WO1994021111A1 - Transgenic mammalian as disease model - Google Patents

Abstract

A transgenic mammal is described which comprises a heterologous gene encoding an effector associated with a disease, the gene comprising at least one control sequence effective to direct expression of the effector substantially exclusively to cells where the effector is expressed in the normal course of the disease. The mammal of the invention is useful in the study of diseases and therapy for diseases.

Description

TRANSGENIC MAMALIAN AS DISEASE MODEL The present invention relates to a transgenic mammalian model for a disease. In particular, the present invention relates to a transgenic mouse which expresses HIV- derived peptides in a tissue-specific manner.

The study of a disease is dependent to a large extent upon the model systems available to researchers which allow the disease to be studied outside the clinical context. Especially in the case of diseases which carry the risk of mortality or permanent disablement, the study of pathological conditions and the efficacy of drugs in human subjects is ethically questionable and often unacceptable. Furthermore, the range of studies which may be undertaken with volunteers is limited.

Much basic research, as well as more advanced studies, is carried out in artificial living systems such as cultured cells, or in animals.

Cultured cells, however, present a number of well- known problems which complicate their use in the study of disease. For example, most transformed cell lines possess characteristics which are very different from those of cells in living organisms or even cells in primary cell culture.

Furthermore, cell-cell interactions in living tissue are well known to affect the behaviour of cells in a manner difficult to reproduce outside of living animals or organ culture.

The use of animals as model systems, although overcoming many of the above disadvantages, is limited by the existence of an animal equivalent of the human disease in question. The most common animal models, rodents, do not share many diseases with humans. The use of animals such as primates is objectionable on ethical as well as financial grounds. It has therefore been sought, in a few cases, to adapt rodents in order to reproduce human diseases therein, so that the disease may be studied. Examples range from the irradiation of mice to produce immunocompromised animals, through the use of specific mutants such as non-obese diabetic (NOD) mice, to the recent use of mice rendered highly susceptible to tumorigenesis (the Harvard Oncomouse; see European Patent Application 0162672).

The study of AIDS is in part frustrated by the lack of suitable models for the disease. Immunodeficiency diseases exist in primates, where the ethological agent is SIV. However, no rodent model exists, nor is it possible to control the effect of viruses such as SIV in order to study individual aspects of its function and pathology.

For example, the function of the nef gene product of HIV is poorly understood.

NEF is encoded by an open reading frame overlapping the 3' HIV LTR (Guy et al ., 1987) and displays a high degree of polymorphism between HIV isolates (Ratner et al . , 1985). Up to 80% of the early, multiply spliced class of viral transcripts encode NEF (Robert-Guroff et al . , 1990), yet its function is unclear. The 27 Kd myristilated protein is expressed at a very high level early in the HIV life cycle (Haseltine, 1991) and is found in the cytoplasm (Franchini et al ., 1986). While NEF shows some sequence homology to G proteins (Guy et al . , 1987), it seems unlikely that it has either GTP-binding or GTP-hydrolysing properties (Backer et al . , 1991). Evidence for (Ahmad and Venkatesan, 1988) and against (Hammes et al . , 1989) the NEF protein having a negative regulatory effect on HIV LTR transcription has been reported. It has been shown that NEF is non-essential for viral replication in cultured T cell lines (Hammes et al . , 1989; Kim et al ., 1989) although it has been observed to accelerate viral replication in primary lymphocytes (De Ronde et al . , 1992).

Extensive evidence to support the essential role of NEF in immunodeficiency virus replication has been obtained in SIV infected rhesus monkeys. A full length SIV NEF product was shown to be necessary for viral replication and subsequent SIV pathogenesis (Kestler et al . , 1991). These data directly demonstrate that there is a strong selective pressure for a functional NEF protein in vivo .

While NEF may provide an important function in the life cycle of the virus, it may have an adverse effect on host, cell function. In vitro , the NEF protein, in the absence of other HIV sequences, has been shown to downregulate the levels of cell surface CD4 on human T cell lines (Garcia and Miller, 1991; Garcia et al . , 1993). A post-translational mechanism has been postulated since downregulation has been observed not at the level of mRNA but in surface CD4 protein. CD4 was found to be localized in the cytoplasm (Garcia and Miller, 1991). NEF has also been found to downregulate mouse and simian cell surface CD4 suggesting a common mechanism of action (Garcia et al . , 1993). Since NEF has such effects on established mature human CD4+ T cell lines, the relevance of such downregulation should be established in vivo .

However, no animal model is described in the prior art due to the difficulties involved in reproducing the pattern of infection of HIV in living animals without infecting the same with an immunodeficiency virus. For this reason, the study of SIV in monkeys has been favoured to date.

If such an animal model were available, it would be useful for the investigation of the properties and effects of not only the nef gene product, but also of a number of biologically active peptides encoded by HIV.

Transgenic animals such as the Harvard Oncomouse are fundamentally incompatible with the study of most diseases with the exception of certain tumours. This is because the Oncomouse expresses oncogenes under the control of a viral LTR promoter/enhancer. Such control sequences essentially lead to widespread yet unpredictable expression of the oncogene, as evidenced by the experimental data referred to in the examples of EP 0162672. The majority of diseases, including viral diseases, involve a tissue-specific event, such as the expression of HIV genes in T-cells. The oncomouse is incapable of producing tissue-specific expression of oncogenes on a repeatable basis.

It has now been established that it is possible to express disease effectors, such as the effector proteins of HIV, in a tissue-specific manner which may be tailored to suit the characteristics of the disease.

According to the present invention, therefore, there is provided a transgenic mammal comprising a gene encoding an effector associated with a disease, the gene comprising at least one control sequence effective to direct expression of the effector substantially exclusively to cells where the effector is expressed in the normal course of the disease.

By "effector", it is intended to denote that the heterologous gene encodes an agent which promotes a biological effect in the course of the disease. Such agents may be polypeptides or nucleic acids. Preferably, the effector is a polypeptide. Examplesof polypeptide effectors include the products of oncogenes such as ras, myc and fos, the aberrant proteins such as tau protein which are implicated in Alzheimer's disease, HIV-specific peptides such as TAT, NEF, REV, VPU, VPR and VIF, bacterial toxins, self-antigens and any other peptides which cause or have a role in disease.

Preferably, the disease is a human disease. Any disease may be studied using the transgenic mammal of the invention, provided that the disease involves the expression of an effector in particular cell types of the host. Examples of such diseases include, but are not limited to, Alzheimer's disease, cancer, autoimmune diseases, cystic fibrosis and infectious diseases of all kinds.

However, it is especially preferred that the invention be applied to viral diseases, especially AIDS. The invention allows the study of a particular aspect of the pathology of the disease to be studied in isolation, or in connection with any other preselected aspects of the disease. For example, in the case of AIDS, the effect of virally-encoded proteins on the immune system may be studied, without the risk of pathological immunodeficiency arising from viral infection colouring the results. Furthermore, the effect of the interaction of two or more virally encoded proteins on the immune system may be studied.

The control sequences used in the invention may be any sequences capable of directing tissue-specific expression in a host animal. For example, combinations of promoters, enhancers and tissue-specific responsive elements may be used. Such combinations have been studied in cell lines and transgenic animals in the prior art and the selection of appropriate combinations will be within the capabilities of a person skilled in the art.

Especially preferred, however, is the use of sequences which confer integration site-independent expression characteristics to the transgene, thus ensuring a consistency of expression between model mammals. For example, the control sequences may comprise Locus Control Regions (LCRs). LCRs are position-independent, copy number- dependent activators of gene transcription which display strong tissue specificity. First discovered in globin genes (Grosveld et al , 1987) these elements are believed to direct the creation of independent regulatory domains within the chromatin structure of a cell genome, thereby ensuring the potential activity of a co-transferred gene.

A number of LCRs other than those for globin genes has now been described. The tissue specificity of each of these differs. Particularly preferred for use in the present invention are the CD2 LCR (Greaves et al . , 1989) which is specific for T-lymphocytes, the macrophage-specific lysozyme LCR (Bonifer et al . , 1990) and the Class II LCR specific for dendritic cells and macrophages (Carson et al . , 1993).

Preferably, the gene used in the invention comprises an LCR together with an appropriate promoter/enhancer to drive transcription of the gene in the intended cell type. For example, greater tissue-specificity can be achieved by the use, in combination, of an LCR and a tissue-specific promoter and/or enhancer.

Appropriate selection of promoter elements may be used to introduce developmental regulation as well as tissue- specific regulation of the transgene. For example, when considering T-cell specific expression, use of an IL-2 promoter and CD2 LCR will ensure that the transgene is expressed only late in the T-cell development cycle, in peripheral T-cells, rather than in the thymus. This mirrors an HIV infection, which would occur initially in peripheral cells.

Moreover, a greater degree of control over the gene can be provided by the use of a regulatable promoter and/or enhancer, which may be susceptible to regulation by, for example, transcription factors (Hu and Davidson, 1987; Kakidani and Ptashne, 1988), hormones, such as glucocorticoids (Picard et al . , 1988), oestrogen (Boehmelt et al . , 1992) or orally administrable non-toxic small molecules, such as tetracycline (Gatz and Quail, 1988; Gossen and Bujard, 1992). The use of such regulatory mechanisms enables the activation or inactivation of genes to be carried out at will, further extending the flexibility of the invention.

The transgenic mammal of the present invention may be any non-human mammal. However, rodents, especially mice, are preferred.

Transgenic mammals may be generated by any technique known in the art. By "transgenic", it is intended to infer that the mammal in question comprises at least one active copy of a heterologous gene in a substantial proportion of the cells of interest. The heterologous gene may be inserted by conventional techniques, such as microinjection of embryos, such that the gene is present in substantially all the cells of the mammal. Alternatively, it may be delivered in a targeted or non-targeted manner to mature animals, using, for example, virus vectors or liposome-based vectors according to techniques known in the art.

The heterologous gene encoding the effector may be present in an episomal state. However, it is preferred that the gene be integrated into the genome of the transgenic mammal.

In a preferred embodiment of the invention, there is provided a transgenic mouse comprising the coding sequence of an HIV peptide under the control of the CD2 LCR. For example, the HIV peptide may be the HIV nef gene product.

In a second aspect of the present invention, there is provided the use of a gene encoding an effector of a disease and comprising at least one control sequence effective to direct expression of the effector substantially exclusively to cells where the effector is expressed in the normal course of the disease in the generation of a transgenic mammal for use as a model for the disease.

The transgenic mammals of the invention are useful for the study of diseases in general and particularly for the study of therapy intended for diseases. For example, agents which target the HIV regulatory proteins NEF, TAT, REV, VPU,

VIF and VPR may be assayed in transgenic animals according to the invention. It will be apparent that regulators derived from viruses other than HIV may be studied in a similar manner.

Accordingly, the invention provides a method for studying a potential therapeutic agent for a disease comprising administering the agent to a transgenic mammal according to the invention.

The invention is described, for the purposes of illustration only, in the following examples, with reference to the Figures, in which:- FIGURE 1 is a schematic representation of the transgene used for the generation of CD2-nef transgenic mice;

FIGURE 2A shows a southern blot of DNA isolated from four transgenic mouse lines;

FIGURE 2B shows a slot blot analysis of RNA from the same lines;

FIGURE 2C shows western blot analysis of spleen tissue extract in the four lines;

FIGURE 3 shows representative FACS analysis for transgenic and non-transgenic littermates performed on cells from the thymus and peripheral lymphoid organs, the spleen and lymph nodes;

FIGURE 4 shows a FACS histogram analysis of surface levels of CD4 on thymocytes from nef transgenic mice;

FIGURE 5 shows the results of an anti-CD3ε mediated activation assay;

FIGURE 6 compares the results of analysis of normal and transgenic thymocytes of the presence of CD4 by direct immunofluorescence;

FIGURE 7 shows a similar experiment to Figure 6, except that the cells have been double-stained for CD4 and subcellular compartments;

FIGURE 8 shows the construction of a CD2 LCR - tat transgene;

FIGURE 9 shows: A. Slot blot identification of transgenic mice carrying the CD2 LCR - tat transgene;

B. RNA slot blot analysis of tissues for tat transgenic mice;

C. S1 protection wrapping of RNA extracted from tissues of tat transgenic mice;

FIGURE 10 shows a FACS analysis of CD4 and CD8 T-cell subsets in tat transgenic mice;

FIGURE 11 shows a northern blot analysis of RNA derived from CD2 - tat transgenic mice, probed with a number of cytokine-specific probes; and

FIGURE 12 shows the observed increase in TNF-β transcription in tat transgenic mice.

Example I

DNA Constructs

The 800 bp BamHI-Smal fragment from either pTG1147 or pTG1191 (gift from Dr. B.Guy) was blunted and ligated into a unique blunted EcoRl site in the first exon of the p2629 CD2 expression plasmid (gift from D.Kioussis) to give either p2629N47 or p2629N91. A 4.5Kb BamHI-NotI fragment containing the 3' CD2 LCR from p2694 (gift from D.Kioussis) was then ligated into the unique BamHI-NotI sites in P2629N47 or p2629N91, resulting in either pCD2nef 1147 or pCD2nef 1191. The 12Kb Sall-NotI fragment from these plasmids was prepared for microinjection into (CBAxC57BL/10) fertilised mouse oocytes as previously described (Grosveld et al . , 1987). Positive founder animals were bred with CBAxC57BL/10) mice and lines were maintained as heterozygotes.

DNA and Expression Analysis

Tail DNA (10 μg) from founder animals was analysed by Southern blot analysis after digestion with Hindlll or Asp718. DNA was run on a 1% agarose/Tris acetate, EDTA gel, blotted onto nitrocellulose and probed with a randomly primed 800 bp BamHI-Smal nef fragment from pTG1147. A 1.2 Kb Thy-1.2 fragment was used as a loading control probe.

Appropriate amounts of pCD2nef 1147 spiked in 10 μg genomic DNA were used as copy number controls. Quantitation was performed on the Molecular Dynamics PhosphorImager.

RNA was prepared using the lithium chloride/urea method (Fraser et al . , 1990). For Northern blot analysis (Sambrook et al . , 1989) 10 μg of RNA was run on a 1% formaldehyde gel, blotted onto nitrocellulose and probed with an 800 bp BamHI-Smal nef fragment from pTG1147. For RNA slot blots (Sambrook et al . , 1989) 5μg of RNA was blotted onto nitrocellulose and probed as above. RNA from the NEF producing CRIP L producer cell line (Schwartz et al . , 1992) was used as a positive control.

FACS analysis

FACS analysis was used to detect cell surface markers on lymphocytes from transgenic mice. The antibodies used were: a PE-conjugated rat monoclonal antibody (mAb) against murine CD4; a FITC-conjugated rat mAb against murine CD8 (both Becton Dickinson, San Jose, CA); a FITC-conjugated hamster mAb against murine CD3e (Pharmingen, San Diego, CA) and a FITC-conjugated rat mAb against murine Thy-1.2 (Sigma Chemical Co., St. Louis, MO). The thymus, spleen and lymph nodes were removed and homogenised to single cell suspensions in FACS medium (αMEM, 5% FCS, lOμg/ml Na azide) on ice. Accurate cell counts were obtained and 10 cells were washed in 5 ml FACS media, pelleted and the supernatant removed. Antibodies were added at a dilution of 1:200 in FACS medium and incubated for 30 min on ice. Cells were washed once with 5 ml of cold FACS medium, once with 5 ml of cold PBS, fixed in 1% formaldehyde/PBS and filtered through fine gauze. Stained cells were analysed with a Becton Dickinson FACSCAN cell sorter and the LYSIS II software package.

Immunoprecipitation and Western blot analysis

Single cell suspensions from thymus and spleen of transgenic and non-transgenic mice were prepared. Erythrocytes were removed by lysis in tris-buffered ammonium chloride. Extracts were prepared by lysing cells in 1 ml of 0.5% Triton X100, 10mM Tris pH 7.5, ImM EDTA, 0.15 mM NaCl, 10 mg/ml bovine serum albumin, 200μM PMSF, 5mM iodoacetamide and 5 μM leupeptin on ice for 15 min. The extracts were cleared by centrifugation (14,000 rpm) for 5 min at 4°C and incubated with 50 μl of normal rabbit serum for 1 hr at 4°C and for 30 min with 100μl of a 10% suspension of protein A- sepharose in lysis buffer followed by 15 min centrifugation at 4°C. Anti-NEF antibodies (HIV-1 HXB3 NEF antisera) to the N and C termini (Hammes et al . , 1989) were then added at a 1:250 dilution to each cell lysate and incubated overnight on ice. Following this the extracts were incubated again with 100 μl of 10% suspension of protein A-sepharose beads for 1 hr. The beads were collected by centrifugation for 15 min at 4°C and washed three times in lysis buffer. Pellets were resuspended in reducing sample buffer, heated at 100°C for 5 min and the supernatants recovered. The supernatants were resolved on 15% SDS-polyacrylamide gels and transferred to nitrocellulose filters by electroblotting. Filters were blocked in 5% low fat dried milk dissolved in phosphate buffered saline with 0.1% Tween-20 (PBS-T) at 4°C overnight. After extensive washing in PBS-T the filters were incubated for 1 hr at room temperature with a mAb to NEF (AE6) diluted at 1:1000 in PBS. After further washing in PBS-T the filters were incubated with horseradish peroxidase- conjugated goat anti-mouse IgG antibody (Amersham) at 1:5000 dilution, washed extensively in PBS-T and the immune complexes visualised by the ECL detection system (Amersham) and autoradiography.

T cell proliferation assays

Thymocytes and erythrocyte depleted splenocytes were cultured in 200 μl of αMEM, 10% FCS, 2 mM glutamine, 10 U/ml Penicillin, 100 μg/ml Streptomycin and 50 μM β- mercaptoethanol. Cells were cultured in microtiter wells from a density of 8 x 10 per well for splenocytes. Thymocytes were stimulated with anti-CD3e (145-2C11) mAb (0.36 μg/well) and 5 ng/ml PMA (Sigma) or with 5 ng/ml PMA and 500 ng/ml ionomycin (Sigma) or 5 ng/ml PMA alone as a control. Splenocytes were stimulated as above except no PMA was added with anti-CD3e antibodies. Controls contained no PMA. 48 hours after stimulation cells were labelled for 16 hours with 1 μCi/well of ³H thymidine (Amersham) before harvest. The incorporated radioactivity was precipitated on glassfibre filter paper and subsequently counted by liquid scintillation.

Immunofluorescence

Indirect immunofluorescence was used to detect the intracellular and surface distribution of CD4 on lymphocytes from transgenic mice. The antibodies used were a rat mAb against murine CD4 (Pharmingen, San Diego, CA) detected via a FITC-conjugated goat anti-rat IgG (Calbiochem., La Jolla, CA); a golgi-specific rabbit anti-α mannosidase II antibody (gift from Dr. K. Moremen, Univ.Georgia); an endoplasmic reticulum-specific rabbit antibody, anti-ERC55 (gift from K.Weis and A.Lamond, EMBL, Heidelberg) and; an anti-p56 lck [RNGS] rabbit antiserum (gift from M.Marsh, Univ. Coll., London) all three detected via a Texas Red-conjugated goat anti-rabbit IgG (Calbiochem., La Jolla, CA).

Whole thymus and lymph nodes were homogenised to a single cell suspension in PB2 and filtered through fine gauze. Approximately 3x10⁵ cells were spread on 10 mm poly- lysine coated coverslips and were allowed to attach for 2- 3 min. Cells were fixed with 3.7% paraformaldehyde in CSK (100mM NaCl, 300mM sucrose, 10mM PIPES pH6.8, 3mM MgCl, 2mM EGTA pH6.8) for 10 min with periodic swirling, followed by three 5 min washes with PBS. Cells were permeabilised with 0.5% Triton-X in CSK for 15 min, followed by three 5 min washes with PBS.

To stain the cells the coverslips were first overlaid with 5 μl of block solution (0.8% BSA, 0.1% gelatin in PBS) for 15 min. Diluted antibodies in the same block solution were then applied (10 μl) in the following sequence; (1) anti-CD4 (1:200); (2) FITC anti-rat IgG (1:100); then for double staining either; (3) anti-golgi-marker (1:1000), or anti-ER marker (1:20), or anti-p56 lck (1:50); (4) Texas Red anti-rabbit IgG (1:100). The incubation period for each antibody was 30 min at room temperature in a humidified chamber with three 5 min washes with 0.05% Tween-20/PBS between each application. Coverslips were mounted with a drop of Univert (BDH, Poole, UK) containing 100 mg/ml of DABCO (Sigma Chemical Co., St. Louis, MO) as anti-fading agent, and cells were viewed under a Zeiss Axiophot fluorescence microscope.

CD2 nef transgenic mice express NEF in thymocytes and peripheral T cells

Four transgenic lines of mice were produced with a construct containing the human CD2 promoter and LCR element and a 621 bp nef fragment (Figure 1). Two different alleles of the HIV-1 nef gene were used to examine the effects of NEF in vivo , 1147 with threonine at amino acid position 15 and 1191 with an alanine at position 15. In vitro studies show CD4 downregulation with both alleles (Guy et al . , 1990) but only the 1147 allele is phosphorylated at position 15. DNA from the four lines A (1147), F (1147), B (1191) and D (1191) was analyzed by Southern blot analysis and copy numbers were determined to be 6, 25, 48 and 26 respectively (Figure 2A). Slot blot RNA analysis demonstrated expression of the nef transgene in all four lines (Figure 2B). Expression was tissue specific and observed only in the thymus and spleen (Figure 2B). Quantitation of RNA by probing for glucose-6-phosphate dehydrogenase transcripts and phosphorimaging demonstrated that expression levels were consistent with copy number dependent expression in the four different lines. For A:F:B:D having transgene copy numbers 6:25:48:26 respectively, the ratio of RNA expression levels in thymus was 6:22:30:22. Western blot analysis confirmed expression of NEF protein in the spleen (Figure 2C) and the thymus (not shown) of all four transgenic mouse lines.

Thymocytes and peripheral T cells populations are altered in nef transgenic mice

To examine whether NEF had any effects on the T cells of the transgenic mice, FACS analysis was performed on cells from the thymus and the peripheral lymphoid organs; spleen and lymph nodes. Antibodies specific for CD4 and CD8 were used for analysis of distinct T cell subpopulations. Figure 3 shows representative FACS analysis for transgenic and non- transgenic littermates. In all lines we observed a decrease in the percentage of CD4 single positive (SP) thymocytes and a concomitant increase in the percentage of CD4/CD8 double positive (DP) cells. While the percentages of double negative (DN) and SP CD8 cells remained similar in lines A and D and line B (Figure 3B), the percentages of CD8 SP cells slightly decreased and the DN cells increased in line F (Figure 3A). Similar statistically significant changes were found in all lines (Table I) when comparisons were made of data from twelve litters of mice (three transgenic and three non-transgenic littermates per experiment). Cell counts of the total number of thymocytes in lines A, B and D showed a decrease of 10% on average, but a large decrease (approximately 70%) in thymocyte cell number was observed in line F (Table II). The greatest reduction in absolutenumbers for line F was observed in the CD4 SP subset and resulted in 15 fold fewer cells. The total number of CD8 SP cells was decreased 12 fold while the number of DN cells were not significantly changed. In the D transgenic line although the reduction in total number of thymocytes was less than line F, again a significant decrease was observed in the CD4 SP subset.

When peripheral T cell populations were examined, variable decreases in CD4 SP cells in the lymph nodes and spleen were seen for lines A, B and D. However, line F exhibited large decreases in the percentage of CD4 SP cells in the lymph nodes. Normally non-transgenic lymph nodes contained 51.1% (+/- 3.4) CD4 SP while only 14.7% (+/- 6.7) were found in nef transgenic lymph nodes, representing a 3.5 fold decrease. A decrease from 23.5% (+/- 1.9) to 8.5% (+/- 3.9) was observed in the CD8 SP subset. In addition, the total number of cells in the spleen was unchanged in the A, B and D lines but decreased in line F by a factor of 1.5 in the CD4 SP compartment. Thus, nef gene expression consistently results in a decrease of the absolute number of CD4 SP cells in the thymus and in one line this effect extends to the peripheral lymphoid organs.

NEF expression results in a downregulation of CD4 on the surface of developing T cells

Since others have found downregulation of CD4 on T cell lines transfected with the 1147 and 1191 alleles of the nef gene, we investigated whether T cells in nef transgenic mice also downregulate CD4 in vivo . FACS histogram analysis revealed that surface levels of CD4 were decreased in thymocytes from nef transgenic mice (Figure 4). The downregulation was observed in all four lines (Table III), for both nef alleles. As controls, antibodies specific for CD8, CD3 and Thy-1 were used to examine the specificity of the downregulation effects. With both nef transgenic alleles, CD8 levels were found to be slightly decreased and the normally high CD3 expressing population of thymocytes was found to be greatly reduced in line F and less so in the other three lines. This loss of CD3 high cells correlates well with the loss of CD4 SP cells in the thymus. Thy-1 levels did not change, thus demonstrating the specific effects of NEF.

To determine which subset of cells was downregulated for CD4 levels, we performed histogram analysis on gated CD4 SP and DP thymocyte populations. In twelve litters of mice, consistent downregulation ranging from 17% to 72% of normal levels of surface CD4 was observed on all DP populations. Thymocytes from the F line showed the greatest levels of downregulation. Some downregulation (but not to the extent in the DP population ) was also seen on SP cells ranging up to a 28% decrease from normal levels. When lymph node T cells were examined, some but not consistent downregulation of CD4 was observed only in line F, suggesting that CD4 low expressing cells rarely leave the thymus for the periphery.

T cell activation is decreased in nef transgenic mice

Mitogen induced or anti-CD3e mediated activation assays were performed to examine whether downregulation of CD4 or loss of CD4+ cells haα negative effects on thymocyte or peripheral T cell activation. Proliferation of thymocytes as measured by H thymidine incorporation after activation via the calcium ionophore (ionomycin) and phorbol ester (PMA) revealed small differences between nef transgenic and non-transgenic cells of both alleles demonstrating that the total response of transgenic thymocytes to mitogen is not impaired. However, when cells were activated via the T cell receptor-CD3 complex with anti-CD3e antibodies and PMA, a measurable difference between transgenic and non-transgenic thymocytes was observed (Figure 5B). This decrease in activation was seen in transgenic lines carrying both the nef 1147 allele and the 1191 allele. In the experiments the shift of the titration curves to the right clearly indicates that more transgenic thymocytes are required to achieve the equivalent activation levels observed in non-transgenic thymocytes. These changes correlate well with the quantitative loss of SP CD4 cells in the transgenic thymuses (Table I). In addition, decreases in CD3e mediated activation were observed in proliferation assays performed on peripheral T cells from Line F and correspond to the decrease in the number of CD4 SP cells in the lymph nodes.

Changes in T cell subsets are a consequence of intracellular sequestration of CD4.

In vitro, NEF has been shown to have no effect on the steady-state levels of CD4 mRNA or CD4 protein and the surface downregulation was found to be a consequence of intracellular localization of CD4 (Garcia and Miller, 1991). Thus, we examined thymocytes from nef transgenic mice for the presence of intracellular CD4. Indirect immunofluorescence was performed on permeablized thymocytes with anti-CD4 antibody. As shown in Figure 6A, non- transgenic permeablized cells have normal cell surface expression of CD4 whereas transgenic thymocytes (Figure 6B) express only low levels. Instead, CD4 was observed as a singular brightly staining area within the cytoplasm of nef transgenic thymocytes. To determine whether CD4 was localized to any particular subcellular compartment, we stained thymocytes with antibodies specific for the endoplasmic reticulum (ER) and the α-mannosidase II protein of the golgi apparatus. Double staining with CD4 and compartment specific antibodies revealed that CD4 was sequestered within the specific region stained by the anti- golgi (figure 7, A and B) but not the anti-ER antibodies (Figure 7, C and D). While CD4 colocalises with the golgi marker, due to the coalescence of these organelles in the perinuclear region we cannot rule out an endosomal localization for CD4. NEF expression was found throughout the cytoplasm of transgenic thymocytes as previously described for HIV infected cells (Franchini et al . , 1986; Ovod et al . , 1992). Since the cytoplasmic domain of CD4 is known to interact directly with the tyrosine kinase p56 lck (Shaw et al . , 1989; Veillette et al . , 1988), double staining with CD4 and p56 lck specific antibodies was performed to determine whether the NEF mediated downregulation of CD4 also affected the cellular location of p56 lck. Although the intensity of membrane associated staining for p56 lck does not change between transgenic and non-transgenic thymocytes, intracellularly p56 lck appears to colocalise with CD4 in CD4 downregulated cells (Figure 7, E and F).

We have demonstrated that HIV-1 has an effect on the CD4 subsets of thymocytes which extends to the peripheral T cells in transgenic mice. Due to the CD2 gene regulatory elements, nef begins expression in the transgenic mice very early in T cell differentiation while the cells are in the CD4/CD8 DN stage in the thymus and continues in the DP and SP thymocytes as well as in the peripheral T cells (Kamoun et al . , 1981; Lang et al . , 1988; Owen et al . , 1988). We observed a dramatic decrease in the levels of cell surface CD4 on the DP subset of thymocytes and a significant reduction in the number of CD4 SP thymocytes in all four mouse lines. Additionally, we observed a reduction in the peripheral CD4+ T cells of the F transgenic line but no significant loss of CD4+ cells occurred in the periphery of the other three transgenic lines. This variation is most likely due to threshold effects of nef expression and/or allelic differences. Others have reported that high levels of nef expression are required for in vitro downregulation of CD4 (Schwartz et al . , 1993).

The decrease in CD4 SP cell number could be due to a NEF specific cytopathic effect, as has been indicated previously (Luria et al . , 1991; Skowronski et al . , 1993). However, we have found nef RNA and protein in both the thymocytes and splenocytes of the transgenic mice demonstrating that NEF is not directly toxic to T cells (SP or DP). Alternatively, NEF mediated depletion of CD4 SP cells may be a consequence of aberrant positive selection in the thymus. The effect of NEF is most likely to occur early in the T cell differentiation pathway at the DP stage, initiated by the downregulation of CD4. In normal positive selection, DP cells interact with MHC class I and II molecules presented by the cortical thymic epithelium and receive a signal to expand (Berg et al . , 1989). Mice deficient in MHC Class II (Cosgrove et al . , 1991) and mice treated with anti-class II antibodies (Kruisbeek et al . , 1983) show defective development of their CD4+ T cells and clearly demonstrate that CD4 must interact with Class II for expansion to occur. Additionally, in mice homozygous for CD4 which have reduced levels of CD4 on the surface of DP thymocytes (Rahemtulla et al . , 1991), decreased numbers of CD4 SP cells have been found in the thymus. These studies along with our results strongly support a NEF mediated effect on positive selection in the thymus through downregulation of cell surface CD4. This could result in an alteration in the interaction of transgenic thymocytes with MHC class II molecules and lead to the maturation of fewer SP CD4 thymocytes.

It has been shown that thymocyte activation assays with anti-CD3e antibody result in proliferation of only SP cells (Havran et al . , 1987; Weiss et al . , 1987). Differences in thymocyte activation could be due to quantitative changes in the total number of SP cells or the level to which individual SP cells can be activated. Our results concerning the effects of NEF on the in vivo immune system indicate that equal numbers of transgenic thymocytes are stimulated to a lesser degree than those from non- transgenic littermates. This is most likely due to the quantitative loss of the CD4 SP subset (which are CD3 high expressing cells). When the levels of activation in the transgenic thymocytes are corrected for the depleted SP cells, they become comparable to those from non-transgenic littermates. Further analysis is required to determine if the activation level of individual thymocytes is also perturbed. These data are in direct contrast to those of others (Skowronski et al . , 1992) where, despite large losses in CD4 SP cells, Nef transgenic thymocytes are hyperactivated through anti-CD3 stimulation. The nef transgene in these mice is controlled via murine CD3 gene regulatory elements which direct expression of nef much later in T cell ontogeny than CD2 elements (Yagita et al . , 1989). Thus developmental expression differences may account for the opposing data. Alternatively, position effects on the transgene could play a role in variable NEF expression on T cells since, unlike human CD2 (Greaves et al . , 1989), no elements have been identified in the CD3 gene to confer position independent expression (Lacy et al . , 1983; Lee et al . , 1992).

At present, it is known that both DP and SP human fetal thymocytes can be infected in vitro (De Rossi et al . , 1990; Hayes et al . , 1992) and that fetal thymus tissue in utero harbours HIV (Courgnaud et al . , 1991). Since our studies demonstrate that NEF has an effect on DP and SP thymocytes, nef transgenic mice may serve as useful models of in utero or paediatric HIV infection. The thymus probably acts as a site of T cell differentiation and maturation throughout life (Steinmann, 1986) and thus dysfunction of thymopoiesis may be a pathogenic mechanism for HIV as recently suggested in the SCID-hu model (Bonyhadi et al . , 1993; Aldrovandi et al . , 1993). HIV infection of adult thymus/liver implants affects DP and SP thymocyte percentages and numbers, with the DP population harbouring greater than 90% of the virus. Taken together, these studies may provide clues as to why HIV infected patients become depleted for CD4 cells over a long period of time (Fauci, 1986). The ability of HIV to infect DP cells could lead to effects on the peripheral T cells as we observe in some of the nef transgenic mice. We are presently examining the response to antigen of CD4 cells in the peripheral lymphoid system of nef transgenic mice.

Given that NEF does cause CD4 downregulation, what is the mechanism of this action? We have shown that downregulated CD4 is sequestered intracellularly and is colocalised with a marker specific to the golgi and not one to the ER. Thus, NEF may directly interact with CD4, block it in the cytoplasm and prevent it from reaching the cell membrane or it may act indirectly through other proteins necessary for mediating cell surface appearance of CD4. It is known that HIV gpl60 env glycoprotein precursor also causes CD4 downregulation (Stevenson et al . , 1988; Crise et al . , 1990) and that HIV-l Vpu induces the rapid degradation of CD4 in the ER of HIV infected cells (Willey et al . , 1992). We show here that in the absence of gpl60, CD4 is downregulated specifically by NEF and that CD4 colocalises to the perinuclear region. Further to this, the interaction of CD4 with cytoplasmic tyrosine protein kinase p56 lck (Veilette et al . , 1988; Shaw et al . , 1989) has been examined. It has been shown that p56 lck and gpl60-CD4 form a ternary complex in the ER (Crise and Rose, 1992). Indirect immunofluorescence of p56 lck in the thymocytes from our nef transgenic mice shows no decrease in p56 lck membrane expression but some colocalisation intracellularly with CD4. It will be necessary to determine whether there is a direct association between NEF, CD4 and p56 lck. Considering the differences between NEF and gpl60 mediated CD4 downregulation and data showing that NEF is expressed at high levels (Franchini et al . , 1986) very early in the HIV life cycle from a Rev independent RNA (Haseltine, 1991), NEF most likely acts to modify CD4 localisation before the production of the env precursor gpl60. Thus, NEF, gpl60 and Vpu could perform different functions at different stages in a viral mechanism to downregulate its CD4 receptor.

Why would the virus need such an elaborate mechanism (s) to downregulate the CD4 receptor? By removing the cell surface receptor, virus producing cells can be protected from superinfection. This phenomenon is known as interference and has been well characterised for retroviruses (Weller et al . , 1980; Stevenson et al . , 1988; Crise et al . , 1990; Heard and Danos, 1991). Other viruses have enzymes to degrade their receptor, e.g. influenza virus (Muchmore and Varki, 1987) and coronavirus (Vlasak et al . , 1988). Furthermore, it has been demonstrated that high levels of CD4 block HIV virion formation and that downregulation is important for efficient production of infectious virus (Marshall et al . , 1992. Thus, our transgenic mice could serve as a useful model to further elucidate the mechanism of NEF mediated CD4 downregulation, to study effect of NEF on the host developing immune system and for the testing of NEF inhibitors that may have a therapeutic effect against HIV replication in vivo .

Example II

Transgenic mice expressing the HIV-2 tat gene were constructed, in order to study the effect of the TAT transactivator on T-cells. The tat transgene was expressed under the control of the CD2 LCR and regulatory sequences in order to achieve T-cell specific expression.

The procedures used were as for Example I, unless otherwise specified.

Exon 1 (encoding aa 1-72) of the HIV-1 tat gene was inserted downstream of the transcriptional start site in the first exon of the human CD2 gene. A stop codon was constructed in the sequence of human CD2 exon 2 so as to eliminate the production of CD2 protein. The human CD2 LCR element was ligated to the 3' end of the construct. A Sall- Notl fragment was injected into fertilized mouse eggs. At least three transgenic lines were created. Lice C (2 in Figure 9A) contains 70 copies and line E (4 in Figure 9A) contains 40 copies.

S1 nuclease RNA protection was performed on various tissues from a transgenic and a non-transgenic mouse using a tat exon 1 probe. As shown in Figure 9B, only thymus expressed TAT highly. Spleen expressed TAT only to low levels. No expression was observed in the kidney or liver of the transgenic mouse or in any of the tissues of the control non-transgenic (Figure 9C).

Thymocyte and peripheral T cell CD4 and CD8 subsets are not perturbed by expression of the CD2-tat transgene.

In order to determine whether the CD4 and CD8 T cell subsets are affected by the overexpression of HIV-TAT, antibody staining and FACS analysis was performed on thymocytes, spleen and lymph node cells fron CD2-tat transgenic mice (Figure 10). Single cell suspensions were prepared from the tissues of line C and line E transgenic mice (samples 3 and 4 respectively) and their non-transgenic littermates (samples 1 and 2). PE labelled CD8 and FITC labelled CD4 antibodies were incubated with the cells and FACS analysis performed. As shown in the contour plots, no changes in the percentage of double negative, double positive or single positive subsets were found in the thymus of transgenic mice. Furthermore, no changes in the percentage of CD4 or CD8 single positive subsets were found in the spleen or lymph nodes of the transgenic mice when compared to the non-transgenics. Thus, high level expression of HIV-TAT does no affect subset distribution in vivo in the lymphoid organs.

Cytokine gene expression is affected by the presence of HIV- TAT

Northern blot analysis was performed on RNA from CD2- tat transgenic mice to test for quantitative differences in cytokine gene expression (Figure 11). Total mRNA was prepared from thymocytes of line C transgenic mice (C+1 and C+2) and a non-transgenic littermate (C-) and from thymocytes of line E transgenic mice (E+1 and E+2) and a non-transgenic littermate (E-). 10 μg of RNA was loaded per lane on a formaldehyde agarose gel. RNA was transferred onto a filter and hybridized with a β-actin probe as a RNA quantitation control and a tat probe for verification of transgene expression. The filter was rehybridized several times with probes for cytokine genes TGF-β, IL-4R, TNF-β and TNF-α. Autoradiagrams of the Northern blot demonstrate an increase in expression of TGF-β , IL-4R and TDF-β gene expression in the TAT transgenic mice. However, hybridization signal with the TNF-α probe suggests no change or a decrease in TNF-α gene expression in the TAT positive mice.

The results of this Northern blot were quantitated on a Phosphorlmager and fold increases or decreases were calculated. As shown in the table, when signal was normalized against the β-actin quantitation control, TNF-β, IL-4R and TGF-β steady state mRNA was increased in the transgenic mice. TGF-β levels increased on average 2.8 fold, IL-4R levels increased 3.5 fold and TNF β levels increased 2.3 fold. TNF-α levels decreased by about 25%. Thus, the expression of TAT in thymocytes has an effect on cytokine gene expression. Quantitation of Cytokine RNAs in CD2-TATTransgenic Mice

Fold RNA change as compared to non-transgenic littermates. TAT induced transcriptional upregulation of TNF-β leads to overproduction of functional TNF-β as measured by cytotoxicity.

Northern blot analysis of RNA from TAT transgenic lines C and E demonstrated an increase of 2.3 fold in TNF- β transcription. In order to test whether this resulted in increased TNF-β protein production, we performed cytotoxicity assays with cell lysates of T cells from TAT transgenic mice (C+1, C+2, E+1 and E+2) and non-transgenic littermates (C- and E-). Active equivalents of TNF-β protein in each sample was quantitated against a known TNF- β protein standard (in units). As shown in Figure 12, all four transgenic mice produced significantly higher levels of TNF-β (2-4 fold) as compared to the non-transgenic controls.

The invention is described hereinbefore by way of example only, and numerous modifications of detail will be apparent to those skilled in the art.

References

Ahmad, N. and Venkatesan, S (1988) Science, 241, 1481-1485.

Aldrovandi, G.M., et al., (1993) Nature, 363. 732-736.

Backer, J.M. et al., (1991) AIDS Res.Hum.Retroviruses, 7, 1015-1020.

Berg, L.J., et al., (1989) Cell, 58, 1035-1046.

Boehmelt et al., (1992) EMBO J., 11, 464-4652

Bonifer et al., (1990) EMBO J. , 9, 2843-2848

Bonyhadi, M.L. et al., (1993) Nature, 363, 728-732

Carson, S., and Wills M.V. (1993), NAR, 21, 2065-2072

Cosgrove, D. et al., (1991) Cell, 66, 1051-1066.

Courgnaud, V., et al., (1991) AIDS Res.Hum.Retroviruses, 7,

337-341.

Crise, B., et al., (1990) J.Virol., 64, 5585-5593.

Crise, B. and Rose, J.K. (1992) J.Virol., 66, 2296-2301. de Ronde, A., et al (1992) Virology, 188, 391-395.

de Rossi, A., et al., (1990) AIDS Res.Hum.Retroviruses, 6,

287-298.

Fauci, A.S. (1986) Proc.Natl.Acad. Sci. USA, 83, 9278-9283. Franchini, G. et al., (1986) Virology, 155, 593-599.

Fraser et al., (1990) ???

Garcia, J.V. and Miller, A.D. (1991) Nature, 350, 508-511.

Garcia, J.V. et al., (1993), J.Virol, 67, 1511-1516.

Greaves, D.R. et al., (1989) Cell, 56, 979-986.

Gatz and Quail, (1988) PNAS, 85., 1394-1397.

Gossen and Bujard (1992) PNAS, 89., 5547-5553.

Grosveld, F.G. et al., (1987) Cell, 51, 975-985.

Guy, B. et al., (1987), Nature, 330, 266-269.

Guy, B. et al., (1990) Virology, 126, 413-425.

Hammes, S.R. et al., (1989), Proc.Natl.Acad. Sci USA, 86..

9549-9553.

Haseltine, W.A. (1991), FASEB J., 5, 2349-2360.

Havran, W.L. et al., (1987) Nature, 300, 170-173.

Hayes, E.F. et al., (1992) AIDS, 6, 265-272.

Heard, J.M. and Danos, o. (1991) J.Virol., 65., 4026-4032.

Hu and Davidson, (1987), Cell, 48, 555-566.

Kakidani and Ptashne (1988), Cell, 56, 161-167.

Kamoun, M. et al., (1981) J.Exp.Med., 153, 207-212. Kestler, H.W. et al., (1991) Cell, 65, 651-662

Kim, S., et al., (1989) Proc. Natl. Acad. Sci. USA, 86,

9544-9548.

Kruisbeek, A.M. et al., (1983) J.Exp.Med., 157, 1932-1946. Lacy, E. et al., (1983), Cell, 34, 343-350.

Lang, G. et al., (1988) EMBO J. , 7, 1675-1682.

Lee, N.A. et al., (1992) J.Exp.Med., 175, 1013-1025.

Luria, S. et al., (1991) Proc.Natl.Acad. Sci. USA, 88, 5326-

5330.

Marshall, W.L. et al., (1992) J.Virol., 66, 5492-5499.

Muchmore, E.A. and Varki, A. (1987) Science, 236, 1293-1295.

Ovod, V. et al., (1992), AIDS, 6, 25-34.

Owen, M.J. et al., (1988), Eur. J. Immunol., 18 ., 187-189.

Picard, et al., (1988), Cell, 54, 1073-1080.

Rahemtulla, A, et al., (1991) Nature, 353, 180-184.

Ratner, L. et al., (1985) Nucleic Acids Res., 13, 8219-8229.

Robert-Guroff, M. et al., (1990), J.Virol., 64, 3391-3398.

Sambrook, J. et al., (1989) Molecular Cloning: A Laboratory

Manual. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, New York.

Schwartz, O. et al., (1992) Aids Research Hum.Retroviruses,

8, 545-551.

Schwartz, O. et al., (1993) J..Virol., 67, 3274-3280

Shaw, A.S. et al., (1989) Cell, 59, 627-636.

Skowronski, J. et al., (1993) EMBO J., 12, 703-713.

Steinmann, G.G. (1986) Curr. Top. Path., 75, 43-88.

Stevenson, M. et al., (1988) Cell, 53, 483-496.

Veilette, A. et al., (1988) Cell, 55, 301-308.

Vlasak, R. et al., (1988) Proc. Natl. Acad. Sci. USA, 85, 4526-4529.

Weiss, A. et al., (1987) J. Immunol., 139, 3245-3250.

Weller, S.K. et al., (1980) J.Virol., 33, 494-506.

Willey, R.L. et al., (1992), J.Virol., 66, 7193-7200.

Claims

CLAIMS :

1. A transgenic mammal comprising a heterologous gene encoding an effector associated with a disease, the gene comprising at least one control sequence effective to direct expression of the effector substantially exclusively to cells where the effector is expressed in the normal course of the disease.

2. A transgenic mammal according to claim 1 wherein the effector is a polypeptide.

3. A transgenic mammal according to claim 1 or claim 2 wherein the disease is a human disease.

4. A transgenic mammal according to any preceding claim wherein the disease is a viral disease.

5. A transgenic mammal according to claim 4 wherein the effector is a viral regulatory protein.

6. A transgenic mammal according to claim 5 wherein the viral regulatory protein is a HIV regulatory protein selected from NEF, TAT, REV, VPU, VIF and VPR.

7. A transgenic mammal according to any preceding claim wherein the control sequences of the heterologous gene comprise a Locus Control Region (LCR).

8. A transgenic mammal according to any preceding claim wherein the control sequences of the heterologous gene comprise a tissue specifically or developmentally regulated promoter element.

9. A transgenic mammal comprising the coding sequence of a HIV regulatory protein under the control of the CD2 LCR.

10. The use of a gene encoding an effector associated with a disease and comprising at least one control sequence effective to direct expression of the effector substantially exclusively to cells where the effector is expressed in the normal course of the disease in the generation of a transgenic mammal for use as a model for the disease.

11. A method for studying a potential therapeutic agent for a disease comprising administering the agent to a transgenic mammal according to any one of claims 1 to 9.

12. A transgenic mammal as hereinbefore described with reference to the examples.