WO1996014420A1

WO1996014420A1 - Inducible cell ablation

Info

Publication number: WO1996014420A1
Application number: PCT/GB1995/002596
Authority: WO
Inventors: John Clark; Thomas Connors; Barry Gusterson; Richard Knox
Original assignee: Cancer Research Campaign Technology Limited; Agricultural And Food Research Council
Priority date: 1994-11-04
Filing date: 1995-11-06
Publication date: 1996-05-17
Also published as: AU3812395A; GB9422264D0

Abstract

The present invention provides a method of producing a transgenic non-human animal, which method comprises incorporating into the genome of the non-human animal at least one nucleotide sequence (A) comprising a sequence (i) encoding a nitroreductase which is capable of converting a prodrug into a cytotoxic drug.

Description

INDUCIBLE CELL ABLATION

The present invention relates to inducible cell ablation in transgenic animals. Transgenic animals are primarily used as research tools and model systems for the investigation of the effects of existing and newly discovered drugs. Animals such as the "Oncomouse" can now be produced which are particularly susceptible (or resistant) to specific diseases (EP-B-0 169 672). The action of a drug against the specific disease may consequently be investigated more readily.

The same technology has been used with limited success in the field of inducible cell ablation. This technique allows an investigator to destroy a specific cell type within the whole animal. For example, the anterior pituitary gland produces GH, PRL, TSH and gonadotrophins from different specialised cells which are interspersed amongst each other. It is impossible to produce an experimental animal with a single hormonal deficiency by conventional surgery. This problem has been solved by transgenic ablation.

Transgenic mice have been produced carrying the gene encoding the diphtheria A toxin linked to the regulatory sequences from the growth hormone (GH) gene (Behringer et al, Genes Dev. 2: 453-461, 1988). In these animals the pituitary somatotrophs and lactotrophs are destroyed specifically but the other pituitary cells are unaffected. A major problem with this approach, however, is that the target cells are ablated as soon as the toxin is produced. This is determined by when the tissue-specific promoter is activated and, as such, is not directly under the investigator's control. A second problem is that the diphtheria toxin A chain is a very potent toxin: one polypeptide chain in the cell is thought to be sufficient for killing. Any leaky expression of the transgene in other tissues has deleterious effects, often lethality. Thus, with this approach the investigator has little or no control over the threshold for ablation. An alternative method that avoids these problems involves targeting an inducible toxic phenotype to a particular cell type. This is accomplished by using a gene encoding an enzyme that converts an inactive agent or pro-drug to a cellular toxin. Thus, when expression of the herpes virus type 1 thymidine kinase gene (HSVTK1) is targeted to the pituitary gland using the GH promoter, treatment of the mice with the inactive pro-drug FIAU results in specific ablation of the lactotrophs and somatotrophs. HSV-TK1 phosphorylates FIAU which, as a nucleotide analogue, is highly toxic to the cell. In these experiments the ablated cells regenerate when FIAU administration is discontinued.

The potential for inducible ablation is considerable. The approach requires that (a) the foreign enzyme is not toxic to the cell in the absence of the pro-drug, (b) that the pro-drug is neither toxic nor converted to an active form by endogenous enzymes, (c) that the foreign enzyme converts the prodrug to a toxic product and (d) that once produced the toxin should not affect nearby cells.

Although inducible ablation based on the use of HSVTK1 meets these basic criteria, there are a number of disadvantages associated with the approach. A major limitation is that, for some cell types, ablation requires cell division. This is because of the mechanism of cell killing. The phosphorylated products of the prodrugs, such as FIAU or ganciclovir, are nucleotide analogues. These are incorporated into cellular DNA during replication, causing chain termination and thereby killing the dividing cell. Thus FIAU-treated transgenic mice expressing HSVTK1 in the lactotropes of the pituitary have normal pituitaries; the lactotropes are not ablated because prolactin expression occurs post-mitotically (Borrelli et al. , Nature, 339: 538-541, 1989). Although the incorporation of nucleotide analogues into replicating DNA is thought to be the primary mechanism of HSVTK1-mediated ablation there may be some exceptions. Thus, thyroid follicle cells have been ablated using this approach in the absence of cell division (Wallace et al. , Endoc. 129: 3217-3226. 1991). In this case the mechanism of cell killing is not understood. There are also other disadvantages associated with the HSVTK1 approach. The prodrugs used in this approach (FIAU, ganciclovir) are very costly, discouraging their use in larger transgenic animals. Nor is it known whether these compounds will cross the blood brain barrier. If they do not, then it effectively rules out this approach for the ablation of specific neural types as a means of studying brain function. Large doses of active drug must be present for long periods in order to produce cytotoxicity and the active drugs may migrate and kill other cells (bystander effect). Finally, HSVTK1 reporter genes are expressed in the testis regardless of the nature of the promoter to which they are linked. This invariably causes male sterility, which effectively prohibits the establishment of breeding lines in mice.

The present invention addresses these problems by introducing into a mammalian germline a DNA sequence encoding a nitroreductase. The present invention provides a method of producing a transgenic non-human animal, which method comprises incorporating into the genome of the non-human animal at least one nucleotide sequence (A) comprising a sequence (i) encoding a nitroreductase (NR) which is capable of convening a prodrug into a cytotoxic drug. The present invention also provides a transgenic non-human animal whose germ cells and somatic cells comprise at least one nucleotide sequence (A), preferably as a result of incorporating into the animal genome or into the genome of an ancestor of said animal said nucleotide sequence, which at least one nucleotide sequence (A) comprises a sequence (i) encoding a nitroreductase which is capable of converting a prodrug into a cytotoxic drug.

The present invention further comprises a method of converting a prodrug into a cytotoxic drug, which method comprises exposing an animal according to the invention or produced by the method of the invention to a prodrug capable of being converted to a cytotoxic drug by the nitroreductase encoded by the sequence (i). Figure 1 shows an immunoblot of HB4a/R5P8 and HB4a/NR probed for

nitroreductase protein expression using the Rb 685/ntr antibody.

Figure 2 shows expression of nitroreductase in transfected breast epithelial cells. Figure 3 shows reduction of ³H-CB1954 by nitroreductase transfected 4 A breast cell line.

Figure 4 shows the production of plasmid pFG12.

Figure 5 shows the formation of DNA interstrand crosslinks in nitroreductase transfected breast cells.

Figures 6A and 6B show the construction of pBJ41.

Figures 7 shows an immunoblot of tissue extracts from transgenic mice carrying the nitroreductase gene.

The non-human animal according to the invention is generally avian, piscine, amphibian, arthropodal or mammalian, for example bovine, ovine, murine, feline, canine, simian, equine, porcine or lagomorphous. Preferably the animal is a rodent, such as a mouse, a rat, a hamster or a guinea pig, a rabbit, or a livestock animal such as cattle, a sheep, a goat or a pig.

According to the present invention, a nitroreductase is an enzyme, fragment or homologue thereof capable of reducing a nitro group in various compounds to the corresponding hydroxy lamino group.

The nucleotide sequence (i) encoding the nitroreductase preferably comprises the oligonucleotide of the sequence shown in SEQ ID NO: 1 , a fragment thereof or oligonucleotide hybridisable thereto. An oligonucleotide capable of hybridising to the oligonucleotide of SEQ ID NO: 1 or fragment thereof will generally be at least 70%, preferably at least 80 or 90% and more preferably at least 95% homologous to the oligonucleotide of SEQ ID NO: 1 or fragment thereof over a region of at least 20, preferably at least 30. for example 40, 60 or 100 or more contiguous

nucleotides. The sequence of the oligonucleotide may be varied by deleting at least one nucleotide, inserting at least one nucleotide or substituting at least one nucleotide in the sequence.

The oligonucleotides may be RNA or DNA. The oligonucleotide fragments typically will be at least 10, for example at least 20, 30, 40, 60 or 100 nucleotides long.

One, two, three or more nucleotide sequences (A) comprising a sequence encoding a nitroreductase may be incorporated into the genome of the animal. The nitroreductase encoded by the nucleotide sequence (i) is preferably bacterial nitroreductase, for example a nitroreductase which is a flavoprotein having a molecular weight in the range 20 to 60 kDa, which requires NADH or NAD(P)H or analogues thereof as a cofactor and which has a km for NADH or NAD(P)H in the range 1 to 100 μM, for example as described in EP-A-540 263. Typically the nitroreductase is the same as that from E.coli. Salmonella or Clostridia organisms.

Preferably the nitroreductase of the invention is a nitroreductase having the sequence of SEQ ID No: 2, a fragment thereof or homologue thereof. The sequence of the polypeptide may be varied by deleting, inserting or substituting at least one amino acid.

A nitroreductase of SEQ. ID No. 2 in substantially purified form will generally comprise the protein in a preparation in which more than 90%, eg. 95%, 98% or 99% of the protein in the preparation is that of the SEQ. ID No. 2.

A homologue of the SEQ. ID No. 2 will be generally at least 70%, preferably at least 80 or 90% and more preferably at least 95% homologous to the protein of SEQ. ID No. 2 over a region of at least 20, preferably at least 30, for instance 40, 60 or 100 or more contiguous amino acids. Generally, fragments of SEQ. ID No. 2 or its homologues will be at least 10, preferably at least 15, for example 20. 25, 30, 40, 50 or 60 amino acids in length.

The nucleotide sequence (i) encoding the nitroreductase is preferably operably linked to a regulatory sequence, such as a tissue specific promoter, which directs expression to a desired cell type. The nucleotide sequence (A) may therefore comprise nitroreductase encoding sequence (i) and optionally a regulatory sequence and/or non-coding sequences. "Operably linked" refers to a juxtaposition wherein the promoter and the

nitroreductase-coding sequence are in a relationship permitting the coding sequence to be expressed under the control of the promoter. Preferably the nitroreductase coding sequence is juxtaposed downstream in cis, that is at the 3' end of the promoter sequence. Additionally there may be elements such as non-coding sequence between the promoter and coding sequence which are not native to either the promoter or the coding sequence.

According to the one embodiment of the present invention the oligonucleotide of sequence SEQ ID No.3 encoding nitroreductase (NR) in plasmid pNRR8/3 is flanked at both the 5' and 3' ends by non-native sequences. At the 5' end the sequence AAGCTTTCACATTGAGTCATT directly precedes the NR coding sequence. This non-native sequence is introduced in a PCR cloning step and comprises

i) a Hindlll cleavage site to facilitate the manipulation of the NR DNA encoding segment, and

ii) a so-called "Kozak consensus" sequence thought to improve the translation of mRNA sequences in eukaryotic cells.

At the 3' end the sequence GGATCC directly follows the NR encoding sequences. This non-native sequence is introduced in a PCR cloning step and comprises a BamHI cleavage site to facilitate the manipulation of the NR DNA encoding segment. The positioning of the HindIII and BamHI sites defines the 5' - 3' orientation of the NR encoding sequences (see Figure 4).

Such non-coding sequences can be incorporated into the genome if they do not impair the correct control of the coding sequence by the promoter. Suitable promoters include tissue and tumour specific promoters, such as, for example, the promoter from a milk protein gene, the CEA gene promoter, or the CA-125 gene promoter. Promoters from milk protein genes include the β-lactoglobulin (BLG) promoter, preferably from sheep (see Fig. 4), the α-lactalbumin promoter and the whey acidic protein promoter. Promoters expressed in other cell types include the LHβ promoter which targets expression to the gonadotroph cells of the pituitary gland.

A single promoter may be selected which allows ablation of a single cell type or more than one cell type, for example 2. 3 or 4 cell types. Preferably a single promoter allows ablation of only one cell type. A cell type may be defined anatomically, biochemically, histologically or developmentally.

Although it is preferred to juxtapose to the NR encoding sequences (i) a native promoter sequence, for example a native mammalian sequence, modified promoter sequences which are capable of selectively hybridizing to the native sequence may be incorporated in the genome. A promoter sequence capable of selectively

hybridizing to the native promoter sequence will be generally at least 70% , preferably at least 80 or 90% and more preferably at least 95% homologous to the promoter region or fragment thereof over a region of at least 20, preferably at least 30, for instance 40, 60 or 100 or more contiguous nucleotides.

In general, those of skill in the art will appreciate that some regions of promoters will need to be retained to ensure tissue specificity of expression whereas other regions of the promoter may be modified or deleted without significant loss of specificity. The nucleotide sequence (A) comprising, for example, promoter, NR encoding sequences and any additional sequences may be incorporated into the genome by conventional methods, such as. for example, pronuclear injection or ballistic techniques or by the use of embryonic stem (ES) cells. Alternatively the NR encoding sequences may be incorporated into the genome using gene targeting methodologies. In this way the expression of the NR sequences would be under the control of endogenous sequences to which they have been targeted. According to this embodiment the NR encoding sequence is incorporated into an endogenous gene by means of a targeting vector and ES cells so that it is downstream of an

endogenous promoter sequence.

Pronuclear injection may involve injection into a pronucleus of a fertilised one cell egg. The fertilised egg may then be transferred into the oviducts of pseudo-pregnant foster mothers and allowed to develop.

The nucleotide sequence (A) comprising, for example, promoter sequence, nitroreductase encoding sequence (i) and any additional sequences, or just sequence (i), may be injected with a promoter gene, such as an unmodified promoter gene by the technique described in WO92/11358. For example, if the promoter used for nitroreductase expression and included in the sequence (A) is the BLG promoter, then either the nucleotide sequence (A) is injected into the embryo simultaneously, for example as a mixture, with the unmodified BLG gene, or the nitroreductase encoding sequence and the unmodified BLG gene are introduced sequentially or separately. When the nucleotide sequence (A) and unmodified promoter gene are introduced sequentially, the nucleotide sequence (A) may be introduced before the unmodified promoter gene or vice versa.

Co-introduction may also be achieved by covalently or otherwise linking the first DNA sequence, for example the unmodified promoter gene, and second DNA sequence for example the nucleotide sequence (A): the two sequences may be linked in a single DNA molecule. In this embodiment of the invention it is possible to tailor quite precisely the arrangement of the two sequences to be introduced; for example, a first sequence may be sandwiched between two second sequences, and/or the tandem nature of the sequences (i.e. whether they are head to head or head to tail) can be fixed.

The prodrug which will be used in conjunction with the animal of the invention will be a compound which can be convened by the nitroreductase encoded by the nucleotide sequence (i) into an active drug. Desirably, the toxicity of the prodrug to the animal being treated will be at least one order of magnitude less toxic to the animal than the active drug. Preferably, the active drug will be several, eg 2, 3, 4 or more orders of magnitude more toxic.

Suitable prodrugs include nitrogen mustard compounds and other compounds such as those described in WO93/08288 or EP-A-540 263. Preferred prodrugs are compounds of the general formula:

and:

where R¹ and R² are groups such that the compound R¹NH₂ and R²OH are cytotoxic compounds.

It is preferred that compounds R¹NH₂ and R²OH are aromatic cytotoxic compounds and the compounds R¹NH₂ can be any one of the well known nitrogen mustard compounds, for example based on p-phenylene diamine. Thus, the compound R¹NH₂ can be:

or analogues of this compound with the general structure IV

where R' and R" are H, F or CH₃, and particularly where

R' = H and R" = CH,;

or R' = CH, and R" = H:

or R' = H and R" = F;

or R' = F and R" = H.

Further types of amino cytotoxic compounds that can be used in accordance with the present invention are compounds such as actinomycin D. doxorubicin. daunomycin and mitomycin C. The structure of the pro-drugs derived from actinomycin D and mitomycin C are shown below as V and VII respectively.

Similar p-nitrobenzyloxy derivatives can be made at the amino substituent of other actinomycins and of the other cytotoxic compounds of the type mentioned above.

In addition to forming p-nitrobenzyloxycarbonyl derivatives at an amino group on a cytotoxic compound, similar derivatives can be made at a hydroxy group,

particularly a phenolic hydroxy group of a cytotoxic compound. Here, attention is directed at the phenolic nitrogen mustard compounds, and the compound of formula VIII:

Suitable prodrugs also include other aromatic nitro compounds such as 5-chloro-2,4-dinitrobenzamide, 3,5-dinitrobenzamide, 3-nitrobenzamide, 4-nitrobenzamide and 5-nitro-2-furfuraldehydesemicarbazone (nitrofurazone).

Particularly preferred prodrugs are CB 1954 (5-(aziridin-1-yl)-2,4-dinitrobenzamide), SN 23862 (5-(bis(2'-chloroethyl)amino)-2,4-dimitrobenzamide) and analogues of CB 1954 or SN 23862, such as, for example descarboxamido CB 1954 (1-aziridin-1-yl-2,4-dinitrobenzamide - known as CB 1837), N,N-dimethyl CB 1954 (N,N-dimethyl-(5-aziridin-1-yl)-2,4-dinitrobenzamide - known as CB 10-107), CB 10-199, CB 10-200, CB 10-201, CB10-217, CB10-021 and CB 10-214.

The prodrugs which may be used in the system of the present invention generally comprise a cytotoxic drug linked to a suitable protecting group. Generally the protecting group is removable by a nitroreductase as defined herein or is converted into another substituent by a nitroreductase as defined herein. Alternatively the prodrug is converted by the nitroreductase directly to an active form. In the case of CB1954 the active form is a mixture of the 2- and 4-hydroxy lamino derivatives. These are formed in equal proportions by the nitroreductase (Knox et al, Biochem. Pharmacol. 44 : 2297-2301. 1992). In the case of the 4-hydroxylamine (5-(aziridin-1-yl)-4-hydroxylamino-2-nitrobenzamide) this can become a species capable of binding to DNA and producing interstrand crosslinks, by a direct, non-enzymatic, reaction with either acetyl coenzyme A. butyl and propyl coenzyme A or s-acetylthio-choline. It is thought that the ultimate. DNA reactive species of this derivative of CB 1954 is 4-(N-acetoxy)-5-(aziridin-1-yl)-2-nitrobenzamide (Knox et al, Biochem. Pharmacol. 42 : 1691-1697, 1991). In the case of SN 23862 only a single product is produced by the nitroreductase and this is the 2-hydroxylamine. This hydroxylamine can also be activated to a DNA crosslinking agent by a direct reaction with a thioester.

The prodrug may include any suitable group which can be removed by or modified by a nitroreductase defined herein in such a manner that the group is unstable and undergoes "self immolation" to provide the cytotoxic drug.

The prodrugs of the system of the invention are conveniently prepared by methods of chemical synthesis. For example p-nitrobenzyloxycarbonyl compounds are conveniently prepared by methods of chemical synthesis. For example, the amine or hydroxy cytotoxic compounds can be reacted with 4-nitrobenzyl chloroformate under anhydrous conditions in the presence of a hydrogen chloride acceptor, particularly an alkylamine such as triethylamine. This reaction can be carried out in a dry organic solvent such as chloroform and the resulting compound isolated from the organic solvent by conventional methods such as chromatography. Nitroreductases of the present invention are capable of reducing a nitro group in various substrate molecules and we have found that the nitroreductases are particularly useful in their ability to reduce the nitro group of various p-nitrobenzyloxycarbonyl derivatives of cytotoxic compounds to give self-immolative compounds that automatically decompose to release cytotoxic compounds. Generally the nitroreductase reduces the nitro group to the corresponding hydroxylamino group.

The interest in the present approach resides in the fact that the various cytotoxic compounds containing amino or hydroxy substituents, particularly aromatic amino or hydroxy substituents. give rise to p-nitrobenzyloxycarbonyl derivatives of the amino or hydroxy group which exhibit considerably less cytotoxicity than the amino or hydroxy parent compound. Thus, it is possible to use the p-nitro-benzyloxycarbonyl derivatives as prodrugs in a system of the type discussed above where the prodrug is converted into the cytotoxic form under the influence of a polypeptide expressed within a cell.

For example, compounds of formulae (I) R¹-NH-CO.O-CH₂-Ph-NO₂ (I) where Ph is a phenylene ring and R¹ is a group such that R-NH₂ is a cytotoxic compound; and (II) R²-O-CO.O-CH₂-Ph-NO₂ (II) where Ph is as defined above and R² is a group such that R-OH is a cytotoxic compound may be used as a prodrug in the method or animal of the invention, in conjunction with a nitroreductase defined herein, including the E.coli nitroreductase described in WO93/08288. While the present invention is not dependent, for its definition, upon the exact mode of action of the nitroreductase on the prodrug, for compounds of formula I or II. it is believed that the nitro group of the p-nitrophenylbenzyloxy-carbonyl residue is converted to the corresponding hydroxylamino group. The resulting p-hydroxyl-aminobenzyloxycarbonyl compound automatically degrades under the reaction conditions used for the enzymatic reduction to release the cytotoxic compound. p-Hydroxylaminobenzyl alcohol and carbon dioxide are produced as by-products. The reaction probably proceeds as follows:

_:

For use in the method or animal of the present invention all types of prodrug should be able to enter cells. Accordingly, modifications may be made in the prodrug, eg to make the prodrug more or less lipophilic.

In order to bring about the reduction of the prodrug with the nitroreductase described herein, it is necessary to have a cofactor present in the reaction system. Nitroreductase requires NAD(P)H as cofactor. Since NAD(P)H has a very short serum half-life, concentrations in the blood stream are very low. Accordingly, any nitroreductase produced according to the system of the invention released into the blood stream by cell lysis will be unable to activate any circulating prodrug owing to the absence of NAD(P)H. Thus the presence or absence of cofactor allows a greater selectivity so that prodrug is activated only within cells.

The exact dosage regime will, of course, need to be determined for each treatment. This will be controlled by the exact nature of the prodrug and the cytotoxic agent to be released from the prodrug but some general guidance can be given. The treatment will normally involve parenteral administration of the prodrug.

Intraperitoneal administration or administration by the intravenous route is frequently found to be the most practical. The amount of prodrug required will depend on the type and number of cells being ablated and the animal being treated.

In general a suitable, effective dose will be in the range 1 μg to 10 g per kilogram body weight of recipient per day, preferably in the range 0.01 to 100 mg per kilogram body weight per day and most preferably in the range 0.1 to 50 mg per kilogram body weight per day for intraperitoneal administration. The dose may, if desired, be presented as two, three, four or more sub-doses administered at appropriate intervals throughout the day. These sub-doses may be administered in unit dosage forms, for example, containing 1 μg to 1000 mg, preferably 0.01 to 100 mg and most preferably 0.1 to 50 mg of active ingredient per unit dosage form. The daily dosage may also be administered on up to 5. for example up to 3, sequential days. While it is possible for the compounds to be administered alone it is preferable to present them as pharmaceutical formulations. The formulations of the present invention comprise at least one active ingredient, as above defined, together with one or more acceptable carriers thereof and optionally other ingredients which may be therapeutic. The carrier(s) must be "acceptable" in the sense of being compatible with the other ingredients of the formulation and not deleterious to the recipients thereof.

The cells which may be ablated according to the invention include, for example, breast epithelial cells, tumour cells, fat cells, neural cells and pituitary cells.

Typical tumour cells which may be ablated according to the invention include cells of sarcomas, including osteogenic and soft tissue sarcomas, carcinomas, e.g., breast-, lung-, bladder-, thyroid-, prostate-, colon-, rectum-, pancreas-, stomach-, liver-, uterine-, and ovarian carcinoma, lymphomas, including Hodgkin and non-Hodgkin lymphomas, neuroblastoma, melanoma, myeloma. Wilms tumor, and leukemias, including acute lymphoblastic leukaemia and acute mveloblastic leukaemia, gliomas and retinoblastomas. The invention is illustrated by the following Examples.

1. In vitro Expression of Nitroreductase (NR) in Breast Epithelial Cells a) Cloning Nitroreductase

Two primers were used to amplify the coding region from a plasmid supplied by the Public Health Laboratory Service (PHLS), and designated NTR1003.

The sense primer is:

5'CGCAAAAAAGCTTTCACATTGAGTCATTATGG3'

This was designed to add a HindIII site, knock out an upstream ATG. and improve the initiation site for translation in mammalian cells.

The antisense primer is:

5'CGGCAAGGGATCCTTACACTTCGGTTAAGGTGATG3'

This was designed to add a BamHI site. The coding region was amplified using Pfu polymerase, and the Hindlll-BamHI fragment was directionally cloned into pREP8 (Invitrogen). A clone with the correct restriction map was sequenced from the RSV enhancer region to confirm that the 5' end of the clone was correct and was designed NRR8/3. pREP8 and pNRR8/3 were transfected into E.coli NFR-343 (lacking nitroreductase), and ampicillin resistant colonies were selected for assay. pREP8 and pR8NR were purified for mammalian cell transfections using "Qiagen" reagents. b) Expression of Nitroreductase In Epithelial Cells

HB4a, an SV40 conditionally immortalised human breast (lumenal) cell line was transfected with pREP8 and pNRR8/3 using calcium phosphate precipitation, and stable transfectants selected under ImM Histidinol. For each plasmid 50-60 drug resistant colonies were obtained which were then pooled and continuously maintained under selection. The pooled populations of HB4a/REP8 and HB4a/NR were expanded and examined for nitroreductase protein expression. Immunoblotting of whole cell lysates using a rabbit polyclonal antibody, rb 6s4/ntr, showed HB4a/NR to express large amounts of nitroreductase protein which bands at the appropriate weight as compared to recombinant E.coli nitroreductase protein (Figure 1). As expected HB4a/REP8 did not express any nitroreductase protein. c) Enzymatic Reduction of CB 1954

CB 1954 (100μM) and NADH (500μM) were incubated with a cell lysate prepared by sonication (250μl, 1 mg/mL protein) in 10mM sodium phosphate buffer (pH7) at 37°C. At various times aliquots (10μl) were injected onto a Partisphere SCX (110 × 4.7mm) HPLC column and eluted isocratically (2ml/min) with 50mM NaH₂PO₄ in 1 % (v/v) methanol. The eluate was continuously monitored for absorption at 260, 310, 340nm using a diode-array detector. Alternatively, [U-³H]CB 1954 was added to give an activity of 1.6 × 10⁵ dpm per nmole). Samples (0.3ml) were collected and their tritium activity determined by liquid scintillation counting. Protein concentration was determined by a standard method (Biorad), calibrated against bovine albumin. d) Determination of cell survival

Cells were trypsinised and resuspended in fresh media at 2 × 10⁵ cells per mL.

They were then treated for two hours with CB 1954 (10 μl of appropriate stock in DMSO). The treated cells were then spun down, washed with fresh media, and plated out in triplicate at various concentrations. Cells were fixed and stained after 14 days incubation at 37°C in a humidified atmosphere of 5% CO₂ and 95% air. e) Expression in HOSPIP cells

The modified gene described above was ligated into the vectors pREP4 and pREP8. These two vectors with high level constitutive transcription from the RSV LTR have Hygromycin and Histidinol selectable markers respectively for coexpression of recombinant proteins. Both of these constructs have been transfected into the rat mammary carcinoma cell line HOSPIP by the calcium phosphate technique and are currently under appropriate selection conditions to isolate resistant clones. f) Determination of DNA interstrand crosslinks

HB4a/NR cells were radiolabelled by growth for 48 hours in [³h]-thymidine. The cells were then treated with either 0, 10 or 50 μM CB 1954 for 24 hours and their DNA then analysed by sedimentation in alkaline sucrose.

Results

Enzymatic Reduction of CB 1954

HB4a/NR cells but not HBa/REP8 cells showed a time dependent decrease in the concentration of CB 1954 (Fig 2). Examination of the traces also indicated the formation of the 2- and 4-hydroxy lamino reduction products of CB 1954. This was confirmed by the use of radiolabelled prodrug (Fig. 3). The protein concentration of the HB4a/NR lysate in the assay mixture was determined to be 0.3 mg/mL and the enzyme activity estimated to be 0.05 μg/mL by comparison with the pure protein (see Figs. 2 and 3). Thus NR activity is 1.7 μg/mg cell protein ( ~ 1.7 μg/10⁶ cells). Cell Survival

The plating efficiency of both the HB4a/NR and HB4a/REP8 cell lines was poor and cells only grew when plated at a high initial cell density (1 × 10⁴ per dish). These plates were scored by eye for cell growth. Results are shown in Table 1. There is a dramatic difference in cell survival between the two cell lines. After a 2 hour exposure the HB4a/NR line exhibits cytotoxicity at 1.0μM while the HB4a/REP8 line shows no cytotoxicity until the dose is > 100μM.

DNA Crosslinked Formation

Confirmation that the cyctotoxicity of CB 1954 in HB4a/NR cells is due to its enzvmic reduction is obtained by demonstrating the ability of this compound to form DNA interstrand crosslinks (Fig. 5). Increasing amounts of CB 1954 produce a progressive increase in the amount of DNA that is crosslinked as indicated by the increasing proportion of DNA of higher molecular weight which sediments further into the alkaline sucrose gradient (sedimentation is from left to right) (Fig. 5). DNA strand breakage (either frank breaks or alkali labile sites) is also observed and for clarity the sedimentation profiles are shown relative to a modelled control of the same molecular weight as the strand-broken experimental DNA. 10μM CB 1954 produces about 6 crosslinks and 25 breaks per 10⁹ daltons of DNA whilst 50μM CB 1954 produces 12 crosslinks and 28 breaks. These effects are not seen in untreated cells.

2. Production of Transgenic Animal

Transgenic Example I: Generation of a NR transgene designed to express in the mammary gland: construction of pFG12.

A 679 bp HindIII-BamHI fragment was excised from the NR plasmid pNRR8/3 (SEQ ID No.3). This comprises ORF B encoding E.coli nitroreductase plus synthetic DNA sequences at the 5 ' and 3 ' ends terminating in HindIII and BamHI cleavage sites respectively. The 5' overhangs generated by the restriction endonucleases were filled in using Klenow polymerase so generating a "blunt" ended fragment.

The plasmid pBJ41 was linearised at the unique EcoRV site. This plasmid contains a modified BLG gene comprising 4.2 kb of 5' flanking sequence. BLG cDNA sequences comprising BLG exons I, part V, VI, VII and 1.9kb of 3' flanking sequence. The EcoRV was introduced into the 5' UTR of exon 1 allowing for the introduction of foreign DNA segments at this site which is just downstream of the BLG promoter. The expression of foreign DNA sequences introduced at this site may thus be directed to the mammary gland by the BLG promoter. The

construction of pBJ41 is shown in Figures 6 A and 6B.

The blunt-ended HindIII-BamHI fragment from pNRR8/3 and the EcoRV-linearised pBJ41 were ligated together to produce the plasmid pFG12 (Figure 4). The 679 bp Hind III-BamHI fragment excised from pNRR8/3 (stippled box) was inserted at the unique EcoRV site of plasmid pBJ41 in the orientation shown. pBJ41 is derived from the sheep BLG gene. It contains 4.2 kb of 5' flanking sequence and 1.9 kb of 3' flanking sequence (thick lines) and BLG cDNA sequences (unshaded box) corresponding to exons 1 , pan V, VI and VII as indicated. The BLG derived sequences are inserted into pUC18 (dotted lines) at the Sail and XbaI sites, which can be used to excise the BLG-NR DNA segment for micro injection. This plasmid contains the NR sequences inserted in the correct 5 '-3' orientation for expression at the EcoRV site of pBJ41. The structure of this construct was confirmed by restriction mapping with BgIII and by DNA sequencing. The ~ 9.9 kb SalI-XbaI insen comprising BLG and NR sequences was isolated from the plasmid vector by digestion with SaiI + XbaI, gel electrophoresis and purification of the fragment from the gel. Transgenic Example II: Production and identification of transgenic mice.

The 10.5 kb XbaI-SalI fragment from the BLG plasmid pSS1tgXS (described in WO-A-8800239) was co-injected with the ~ 9.9 kb SalI-XbaI fragment prepared from pFG12. The NR construct was co-injected with the BLG gene in a 3 (BLG): 1(NR) ratio according to the method described in Example 2 of WO-92-11358.

Tail DNA was prepared from putative transgenic mice and analysed by Southern blotting. The DNA was cut with EcoRI, electrophoresed on 1 % agarose gels, transferred to Hybond N and then probed with radioactively labelled BLG or NR sequences. A number of these founder animals were shown to carry BLG sequences and the approximate copy number was estimated by reference to copy number controls. Three of these animals (RED 14, RED20 and RED40) were also shown to carry NR sequences. Transgenic Example 3: Breeding RED transgenic mice

The founder mice RED 17, RED 20 and RED 40 were bred to establish lines. All three mice were fertile and both RED 20 and RED 40 transmitted NR sequences to the next and subsequent generations with the expected Mendelian frequency. The RED 40 founder mouse was male whilst the RED 20 founder was female. No infertility problems were encountered for either male or female mice in the two lines. This contrasts with transgenic mice carrying the HSV tk1 gene in which the males are very often infertile due to aberrant expression of the tk gene in the testis.

Transgenic Example 4: Expression analysis of RED transgenic mice

Expression of the BLG-NR transgene was investigated in lactating transgenic females from the RED 20 and the RED 40 lines. G1 (generation 1) mice were mated and sacrificed at day 12 of lactation and mammary and liver samples taken. The mammary and liver samples were homogenised and the cell extracts electrophoresed and investigated by Western blotting with a rabbit antibody specific to E. coli NR. NR protein was detected and sized at the expected MW of 24kD in the mammary gland extracts from RED 40 individuals (Figure 7).

The Western blot was carried out as follows:

NR Purified E.coli NR fromCAMR Porton Down

1 NR Transgenic mouse RED 20-3

2 NR Transgenic mouse RED 20-5

3 NR Transgenic mouse RED 40-11

4 NR Transgenic mouse RED 40-12

CON Control BRAT 62-5-12

RHS caption 'NR' indicates the expected size for E.coli NR at 24kD Blot development:

Rabbit 654 anti-NR @ 1/1000

Amersham ECL Detection

As expected no expression of NR was detected in the livers of these animals. In line RED 20 NR expression was not detected in either the mammary gland or the liver, suggesting that in this line the transgene may be silent. The reason for this silencing is not known, but this is a common occurrence in transgenic

experimentation.

Expression of NR was also examined immunocytochemically using histological sections of mammary tissue from line RED40 and antibody specific to E. coli NR. These experiments demonstrated that the NR protein was localised specifically in mammary epithelial cells in the NR expressing mice. Controls were negative.

Expression of NR was also examined by Northern blotting analysis of mammary gland RNA. Mice from line RED40 to which had not been administered the ablating agent (see below) exhibited the expected BLG-NR transcript in the mammary gland. These transcripts were not detected in mammary samples from RED20 (not shown).

A Northern blot of mammary gland samples from control and transgenic mice probed with a radioactively labelled NR probe showed that nitroreductase was not expressed in the mammary gland of non-transgenic controls

(F 1.8 and F 1.9) or RED40 mice (47.1, 47.2 and 47.3) after ablation (see below). The Northern blot showed that NR was expressed in the Mammary gland of the non-ablated RED40 mouse (47.5).

Transgenic Example 5: CB1954 Ablation Inducible tissue specific ablation was achieved in transgenic mice from line RED 40 by the administration of the pro-drug CB1954. The drug was made up in 10% acetone/arachis oil at a concentration of 5.0 mg/ml. 0.2 ml doses were administered either once or three times at successive 24 hour intervals. The mock injected mice were injected with acetone/arachis oil. The mice were sacrificed 72 hours after the first injection. Ablation was carried out at two developmental stages:- i) day 14-17 of pregnancy at which stage cell division is still occurring in the mammary gland, ii) day 4-7 of lactation at which stage little cell division is taking place.

No untoward effects were observed in either the RED 40 mice or controls upon CB1954 administration except for the group of lactating RED 40 mice in which the lactation failed and these animals were unable to continue suckling their young. Northern blotting analysis of mammary tissue from RED 40 females after CB1954 ablation showed that the NR expression in the mammary gland was significantly reduced suggesting that destruction of NR expressing cells may have taken place. This was confirmed by histological analyses of the mammary gland tissue from ablated and control mice. At both stages (pregnancy and lactation) CB1954 administration to the RED 40 mice results in massive cell killing in the mammary gland. The effect is specific to the mammary gland as judged by the fact that liver samples from these mice appear the same as the controls. In this histological analysis control mammary gland or liver samples (i.e. from non transgenic mice) also appeared unaffected by CB1954 administration (not shown).

Mammary tissue from CB1954 treated and mock controls was fixed, mounted in paraffin wax, sectioned and stained in H&E. The sections were viewed under the low power (x4 objective). For pregnant mice the CB1954 injected RED 40 mouse received only one dose of CB1954. Substantial cell killing in the mammary elands was evident after CB1954 administration and a substantial fraction of the gland structure had evidently been destroyed.

For lactating mice the CB1954 injected RED 40 mouse received 3 doses of CB1954. CB1954 administration caused massive tissue destruction and the structure of the gland was disrupted.

For liver samples from lactating mice no differences were observable between experimental and control samples.

Claims

1. A method of producing a transgenic non-human animal, which method comprises incorporating into the genome of the non-human animal at least one nucleotide sequence (A) comprising a sequence (i) encoding a nitroreductase which is capable of converting a prodrug into a cytotoxic drug.

2. A method according to claim 1 wherein the nucleotide sequence (A) contains the oligonucleotide of SEQ ID No.1 , fragment thereof or oligonucleotide hybrid isable thereto.

3. A method according to claim 1 wherein the nucleotide sequence (A) contains the oligonucleotide of SEQ ID No.3, fragment thereof or oligonucleotide hybridisable thereto.

4. A method according to claim 1 , 2 or 3 wherein the nitroreductase is of the sequence SEQ ID No.2, fragment thereof or homologue thereof.

5. A method according to any one of the preceding claims wherein the nucleotide sequence (A) comprises a promoter operably linked to the sequence (i) encoding the nitroreductase.

6. A method according to any one of the preceding claims wherein the prodrug is a nitroreductase mustard compound.

7. A method according to claim 6 wherein the prodrug is 1-aziridin-1-yl-2,4-dinitrobenzamide or derivative thereof or 2,4-dinitro-5-(N,N-di-(2-chloroethyl))aminobenzamide or derivative thereof.

8. A transgenic non-human animal whose germ cells and somatic cells comprise at least one nucleotide sequence (A) as defined in any one of the preceding claims, which nucleotide sequence (A) comprises a sequence (i) encoding a nitroreductase which is capable of converting a prodrug into a cytotoxic drug.

9. An animal according to claim 8 wherein the prodrug is as defined in claim 6 or 7.

10. An animal according to claim 8 or 9 which comprises at least one nucleotide sequence (A) as a result of incorporating into the animal genome or into the genome of an ancestor of said animal said nucleotide sequence.

11. A method of converting a prodrug into a cytotoxic drug, which method comprises exposing an animal as claimed in claim 8, 9 or 10 or produced by the method as claimed in any one of claims 1 to 7 to a prodrug capable of being converted to a cytotoxic drug by the nitroreductase encoded by the at least one nucleotide sequence (i).

12. A method of ablating cells in a transgenic animal which method comprises exposing an animal as claimed in claim 8, 9 or 10 or produced by the method as claimed in any one of claims 1 to 7 to a prodrug capable of being converted to a cytotoxic drug by the nitroreductase encoded by the nucleotide sequence (i).

13. A method according to claim 11 or 12 wherein the prodrug is as defined in claim 6 or 7.

14. A method according to claim 12 or 13 wherein the cells are normal.

15. A method according to any one of claims 12 to 14 wherein the cells are breast epithelial cells, pituitary cells, fat cells or neural cells.

16. A method according to any one of claims 12 to 15 wherein the cells are of a single cell type.