . No. 1 to 4 and a transfer plasmid chosen from the sequences according to SEQJD . No. 5 to 10.

10. The plasmids of the lentiviral gene transfer system for ribozyme mediated repair of a target RNA with the sequences according to SEQJD . No. 1 to 10.

11. The nucleic acid sequences encoding the VLV proteins Gag, Pol, Vif, Rev and Tat of the lentiviral gene transfer system for ribozyme mediated repair of a target RNA comprising the synthetic non-naturally occurring nucleotide-sequences chosen from SEQJD . No. 11 to 15.

12. A method of producing replication-defective virion vector particles, comprising transfecting producer cells with a recombinant lentivirai gene transfer system according to one of the claims 1 to 9 wherein the cells produce replication-defective virion particles.

13. A method according to claim 12, wherein the producer cells are chosen from the cell lines 293 or 293T or 293ts/A1609.

14. Replication-defective virion vector particles obtained by the method according to claim 12 or 13.

15. A method of ribozyme mediated repair of a mutated target RNA or modification of a non-mutated RNA in eukaryotic cells, comprising the steps:

(a) cloning a heterologous gene or a part of a gene encoding for the accurate counterpart of the target RNA to be repaired (or to be modified) into the transfer plasmid of the lentiviral gene transfer system according to one of the claims 1 to 9;

(b) transfecting producer cells with the so produced transfer plasmid and the other plasmids of the lentiviral gene transfer system according to one of the claims 1 to 9;

(c) growing the producer cells under cell culture conditions sufficient to allow production of replication-defective visna virion particles in the cell and collecting the replication-defective visna virion particles from the producer cell;

(d) infection of eukaryotic cells with the replication-defective visna virion particles.

16. A Kit for ribozyme mediated repair of a mutated target RNA or modification of a non- mutated RNA in eukaryotic cells, comprising the recombinant lentiviral gene transfer system according to one of the claims 1 to 9, whereas the ribozyme cassette of the transfer plasmid contains a cloning site, wherein the heterologous gene or a part of a gene encoding for the accurate counterpart of the target RNA to be repaired (or modified) can be inserted and a stuffer sequence flanked by restriction sites, wherein the specific guide sequence can be inserted.