WO1997042336A1 - A long-term and high expression, single gene retroviral vector with applications to progenitor and mature haematopoietic cells - Google Patents

Abstract

The present invention provides a recombinant retroviral vector which is replication-defective and includes a long terminal repeat (LTR) sequence for expression of a heterologous polynucleotide sequence, wherein the vector does not include a heterologous promoter element or a heterologous gene which imparts antibiotic or other resistance. The retroviral vector may be used to introduce foreign genes into mammalian cells and can be used for therapeutic purposes.

Description

A long-term and high expression, single gene retroviral vector with applications to progenitor and mature haematopoietic cells

The present invention relates to a retroviral vector. The present invention also relates to a method of producing the vector and to a method of transferring foreign genes to target cells. Replication-defective retroviruses are frequently employed to transfer exogenous genes to target cells. In the murine system, retroviral gene transfer studies have aimed to investigate haematopoietic neoplasia (5,7,9,12,19,20,26) or model potential gene therapies (8,13,24,27,31,33,34). The success of these investigations has been vaπable, dependent upon properties of the retroviral vector and the target cells. In such model systems, a retroviral vector must efficiently infect immature bone marrow cells, then stably integrate in the host cell genome and be efficiently expressed in the infected cell and its progeny iii vivo (17,21,25). MoMLV- based retroviral vectors, such as LNL6 (3) and N2 (18) are most frequently employed in murine BM reconstitution experiments. These vectors are suited to these experimental systems because they include a portion of the viral gag gene which enables production of high-titres of virus. High viral titres appear to facilitate infection of long-term repopulating haematopoietic stem cells. However despite this advantage, there have been numerous reports of MoMLV-based retroviral vectors that do not function efficiently within the haematopoietic system of reconstituted mice, Several retroviral vectors that apparently function efficiently in haematopoietic cells in vitro, have been found to be transcriptionally repressed within haematopoietic cells of long-term repopulated mice (2,6,12.16,24,33). In human gene therapy clinical trials, LNL6 and its derivatives have been used extensively and are the vectors of choice (4).

We have previously described an LNL6-derived retroviral vector that was transcriptionally repressed in bone marrow repopulated mice (5). Interestingly, this vector spontaneously generated a recombinant provirus which, as a consequence of an internal deletion, was transcriptionally activated. Through a detailed structural comparison of the parental and deleted proviruses, it was shown that transcriptional repression of the original vector was mediated by exogenous sequence included in the parental vector (22). Specifically, a human γ- actin promoter and/or neomycin resistance gene (neo) appeared to be responsible for transcriptional suppression.

There has been a lack of evidence to date, however, to show that retroviral vectors can be stably integrated and reliably expressed in haematopoietic cells in vivo. Previous studies utilising replication-defective retroviruses as vectors for gene transfer have proven to be variable and preliminary data from clinical protocols has shown that existing retroviral vectors have only limited efficiency in transducing haematopoietic stem cells; high levels of marked progeny cells are not produced.

The present inventors have now designed and constructed a retroviral vector which is surprisingly efficient in that it can be stably integrated into, and reliably expressed within, target cells in-vivo.

Accordingly, in a first aspect the present invention provides a recombinant retroviral vector which is replication-defective and includes a long terminal repeat (LTR) sequence for expression of a heterologous polynucleotide sequence, wherein the vector does not include a heterologous promoter element or a heterologous gene which imparts antibiotic or other drug resistance. In a preferred embodiment of this aspect of the present invention the retroviral vector includes a packaging region and/or a viral gαg-encoding sequence or a portion thereof. Preferably, the vector includes a multiple cloning site. The vector may be the plasmid pLK as herein described.

A recombinant vector according to the first aspect of the present invention may be used to transfer foreign genes to target cells.

Accordingly, in a further preferred embodiment of the first aspect of the present invention the retroviral vector contains a heterologous gene. Preferably, the vector contains a single heterologous gene.

It will be understood that a retroviral vector according to the first aspect of the present invention may be used to study the effects of a gene of interest. For example, genes involved in tumorigenesis. such as activated N- ras. mutant p53. c-myc or mutants thereof, may be introduced into haematopoietic cells to investigate haematopoietic neoplasia. Alternatively, the retroviral vector may be used for gene transfer to human cells for therapeutic purposes, for example by incorporating tumour suppressor genes such as p53 or p21. or antisense to a cell survival gene such as bcl-2. The present invention also provides mammalian cells or cell lines which have been transfected with a retroviral vector according to the first aspect of the present invention. These are called producer cells as they have the ability to produce replication-incompetent infectious retrovirus particles. In a second aspect, the present invention thus provides a method of producing a retroviral vector which method includes transfecting a packaging cell line with a retroviral vector according to the first aspect of the present invention in cDNA form; culturing the transfected cells; collecting the viral particles produced by the transfected cells; and using these to transduce target cells.

The present inventors have found that the novel vector integrates successfully into mammalian host cells and is efficiently expressed in the infected cell and its progeny in vivo. Accordingly, in a third aspect the present invention provides a method of transferring a foreign gene to a mammalian cell which method includes infecting the cell with a retroviral vector according to the first aspect of the present invention.

In a preferred embodiment of the third aspect of the present invention the mammal is selected from a mouse, rat or human.

In a further preferred embodiment of this aspect of the invention the cell is a haematopoietic cell. The haematopoietic cell may be a bone marrow or mobilised progenitor cell, including the pluripotential haematopoietic cells. As there is no selectable marker in the retroviral vector of the present invention, selection of transfected cells may be accomplished by introduction of an Internal Ribosome Entry Site (IRES) or a cDNA encoding a membrane protein allowing cell sorting. In another embodiment of the invention the mammalian cell is co-infected with a vector which contains a chemically different selectable marker.

In order that the present invention may be more clearly understood, preferred forms thereof will now be described with reference to the following examples in which:- FIGURE 1: Construction of LK. ψ - packaging region, solid boxes - LTRs, shaded boxes - gag.

FIGURE 2: PCR analysis of producer clones.

Lane 1 - pLK DNA as positive control. Lanes 2-4-DNA from producer clones, ψLK09. ψLKl3, and ψLK26 respectively,

Lane 5 - ψ-cre, Lane 6 - negative control, containing no template

FIGURE 3: Northern analysis of producer cell lines. Lane 1 - vμ-cre. negative control,

Lane 2 - ψLK09, Lane 3 - ψLKl3, Lane 4 - ψ2AV, positive control.

FIGURE 4: PCR analysis of Ratl cells infected with LK09.

Lane M - molecular weight markers

Lane 1 - ψLK09, positive control.

Lane 2 - LK09 infected Rat 1 cells.

Lane 3 - parental Rat 1 cells, Lane 4 - negative control.

FIGURE 5: Infection ability of LK in bone marrow cells. Lane M - molecular weight markers Lane 1 - negative control, Lane 2 - positive control (ψ2 AV DNA).

Lanes 3-12-individual bone marrow colonies from LK-infected cultures.

FIGURE 6: PCR analysis of LK-infected FDCP-1 cultures. Lane M - molecular weight markers

Lane 1 - positive control (vμ2AV DNA), Lane 2 - negative control, no template

Lanes 3-18-individual FDCP-1 colonies from methylcellulose cultures.

FIGURE 7: PCR of individual colonies derived from reconstituted mice.

Lane M - molecular weight markers

Lane 1 - positive control (ψ2AV DNA).

Lanes 2-15-individual colonies.

Lane 16 -negative control, no template

FIGURE 8: New constructs derived from LK. ψ - packaging region, solid boxes - LTRs, shaded box - gag. genes inserted - N-ras, p53Ac5. v-myc2. wtp53 (not shown), lacz (not shown).

FIGURE 9: Southern analysis of bone marrow cells infected with LN-ras2. Lane 1 - negative control, no template Lane 2 - positive control (ψ2AV DNA).

Lanes 3-12-individual bone marrow colonies derived from LN- ras2 infected cultures.

FIGURE 10: Southern analysis of bone marrow cells from reconstituted mice. Lane 1 - positive control (tEh/hnyci cells), which is a v-mvc transformed myelomonocytic cell line (11)

Lanes 2-6,9. 10. 13, 15 and 17 - Lv-myc2 infected bone marrow cells,

Lanes 7, 8, 11, 12 and 14-LK-infected bone marrow cells as negative controls.

FIGURE 11: PCR of LN-rαs2 infected bone marrow cells from reconstituted mice.

Lane 1 - negative control, Lane 2 - positive control (v|/2AV DNA),

Lanes 3-6-bone marrow cells from LN-ras2 reconstituted mice. FIGURE 12: Northern analysis of bone marrow cells from LN-rαs2 reconstituted mice.

Lanes 1 and 2 - RNA from bone marrow and spleen, 15 weeks following reconstitution,

Lanes 3 and 4 - RNA from liver and spleen, 20 weeks following reconstitution,

Lane 5 - RNA from liver. 27 weeks after reconstitution,

Lane 6 - positive control (ψN-rαs215).

FIGURE 13: lacz proviral expression. a. X-gal staining in producer cell lines.

Magnification - objective x20, eyepiece xlO.

b. β-gal antibody staining in producer cell lines.

Magnification - objective x40, eyepiece xlO.

c. X-gal staining in donor bone marrow cells. Magnification - objective x40. eyepiece xlO.

d. X-gal staining in bone marrow cells of reconstituted mice.

Magnification - objective x20, eyepiece xlO.

MATERIALS AND METHODS

Cell culture techniques

NIH3T3 and Rat 1 cells were maintained in Dulbecco's modified Eagle's medium (DMEM) containing 10% heat inactivated FBS and 50U/ml penicillin/50_/txg/ml streptomycin. Cells were removed from the dishes with 0.05% trypsin/0.53mM EDTA.

Co-transfections were performed using lϋμg of retroviral vector plasmid and lμg of pMolneo in a standard calcium phosphate precipitation procedure (30). Transfectants were selected in lmg/ml G418 and cloned using cloning cylinders (Belco). To collect viral conditioned medium (VCM), 7xlO^h stable transfectants were seeded in 10ml of DMEM plus 10% FBS in a T75 flask. VCM was harvested 11-16 hours later by collecting the supernatant and clarifying either by filtration through a 0.45 micron filter unit or centrifugation at 3000xg for 10 min. VCM was stored in cryogenic vials at - 80°C. For retroviral infection, Ratl cells were seeded at a density of 2xl0⁵/60mm plate in DMEM plus 10% FBS. After a 16 hour incubation period, the medium was replaced with 4ml VCM. The VCM was supplemented with lOμg/ml polybrene. 20 to 24 hours later, the infected cells were split 1:4 into DMEM plus 10% FBS.

To check for the potential presence of replication competent retrovirus (RCR), a marker rescue assay was performed as follows. VCM was collected from producer cell lines as described above. 2xlO^r> LN3T3 cells (NIH3T3 cells transduced with LNL6) were seeded in 60mm plates in VCM with 8μg/ml polybrene and incubated overnight. The following day, the infection medium was replaced with 4ml DMEM plus 10% FBS and 0.8mg/ml G418. Following passaging in this medium for two weeks, the infected LN3T3 cells (LN3T3IN) were split to 80% confluence. The next day, the LN3T3IN medium was replaced with 2ml of fresh DMEM plus 10% FBS with no added G418 and. NIH3T3 cells were seeded at 5xl0^r> cells/60mm plate. 16 hours later, the medium from the LN3T3IN cells was harvested and added to the NIH3T3 cultures with 8μg/ml polybrene. After a 3 hour incubation period, the NIH3T3 cells were split 1:4 into DMEM plus 10%FBS and 0.5%mg/ml G418. The NIH3T3 cultures were scored for G418 resistant colonies 10 days later.

Oligon u cleotides

Primers for PCR analysis: 1. 5 ^VTGTACACCCTAAGCCTCCGC 3 ^* (MoMLV gag)

2. δ ^{^}TGGCCAGCAACTTATCTGTGT 3 ^v (ψ region MoMLV/MSV)

3. 5 ^%TCTTGACATCTACCGACTGG 3 ^{^} (MoMLV gag)

4. 5^VGGACCACTGATATCCTGTCT 3^{^} (MoLV env)

5. 5^VCAAAAATTCAGACGGAGGCG 3 ^v (ψ region MoMLV/MSV) 6 5 ^v TGGCCAGCAACTTATCTGTGT 3 ^* (ψ region MoMLV/MSV)

7: TGT ACA CCC TAA GCC TCC GC (gag(14)) 8: GATGTAGAGCGGGCCTTTGAG(hp21) 9: TTACCAGGGCAACTATGGCT(murine p53) 10: CCGCCAAGAGGCTAAAGTTGG (5VM5) 11: GTGCTCGTCCGATTGGAGAGA (3VM6)

Bone marrow reconstitution:

Bone marrow cells were harvested from the femurs of 8-week old female Balb/C mice that had been treated 4 days previously with 5-flurouracil (5-FU, 150 mg/kg). 5FU destroys rapidly dividing cells, thereby creating a stem cell enriched bone marrow. Bone marrow cells were incubated for 48h with viral supernatant freshly harvested from 80-90% confluent plates of viral producer cells. The infection was conducted in Iscove's Modified Dulbecco's Medium (IMDM) supplemented with lmg/ml bovine serum albumin (Sigma), 30μg/ml transferrin (Sigma), 8μg/ml polybrene (Sigma). 20% WEHI-cell conditioned medium (WCM) as a source of IL-3 (Lee, et al, 1982). lOng/ml of recombinant stem cell factor (SCF), lOng/ml of recombinant IL-6 (Genzyme), 2mM glutamine, penicillin (50 U/litre), streptomycin (50mg/litre) and 20% foetal bovine serum (FBS). Cells were then washed once in IMDM supplemented with 2% FBS and resuspended in IMDM. Following lethal irradiation, recipient mice were injected intravenously with 1.5xl0^r> (lower dose) - 2xl0^e (higher dose) cells.

Bone marrow cultures

Following retroviral infection (as per bone marrow reconstitution), bone marrow cells were cultured in RPMI1640 (GIBCO) supplemented with 0.2M glutamine, penicillin (100u/l), streptomycin (50mg/l), 10% FBS and 20% WCM. Cells were fed at weekly intervals by a half exchange of medium, non-adherent cells being re-added following centrifugation.

Isolation of viral RNA and determination of viral titre

For isolation of viral RNA. viral particles were precipitated from 900μl of clarified VCM by the addition of 225μl of 40% PEG and 128μl NaCl followed by incubation on ice for 1 hour. After centrifugation at 13.000xg for 10 min, the pelleted viral particles were resuspended in 250μl Rnase free TE, pH 8.0. The suspension was then extracted once with phenol/ chloroform/isoamyl alcohol (24:24:1) and then left on ice uncapped for 10 min to allow residual organic solvents to evaporate. Generally, viral RNA samples were extracted in triplicate. Viral RNA samples were applied to Zeta probe nylon membranes (Biorad) using a clot blot apparatus (Biorad) according to the manufacturer's instructions. Membranes were pre- hybridised in hybridisation solution containing 10% dextran sulphate, 0.5% blotto, 2xSSPE, 1% SDS for 1 to 6 hr at 65"C. Hybridisation was carried out for 6 to 24 hr at 65^ϋC in hybridisation solution containing a α^"3Zp [dCTP]- labelled probe. After hybridisation, the dot blots were rinsed once in 2xSSC then washed once in 2xSSC/0.1% SDS for 15 min at room temperature, once in 0.5xSSC/0.1%SDS at 65°C for 15 min. then finally rinsed in O.δxSSC. Viral dot blots were quantitated by densitometry. Autoradiography was carried out at -70°C.

Transduction Efficiency Clonogenic Assay

As LK contains no selectable marker. PCR techniques were employed in order to determine the transduction efficiency. Non-adherent cells (5xl0⁴ cells pre ml) were added to 0.8% methylcellulose with 25% horse serum, 20% WCM in DMEM. Individual colonies were picked after 7 days, lysed and PCR used to determine the efficiency of viral infection of progenitor cells,

Analysis of proviral integration and gene expression in the haematopoietic tissues of reconstituted animals.

Genomic DNA and cellular RNA were isolated from frozen spleens and processed for probing as previously described (5). Blots were probed with a ^,2P-labelled cDNA ψ-specific probe.

Expression of proviral DNA was determined by reverse transcriptase- based PCR (RT-PCR). RNA samples (lμg) were incubated for 40 min at 37°C in 40mM Tris (pH7.5), 1.5mM MgCl₂, and lOmM NaCl containing 2U of RNase-free Dnase (Promega). Samples were then boiled for 15min. cDNA syntheses was performed in a 20μl of reaction mixture containing 50mM Tris (pH8.3), 50mM KCl, 5mM MgCl₂, lOmM DTT, 0.5mM of each dNTP, 40U of RNasin (Promega), lOOu of MoMLV RTase (Promega) and O.lμM specific 3 - primer. Reactions were incubated at 37°C for 40 min, then 94°C for 3 min to inactivate the enzyme. This reaction mixture was used for PCR- amplification of cDNA. Reactions were performed in lOOμl of reaction mixture containing O.luM of 3'-and 5' specific primers, 0.1 mM of each dNTP, 1.5mM MgCl₂ and 0.5U of Taq-polymerase. The thermal cycling program used was dependent upon the set of primers used. PCR reactions with the RT-step omitted were also performed as a control for possible DNA amplification. In some experiment PCR products resolved on a 2% agarose gel were transferred onto Hybond N membrane and probed with 32p-labelled probe, specific to the gene of interest.

Cloning procedures

To prepare cDNAs for cloning into the new LK vector (and for use as probes), restriction enzyme (Promega) digested DNA fragments were purified from agarose gels using the Gene Clean Kit (Bio inc. 101). Where necessary, cDNAs were blunt-ended with 5 units of T4DNA polymerase (Bresatec) according to the manufacturer's instructions. To prevent self ligation, 100- 500ng of linearised plasmid DNA was de-phosphorylated in a 20μl reaction with 2 units of CIP (New England Biolabs) in the manufacturer supplied buffer at 50"C for 60 min. The enzyme was inactivated by heating to 75°C for 5 min. Litigations were carried out by incubating a 10 fold molar excess of purified cDNAs at 22°C overnight with 100-300ng of plasmid DNA in a buffer containing 40mM Tris.HCl pH7.5, lOmM MgCl₂, lOmM DTT. 0.5mM ATP, 50μg/ml BSA and 5 units T4 DNA ligase (BRL). Ligation mixtures were electroporated into competent E. coli which were prepared according to standard procedures (30). Plasmid DNA was extracted from bacteria as described (30).

To screen for plasmids within bacterial colonies. PCR was performed directly on bacterial lysates. Individual bacterial colonies were lysed by heating at 95°C for 10 min in a 50μl solution of 1% triton-X 100, lOmM Tris.HCl pH7.5, ImM EDTA. The lysate was cleared by pelleting in a microfuge. 5μl of the cleared lysate was used in a 50μl PCR reaction. Immunocytological staining

Cytospin preparations were made and cells were fixed with 2.5% paraformaldehyde and dehydrated with ethanol. A mouse monoclonal antibody to β-galactosidase (Zymed) was used to visualise β-gal expression, with the isotype immunoglobulin used as the negative control. SIH Universal KIT (Sigma Diagnostics) was used for the detection of primary antibody binding in which 3-amino-9-ethylcarbozole was used as the substrate for horseradish peroxidase.

β-galactosidase staining

Cells were incubated overnight in a lmg/ml solution of X-gal substrate (5-bromo-4-chloro-3-indolyl-^β-D-galactopyranoside) with 4mM potassium ferricyanide. 4mM potassium ferrocyanide and 2mM magnesium chloride. Cytospins were made and blue cells were scored microscopically.

RESULTS

CONSTRUCTION OF THE LK RETROVIRAL VECTOR

The plasmid pLK was constructed by replacing the complete neo gene, as well as the following sequences between neo and the Hindlll site: FCV (env C) pol region, some unknown intervening sequence of approx. 20 base pairs and part of the mos sequence of MSV (GENBANK LNL6 SEQUENCE), within pLNL6 (3) with a polylinker

(5 ^v AATTCAGATCTGAGCTCCCCGGGGA TCCTCTAGAA 3 η. neo was removed from pLNL6 by digesting pLNL6 partially with EcoRI and to completion with Hindlll. To create the polylinker. two complementary 35bp oligonucleotides containing recognition sites for EcoRI. Bglll, Sacl, Smal. BamHl, Xbal and Hindlll were synthesised and annealed. This polylinker was ligated to the 4.7 kbp EcoRl/Hindll l fragment of pLNL6. To confirm that the polylinker was successfully inserted, plasmid DNA from transformed bacterial colonies was sequenced with primer 1 and analysed by restriction enzyme digestion. The resultant pLK plasmid. Figure 1, is 4.7kbp and contains a multiple cloning site and all features of pLNL6 (3) except the neo gene.

ESTABLISHMENT AND CHARACTERISATION OF THE LK VIRAL PRODUCER CELL LINE

In order to establish a cell line that produces a high-titre of the new retroviral vector LK. the plasmid pLK. was co-transfected, together with pMolπeσ into ψ-cre packaging cells (10). Since the pMolneσ vector encodes neo, transfectants that express this vector are G418 resistant. For the transfection, the molar ratio of retroviral vector to pMolneo was 10 to 1. Hence, transfectants that express neo were also expected to harbour the co¬ transfected retroviral vector. Transfectants were selected in G418 (lmg/ml) and assayed for viral titre by viral RNA dot blot analysis. The dot blots were hybridised to a ψ region probe. The ψ probe is a 505bp PCR product,

(primers 2 and 3) that extends from the viral splice donor site of pLK into gag. The ψ probe hybridises to nucleic acid from pLK- and pLNL6-derived virus. Viral RNA from the pLK transfectants were compared with dilutions of RNA prepared from the supernatant of the ψ2AV viral producer cell line (5). Since this vector contains neo. the titre of the v|/2AV viral producer cell line could be standardised by determining the ability of this cell line to confer G418 resistance to NIH3T3 cells. The titre of the ψ2AV producer cell line was thereby determined to be 2x10 ' cfu/ml. The results from the RNA comparative analyses indicated a number of clones that were potentially useful as high-titre LK viral producers (data not shown). The cells that were transfected with pLK are referred to as ψLK. The numbers that follow distinguish different clones.

PCR analysis was performed to check proviral integrity within the producer clones. Results are shown in Figure 2. For these PCR analyses, primers 2 and 4 (see Materials and Methods) were used to amplify the LK proviral DNA. Fragments of the expected size were amplified from all clones which were identified as positive for viral RNA by dot blot analysis. Clones such as ψLK26, which by viral dot blot did not appear to be a high-titre viral producer, were found to be negative for viral RNA by dot blot analysis. In addition to the expected LK fragment, PCR with primers 2 and 4 also amplified a larger fragment from these pLK transfectants and from ψ-cre cells. This fragment was probably amplified from the endogenous ψ-cre helper provirus.

Northern analysis was performed to assess proviral transcription in the viral producer cell lines. The size of the transcript expected to be produced by the LK proviruses is 1.7kb. Results are shown in Figure 3.

Hybridisation of Northern blots to the ψ probe revealed full-length provirus- derived transcripts in ψLK09 and ψLKl3. In selecting the most useful viral producer, the data from the Southern analysis. Northern analysis and viral

RNA dot blots were taken into consideration. ψLK09 was considered the most useful clone for gene transfer experiments, with the titre estimated to be 10 'cfu/ml. This clone expressed full-length transcripts from an intact provirus.

The ψLK09 cell line was assayed for RCR using the marker rescue assay. In this test, MLV3T3 cells (32) were used as a positive control. No RCR was detected in undiluted supernatant from any of the viral producers.

ABILITY OF VIRUS TO INFECT REPLICATING CELLS

To test that virus produced by the selected clone was infectious, supernatant collected from ψLK09 was used to infect Ratl cells. DNA isolated from pooled populations of infected cells was assessed for proviral integrants by PCR analysis. An agarose gel showing PCR products from these analyses is shown in Figure 4. Primers 2 and 4 (see materials and methods) were used for the amplification of the LK provirus. These results demonstrate that the provirus is present in the infected Ratl cells indicating that the retroviral vector was able to infect replicating cells. A PCR product was also amplified from the parental Ratl cells with primers 2 and 4 however this was larger that the LK PCR product. The size difference between this fragment and the LK PCR product was more evident when the DNA was run further through the gel. This larger band may represent an endogenous provirus in Ratl cells. ABILITY OF LK TO SUCCESSFULLY INFECT, INTEGRATE AND EXPRESS INTERNAL GENES IN HAEMATOPOIETIC CELLS

In order to determine the ability of LK to integrate and express successfully, an in-vivo murine model designed to infect primitive murine haematopoietic stem cells was used. This work was backed up with in-vitro work.

In several independent experiments, the ψLK09 producer cells were used to infect murine bone marrow haematopoietic cells. Using the transduction efficiency assay described in the Materials and Methods, the percentage of infected cells was determined to be 75-100%. Figure 5 shows an example of one such analysis in which LK proviruses were detected in 100% of clonogenic bone marrow progenitor cells. A similar transduction efficiency was determined for LK-infected FDCP-1 cells. Primers 5 and 6 producing a 290bp product were used.

Comparisons between infected and uninfected FDCP-1 cells demonstrated that LK conferred no biological effect on the cells with no differences being observed in cell morphology, viability or growth rates,

In-vivo reconstitution experiments have demonstrated a transduction efficiency of up to 86% in murine haematopoietic bone marrow cells as can be seen by PCR of individual bone marrow colonies from methylcellulose with primers 5 and 6 in Figure 7.

LK proviral integration has been demonstrated in reconstituted animal haematopoietic tissues up to 13 months following reconstitution while expression of N-rαs reconstituted mice can be demonstrated in 100% of mice with proviral integration at 7 months post reconstitution.

GENE TRANSFER ABILITY OF LK

In order to determine the ability of LK to transfer a foreign gene, several constructs were produced based on LK; N-ras, v-myc, p53-wild type and p53- mutant, lacz and p2lWAFl cDNAs were inserted into pLK and six new vectors produced. These constructs were then utilised in the in-vivo murine model as well as in in-vitro infections and in certain cases, transfection. PRODUCTION OF SINGLE GENE VECTORS FROM pLK

N-ras: pLN-rαs2 was constructed by insertion of an N-rαs 650bp cDNA into the polylinker of pLK. This 650bp fragment was blunt-ended and sub-cloned into the BamHl site of pLK which was also bunt-ended and dephosphorylated. The integrity and orientation of the resultant pLN-rαs2 plasmid was checked by PCR analysis and restriction enzyme digestion. p53 wild type: A 1.2kbp wild-type p53 cDNA (entire coding region) was excised from pLSVNc9 (14) by digestion with Smal and BgUl. The pLK vector backbone was prepared by linearising with BamHl. then blunt-ending and digesting with Bglll. The p53 fragment was directionally cloned into the prepared pLK backbone to produce pLp53wt. Bacterial colonies were screened for plasmids containing the p53 cDNA by PCR analysis and the integrity confirmed by restriction enzyme analysis. p53 mutant: For the construction of a p53 mutant-containing vector, termed pLp53Ac5, a 950bp Kspl/Xhol cDNA fragment of the p53 cDNA within pLp53wt was replaced with a corresponding fragment of p53 cDNA which included a codon 135 mutation. This fragment was then inserted into pLK. Transformed bacterial colonies were screened for p53 containing plasmids by PCR. Apart form the point mutation at codon 135. the structure of the pLp53Ac5 plasmid is identical to pLp53wt. v-myc: phv-myc2 was constructed by the insertion of a Hindlll/Xho 1 1.7kbp v-myc cDNA fragment derived form PAB ( 1) into Hindlll linearised pLK. The transformed colonies were screened for pLv-myc2 plasmids by miniprep and restriction enzyme analysis to check the integrity and orientation of the inserted fragment. The new constructs are shown in Figure 8. lacz: lacz cDNA from pSVβ containing a 3.47 kbp bacterial β- galactosidase gene was cut with Notl. blunt-ended and ligated to BamHl linkers. This was inserted into the BamHl site of pLK. pLWAFl: pLWAFlwas constructed by the insertion of the 2.lkb full length cDNA Notl fragment excised from the plasmid pZLWAFl (35) into the Hindlll linearised pLK. Transformed bacterial colonies were screened for p2lWAFl containing plasmids by PCR analysis and the integrity confirmed by restriction enzyme analysis and sequencing. VIRAL PRODUCER CELL LINES FOR NEW VECTORS

To establish cell lines that produce high titres of the new retroviruses, pLN-rαs-2, pLp53wt and pLp53Ac5 were each co-transfected, along with pMolneo into ψ-cre packaging cells. Transfectants were selected in G418 (500μg/ml). pLv-myc2 was co- transfected with pMohieo into the ψ2 packaging cell line and also selected in G418. pLWAFl was co-transfected with pMohieo into Ψ-cre packaging cells. Transfectants were selected in G418 (800μg/ml). Individual colonies were expanded and virus-producing clones were identified by viral dot blot analysis. PCR analysis was performed to check the proviral integrity within the selected producer clones, and fragments of the expected size were amplified from all of the clones. Northern analysis was performed to assess proviral transcription in the viral producer cell lines and revealed full-length provirus derived transcripts in all selected cones, which are referred to as ψN-rαs215, ψp53wt06, ψp53Ac511 and ψLv-jrryc2, ψLlacz88 and ψLWAFl. The relative viral titre produced by individual clones was estimated by comparison to the amount of virus produced by ψ2AV on viral dot blots, probing with a ³²P- labelled cDNA ψ-specific probe. The clones of highest titre were further examined by analysing their ability to transform Rat-1 cells in-vitro (as described for ψLK09 in material and methods), The ψ2AV viral producer, used previously in bone marrow reconstitution studies (5) was used as a control. All six viral producers. ψN-rαs215, ψp53wt06, ψp53Ac511, ψLv- myc2. ψLlaczδδ and ψLWAFl were assayed for any potential contaminating RCR using the marker rescue assay. No RCR was detected.

ABILITY OF THE NEW SINGLE GENE VIRUSES TO INFECT FDCP-1 AND BONE MARROW CELLS

In-vitro: In-vitro infections of bone marrow cells and FDCP-1 cells were performed.

Table 1. In Vitro Transduction Efficiencies for LV-myc2, LN-ras2, Lp53Ac5 and Llacz88"

VIRAL VECTOR % PROVIRUS % PROVIRAL PROVIRAL POSITIVE INTEGRATION EXPRESSION* (bone marrow (FDCP- 1 cells) colonies)

Lv-myc2 80% ND positive

LN-ras2 80% 95% positive

Lp53Ac5 42% 50% positive

Llaczδδ ND 22% positive

" This table is a composite of experiments performed at various time points

' Proviral expression for the in-vitro assays was determined by RT- PCR of bulk culture lysates of bone marrow and FDCP-1 cells.

These in-vitro results are illustrated in Figure 9 showing one example in which there was 60% infection of colony forming unit bone marrow cells by LN-ras2 as detected by Southern analysis. Overall, however, as shown in Table 1, 80% efficiency was observed.

In-vivo: Three of the new constructs made by inserting the genes of interest into the LK vector were also used to infect bone marrow cells utilising the in-vivo murine reconstitution model (5.22,26). In three independent experiments, donor bone marrow cells infected with one of the three new vectors (LNras-2, Lvmyc-2, Lp53Ac5) were transplanted into myeloablated mice. Table 2 below summarises the results of the integration and expression for these three constructs in reconstituted mice. Integration was determined by the detection of provirus by PCR of bone marrow and spleen cells from reconstituted mice. Expression of the gene was determined by RT-PCR of RNA isolated from tissues (bone marrow or spleen) of reconstituted mice.

Table 2. In Vivo Reconstitution Efficiencies for Lv-myc2, LN-ras-2 and Lp53Ac5

VECTOR % PROVIRUS % PROVIRAL

POSITIVE EXPRESSION"

ANIMALS"

Lv-mvc2 100% 100% LN-rαs2 100% 56%

Lp53Ac5 82% ND " Determined in bone marrow and spleen

These results are further demonstrated by Southern analysis in Figure 10 showing 100% transduction of bone marrow cells by Lv-myc2.

Figure 11 demonstrates 100% reconstitution efficiency of LN-ras2 in bone marrow cells of reconstituted mice after 13 months, by PCR with primers 5 and 6.

The 2.3kb transcript in the Northern blot in Figure 12 demonstrates the long-term expression of LN-ras2.

IN VTVO STUDY OF THE TRANSDUCTION EFFICIENCY OF MURINE LIAEMATOPOIETIC STEM CELL (HSC) INFECTION

Transduction efficiency of HSC infection was analysed in in vivo spleen colony formation assay: 3 derivatives of the LK- virus, namely Llacz, hv-myc-2 and Lp53wt were used to infect bone marrow cells derived from 5- FU-treated donor mice. Lethally-irradiated mice were injected with 2 x 10⁵ mononuclear bone marrow cells previously shown to produce up to 10 colony forming units - spleen on day 12 (CFU-S), a time period optimal for physical isolation of individual CFU-S, CFU-S were individually analysed for proviral sequences following microsurgical dissection using a magnifying lens. Colonies were lysed in PCR-lysis buffer and genomic DNA isolated from individual CFU-S by phenol-chloroform extraction and analysed by PCR using provirus-specific primers: Lv-myc - primers 10 and 11; Llacz - primers 2 and 4; Lp53wt - primers 2 and 9.

Table 3. CFU-S Analysis for Llacz, Lv-myc2 and Lp53wt.

virus % of provirus positive CFU-S

LlacZ 100% (n= 7) *

Lv-myc-2 60% (n= 10)

Lp53wt 40% (n= 15)

Lv-myc2:Lp53wt 1:1 70% (n= 7) (dual infection)

*n - number of individual CFU-S analysed by PCR

Conclusion: Transduction efficiency determined by PCR analysis of in vivo spleen derived colony forming units (CFU-S) correlates with the transduction efficiency of in vitro CFUs in methylcellulose cultures derived from the freshly infected bone marrow cells (Table 1).

ABILITY OF THE NEW SINGLE GENE VIRUSES TO INFECT MURINE BONE MARROW CELLS

In vitro infections of post 5-FU murine bone marrow cells were performed using viral conditioned medium from ΨLWAFl:

Bulk cells - Provirus positive as determined by PCR with primers specific to the packaging site (2 and 4).

Individual colonies - 100% provirus positive (6 colonies analysed) with primers specific to the WAFl insert (primers 7 and 8) to produce a 785bp product. Conditions: 94° - 3'; 94° - 1'. 72° - 2' x 35 cycles, 72° - 10'. wtp53

Genomic DNA samples isolated from the spleens of 3 animals reconstituted with bone marrow cells infected with wtp53 were found to be provirus-positive four months following reconstitution. Standard mouse reconstitution conditions were used. PCR was performed with primers specific to the wtp53 insert (2 and 9) to produce a 1480bp product which was found in all 3 animals.

LK-DERΓVATΓVE CARRYING THE P-GAL REPORTER GENE

This is a valuable tool for marking donor derived bone marrow cells in a reconstitution model.

The murine bone marrow reconstitution model was used to determine whether donor derived and reporter gene marked bone marrow cells could be detected among the multilineage progeny of infected haematopoietic stem cells. This is an apparently benign gene that should confer no growth selective advantage or disadvantage to the cells. Donor bone marrow cells were infected with viral conditioned medium derived from ΨLlacz88 which have been shown to efficiently express β-gal ( new Fig 13a and b). X-gal staining performed in situ revealed up to 14%> of β-gal expressing bone marrow cells in bulk as opposed to previous clonogenic assays (Fig 13c). The transduction efficiency of primary bone marrow cells was somewhat lower as compared to that of myeloid progenitor FDCP-1 cell (> 20%) and lymphoid progenitor LBRM-TG6 (25%) cell lines. This is consistent with the proliferative indices in the 3 cell populations, ie. the rate of cell growth is an important factor for retroviral infection and integration. Two lethally-irradiated mice reconstituted with bone marrow cells infected with viral conditioned medium derived from ΨLlaczδδ were found to express β-gal in their bone marrow cells 16 weeks following reconstitution (Fig id). The presence of the virus in the bone marrow cells did not cause any detrimental effect on haematopoiesis. β-gal expression was found in both mice, with a high number of positive cells (Fig Id). Both the myeloid (bone marrow) and the lymphoid (spleen) compartments were found to be provirus positive, indicating that pluripotent early progenitor cells were efficiently infected with the Llacz virus. LWAFl, CONTAINING p2lWAFl IN LK. INHIBITS GROWTH OF LEUKAEMIC CELLS

Transfection of LWAFl into leukaemic cell line tEMmyc4 (11) caused a loss of clonogenic cell growth relative to LK. Such clones were isolated with G418 (800 μg/ml) following co-transfection of LWAFl with an MoMLV LTR neo construct. Results are shown in Table 4 where in both experiments LWAFl caused a statistically significant drop in clonogenicity of tEMmyc4 cells.

Table 4. p21 Expression in LWAFl inhibits transformed cell growth in semi- solid cultures.

Experiments 1 and 2 summarise the data of 2 clonogenic assays for each experiment. In these experiments, colonies are derived from 2 independent transfections of the cDNA construct into a murine leukemic cell line termed tEMmyc4 (11). Each sample in the clonogenic assay was performed in triplicate, Statistical analysis was performed using the Student's l-test.

DISCUSSION The LK construct was designed for efficiency in haematopoietic cells in vivo. Hence, the retroviral vector was designed to be as structurally simple as possible and therefore includes minimal exogenous sequence and only the necessary cis-acting retroviral sequence. This minimal approach was adopted after consideration of results form our previous investigation which demonstrated that internal exogenous sequence can contribute to vector instability and suppress retroviral expression in vivo. The main disadvantage of single gene retroviral vectors is the absence of a selectable marker. Selectable markers are convenient for selection of transduced cells and are useful for determining viral titres. However, selectable markers, neo in particular, have been associated with repressed LTR-driven expression in various studies. For this reason, selectable markers were not included in the new constructs.

As there is no selectable marker in LK, PCR techniques on infected cells were employed in order to determine the transduction efficiency of the virus. The methylcellulose clonogenic assay employed for this purpose allows the formation of colonies that can be easily picked and used for PCR. Due to the high rate of renewal in the haematopoietic system, it is important to have a successful transfer to early progenitor cells in order to achieve long-term expression. Early progenitor colony formation is achieved by the addition of certain growth factors to the methylcellulose cultures, the use of PCR analysis allows the transduction efficiency of the virus in progenitor cells to be determined.

Results from previous studies utilising replication-defective retroviruses as vectors for gene transfer have been variable and preliminary data from clinical protocols has shown that existing retroviral vectors have only limited efficiency in transducing haematopoietic stem cells and yielding reasonable levels of marked progeny cells (4). The LK vector appears to eliminate these problems by minimising the amount of exogenous sequence within the vector, with the significant advantage of long-term expression in vivo in haematopoietic progenitor and mature cells. It is therefore an ideal vector for gene transfer in vivo. For the purposes of other gene transfer studies. LK could be used as a backbone for the construction of additional retroviral vectors to be used for studying the effects of a gene of interest in murine bone marrow studies in vivo. In particular. LK may have application in the use of tumour suppressor genes such as p53 or p21 or antisense to survival genes such as bcl-2 as potential therapeutics for leukaemia. As well as acting as a control LK may also function in the production of amphotrophic viruses for gene transfer to human cells.

Finally, it will be appreciated by persons skilled in the art that numerous variations and/or modifications may be made to the invention as shown in the specific embodiments without departing from the spirit or scope of the invention as broadly described. The present embodiments are, therefore, to be considered in all respects as illustrative and not restrictive.

REFERENCES

1. Alitalo. K., Ramsay, G„ Bishop, J.M.. Pfiefer. SO and Colby, WW (1983) Identification of nuclear proteins encoded by viral and cellular myc oncogenes. Nature. 306, 274-277

2. Apperley, J.F.. B.D. Luskey. D.A Williams. (1991) Retroviral gene transfer of human adenosine deaminase in murine haematopoietic cells: Effect of selectable marker sequences on long-term expression. Blood 78, 310-317.

3. Bender. M.A., T.D. Palmer, R.E Gelinas, A.D. Miller. (1987) Evidence that the packaging signal of Moloney murine leukaemia virus extends into the gag region. J. Virol. 61. 1939-1646.

4. Blaese. M., T. Blankenstein, M. Brenner. O. Cohen-Haguenauer, B. Gansbacher. S. Russell. B. Sorrentino and T. Velu. (1995) Vectors in cancer therapy: how will they deliver? Can. Gene Ther. 4. 291-297.

5, Bonham. L., K. MacKenzie, S. Wood. P.B Rowe. G. Symonds. (1992) Both myeloproliferative disease and leukemia are induced by transplantation of bone marrow cells expressing v-myc. Oncogene 7. 2219-2229.

6. Bowtell, D.L.. Cory. S., Johnson. G.R., Gonda. T.J. (1988) Comparison of expression in haematopoietic cells retroviral vectors carrying two genes. ).

Virol 62. 2464-2473.

7. Chung. S-W.. P.M.C. Wong, II. Durkin. Y-S. Wu. J. Petersen, (1991) Leukemia initiated by hemopoietic stem cells expressing the v-abl oncogene. Proc. Natl, Acad. Sci. USA 88, 1585-1589.

8. Correll. P.H., J.K, Fink, R.O. Brady, L.K. Perry, S. Karlsson. (1989) Production of human glucocerebrosidase in mice after retroviral gene transfer into multipotential hematopoietic progenitor cells. Proc. Natl. Acad. Sci. USA 86. 8912-8916. 9. Daley, G.Q., R.A. van Etten, D. Baltimore. (1990) Induction of chronic myelogenous leukemia in mice by the P210^bc,/"^bl gene of the Philadelphia chromosome. Science 247. 824-830.

10. Danos, O., R.C Mulligan. (1988) Safe and efficient generation of recombinant retroviruses with amphotropic and ecotropic host ranges. Proc. Natl. Acad. Sci. USA 85, 6460-6464.

11. Stapleton. P., Takayama, Y.. Hartshorn. A., Kalaizis, A.. Rowe, P.B. and Symonds, G. (1991) Tumor progression following transformation of murine monocytes by v-myc: acquisition of immortalization and tumorigenicity, Oncogene 6, 807-817.

12. Dunbar. C. E. Crosier, P.S., Nienhuis, A.W. (1991) Introduction of an activated ras oncogene into murine bone marrow lymphoid progenitors via retroviral gene transfer results in thymic lymphomas. Oncogene Res. 6, 39- 51.

13. Dzierzak, E.A., T. Papayannopoulou. R.C. Mulligan. (1988) Lineage- specific expression of a human β-globin gene in murine bone marrow transplant recipients reconstituted with retrovirus-transduced stem cells. Nature 331. 35-41.

14. Eliyahu, D.. N. Goldfinger, O. Pinhasi-Kimhi. G. Shaulsky, Y. Skurnik, N. Arai, V. Rotter. M. Oren. (1988) Meth A fibrosarcoma cells express two transforming mutant p53 species. Oncogene 3. 313-321.

16. Hoeben. R.C, Fallaux. F.J., van Tilburg. N.H.. Cramer. S.J, van Ormondt, H., Brϊel, E., van der Eb, AJ. (1993) Toward gene therapy for hemophilia A: Long-term persistence of factor Vlll-secreting fibroblasts after transplatation into immunodeficient mice. Hum. Gene Ther. 4, 179-186.

17. Karlsson, S. (1991) Treatment of genetic defects in hematopoietic cell function by gene transfer. Blood 78, 2481-2492. 18. Keller, G., C. Paige. E. Gilboa, E.F, Wagner. (1985) Expression of a foreign gene in myeloid cells derived from multipotent haematopoietic progenitor precursors. Nature 318, 149-154.

19. Keller. G., E.F. Wagner. (1989) Expression of v-src induces a myeloproliferative disease in bone-marrow reconstituted mice. Genes Dev. 3, 827-837.

20. Keliher. M.A., J. McLaughlin, O.N. Witte, N. Rosenberg. (1990) Induction of a chronic myelogenous leukemia-like syndrome in mice with v-abl and

BCRIABL. Proc. Natl. Acad. Sci. USA 87. 6649-6653.

21. Krauss. J.C. (1992) Hematopoietic stem cell replacement therapy. Biochim. Biophys. Acta 1114. 193-207.

22. MacKenzie, K.L.. L. Bonham, G. Symonds (1994) An internal deletion enhances transcriptional activity of a recombinant retrovirus in haematopoietic cells in vivo. J. Virol. 68, 6924-6932.

24. Mclvor, R.S., MJ. Johnson, A.D. Miller. S. Pitts, S.R. Williams, D. Valerio, D.W. Martin, I.M. Verma. (1987) Human purine nucleoside phosphorylase and adenosine deaminase: Gene transfer into cultured cells and murine haematopoietic stem cells by using recombinant amphotropic retroviruses. Mol. Cell. Biol. 7, 838-846.

25. Miller. A.D. (1992) Retroviral vectors. Curr. Top. Microbiol. Immunol. 158, 1-24.

26. Miller. M., G, Symonds. (1993) Induction of pre-B lymphoid leukemia following reconstitution of lethally irradiated mice with v-erb-B virus- infected bone marrow progenitor cells. Cell Growth diff. 4. 125-135.

27. Ohashi, T..S. Boggs, P. Robbins, A. Bahnson, K. Patrene, F-S Wei, J-F Wei, J. Li. L. Lucht. Y. Fei. S. Clark. M. Kimak, H. He, P. Mowery-Rushton, J.A. Barranger. (1992) Efficient transfer and sustained high expression of the human cerbrosidase gene in mice and their functional macrophages following transplatation of bone marrow transduced by a retroviral vector. Proc. Natl. Acad. Sci. USA 89, 1132-1136.

30. Sambrook, J., E.F. Fritsch, T. Maniatis. (1990) Molecular Cloning: A Laboratory Manual, Cold Spring Harbor: Cold Spring Harbour Laboratory Press.

31. Sorrentino, B.P., S.J. Brandt, D. Bodine, M, Gottesman, I. Pastan, A. Cline, A.W. Nienhuis. (1992) Selection of drug resistant bone marrow cells in vivo after retroviral transfer of human MDR1. Science 257. 99-101.

32. Sun. L-Q., D. Warrilow, L. Wang, C. Witherington. J. Macpherson, G. Symonds. (1994) Ribozyme-mediated suppression of Moloney murine leukemia virus and human immunodeficiency virus type I replication in premissive cell lines. Proc. Nail. Acad. Sci. USA 91, 9715-9719.

33. Williams. D.A., S.H. Orkin. R.C. Mulligan. ( 1986) Retrovirus-mediated transfer of human adenosine deaminase gene sequences into cells in culture and into murine haematopoietic cells in vivo. Proc. Natl. Acad. Sci. USA 83, 2566-2570.

34. Wilson. J.M.. O. Danos, M. Grossman. D.H. Raulet. R.C. Mulligan. (1990) Expression of human adenosine deaminase in mice reconstituted with retrovirus-transduced hematopoietic stem cells. Proc. Natl. Acad. Sci. USA 87, 439-443.

35. El-Deiry WS. Tokino T, Velculescu VE, Levy DB, Parsons R. Trent JM, Lin D, Mercer WE, Kinzler KW and Vogelstein B (1993). Cell. 75, 817-825

Claims

CLAIMS:-

1. A recombinant retroviral vector which is replication-defective and which includes a long terminal repeat (LTR) sequence for expression of a heterologous polynucleotide sequence, wherein the vector does not include a heterologous promoter element or a heterologous gene which imparts antibiotic or other drug resistance.

2. A recombinant retroviral vector according to claim 1 which includes sequences derived from Moloney murine leukemia virus (Mo-MLV),

3, A recombinant retroviral vector according to claim 1 or claim 2 which includes a packaging region and/or a viral gag-encoding sequence or a portion thereof.

4. A recombinant retroviral vector according to any one of claims 1 to 3 which includes a multiple cloning site.

5. A recombinant retroviral vector according to any one of claims 1 to 4 wherein the vector is LK.

6. A recombinant retroviral vector according to any one of claims 1 to 4 wherein the vector includes a heterologous gene.

7. A recombinant retroviral vector according to claim 6 wherein the heterologous gene is a gene involved in tumorigenesis or its reversion.

8. A recombinant retroviral vector according to claim 7 wherein the retroviral vector is selected from LN-ras2, Lp53wt, Lp53Ac5. Lv-myc2 and

LWAFl.

9. A mammalian cell or cell line which includes a retroviral vector according to any one of claims 1 to 8.

10, A method of introducing a foreign gene into a mammalian cell which method includes infecting the cell with a recombinant retroviral vector according to any one of claims 1 to 8.

11. A method according to claim 10 wherein the mammalian cell is a haematopoietic cell.

12. A method according to claim 11 wherein the haematopoietic cell is a haematopoietic progenitor cell.

13. A method according to any one of claims 10 to 12 wherein mammalian cell is derived from a mouse, rat or human.