WO2005080574A1

WO2005080574A1 - Transgenic mouse model and method for evaluating glucocerebrosidase deficiencies

Info

Publication number: WO2005080574A1
Application number: PCT/CA2005/000266
Authority: WO
Inventors: Lorne Clarke; Graham Sinclair; Francis Choy
Original assignee: The University Of British Columbia
Priority date: 2004-02-24
Filing date: 2005-02-24
Publication date: 2005-09-01

Abstract

A recombinant mouse model in which the genome of the mouse is modified by introduction of paired recombinase recognition sites integrated in the glucocerebrosidase gene in the genome of the transgenic mouse is useful in assessing glucerebrosidase deficiency. The recombinase recognition sites flank a portion of the glucocerebrosidase gene including as least one exon of the glucocerebrosidase gene and are oriented such that the portion of the glucocerebrosidase gene is excised in the presence of a recombinase that recognizes the paired recombinase recognition sites. Upon exposure to the appropriate recombinase, the mouse produces defective glucocerebrosidase in a manner which can be controlled in a temporal, spatial or cell-type- specific manner.

Description

Transgenic Mouse Model and Method for Evaluating Glucocerebrosidase Deficiencies

DESCRIPTION

Priority Claim

This application claims priority form US Provisional Application 60/546,986, filed February 24, 2004, which application is incorporated herein by reference.

Field of the Invention This application relates to a mouse model and to methods of using the mouse model in the evaluation of glucocerbrosidase deficiences, such as Gaucher Disease.

Background of the Invention Inherited deficiency of the lysosomal hydrolase, glucocerebrosidase (GBA, acid β- glucosidase) EC 3.2.1.45) underlies the autosomal recessive disorder, Gaucher disease. This disorder represents the most common lysosomal storage disorder and is the first lysosomal disorder to be treated by direct enzyme replacement strategies. The global disease incidence is estimated to be 1/50,000 to 1/100,000 individuals, although incidence rates increase to 1/850 in those of Ashkenazi Jewish descent. Although three clinical subtypes are described, type 1 (non-neuronopathic), type 2 (acute neuronopathic), and type 3 (sub-acute neuronopathic), type 1 represents the most prevalent form. While enzyme replacement strategies have been effective in treating some of the complications of type 1 disease, this form of therapy is not generally helpful for type 2 or 3 patients. All forms of this disorder share a group of classic "Gaucher" symptoms consisting of hepatosplenomegaly, pancytopenia, frequent bone crises, and the presence of lipid engorged macrophages known as Gaucher cells. Types 2 and 3 Gaucher disease also include neuronal loss leading to extensive neurological dysfunction in those affected (Beutler and Grabowski, 2001). Although over 100 disease causing mutations in the glucocerebrosidase gene have been identified, there exists a poor correlation between genotype and phenotype with little predictive power, h fact, many patients with well documented common mutations can be relatively asymptomatic. A clear etiology for the onset of hepatosplenomegaly and pancytopenia in this disorder does exist due to the pathological storage of lipid-laden macrophages in the liver, spleen, and bone marrow. Unfortunately, no clear mechanistic model currently exists for the onset of skeletal involvement in this disorder or for neurodegeneration in the neuropathic forms. The study of sphingolipid metabolism in Gaucher disease has been limited to cell culture and in vitro investigations due to the absence of an appropriate animal model. Due to the complexity of disease progression it is clear that a whole organism in vivo approach is required to clarify the pathophysiology of this complex disease. In addition to the study of disease pathogenesis advancements in the development of therapies including direct enzyme replacement, small molecule enzyme chaparonins, substrate inhibitors and gene based therapy would be accelerated by the availability of a glucocerebrosidase deficient animal model. Attempts in the last decade to create a mouse model for complete GBA deficiency or to create mice with severe human point mutations in the GBA gene have generally resulted in a non-viable, perinatal lethal phenotype. Further, until recently, attempts to create mice with common point mutations, or chemically induced models for GBA deficiency have led to animals with no obvious phenotype. Xu et al., Amer. J. Pathol 163: 2093-2101 (2003) disclose mouse models containing specific point mutations in the GBA gene. Mice with these mutations are viable, and exhibit phenotypes in which both glucocerebrosidase storage and the production of small numbers of lipid-engorged Gaucher cells are observed. These phenotypes develop only to a limited extent, however, and are only reported after at least 7 months of life.

Summary of the Invention The present invention provides a recombinant mouse model, wherein the genome of the mouse is modified by introduction of paired recombinase recognition sites integrated in the glucocerebrosidase gene in the genome of the transgenic mouse. The recombinase recognition sites flank a portion of the glucocerebrosidase gene including as least one exon of the glucocerebrosidase gene and are oriented such that the portion of the glucocerebrosidase gene is excised in the presence of a recombinase that recognizes the paired recombinase recognition sites. In an embodiment of the invention, the genome of the mouse is also modified by introduction of a recombinase transgene encoding the recombinase that recognizes the paired recombinase recognition sites. The recombinase transgene may be introduced by crossing the mouse of the invention having recombinase cites integrated in the glucocerebrosidase gene with a recombinant mouse known to express the recombinase to produce progeny mice, which are themselves within the scope of the invention, having both the modified glucocerebrosidase gene and a recombinase transgene. Alternatively, the recombinase transgene can be introduced to mice in accordance with the invention at a selected developmental stage, for example using a viral vector containing the recombinase transgene. Other methods for exposing cells to recombinase may also be used. specific embodiments of the invention, the recombinase recognition sites are be recognized by Flpe recombinase or Cre recombinase. Expression of the recombinase encoded by the transgene is under the control of tissue-specific promoter. By way of non-limiting example, the tissue-specific promoter may be specific to hematopoietic cells or to CNS-specific promoter. The model mouse system of the invention is useful for evaluation of glucose cerebrosidase deficiencies and offers greater versatility over the Xu et al. mouse. The mouse model of the invention can be used to evaluate different types of conditions associated with glucocerebrosidase deficiency because me expression of the recombinase, and hence the occurrence of a glucocerebrosidase deficiency, can be controlled in a tissue-specific manner. Further, because the onset of glucocerebrosidase deficiency can be controlled temporally, the affects of the deficiency can be observed earlier, including in utero, or later, as desired for a given condition. This model also has the capability of providing a resource of murine cells lines and/or tissue extracts, not restricted to the following examples: bone marrow, fibroblasts, neural stem cells and hepatocytes for invitro investigations of GBA deficiency and/or invitro CRE excision studies.

Brief Description of the Figures

Fig 1A shows the native exon structure of GBA;

Fig. IB shows a targeting construct in accordance with the invention;

Fig. 1C shows the targeting construct after removal of the selection marker; and

Fig. ID shows the construct after excision of a portion of the GBA gene.

Detailed Description of the Invention The following definitions are relevant to the present application and the claimed invention: A "recombinase" is an enzyme that catalyzes new phosphodiester bonds between a pair of short, unique DNA target sequences, or "recombinase recognition sites". Examples of recombinase enzymes include the Cre enzyme, derived from bacteriophage PI, which recognizes lox sites as described in US Patent No. 4,959,317, which is incorporated herein by reference; and the FLP enzyme, derived from Saccharomyces cerevisiae, which recognizes ERE sites as described in US Patent No. 6,774,279, which is incorporated herein by reference. The lox? recognition site has the sequence

ATAACTTCGATATAGCATACATTATACGAAGTTAT (Seq. ID No: 1).

The FRT recognition site has the sequence

GAAGTTCCTATTCTCTAGAAAGTATAGGAACTTC (Seq. ID No. 2).

Other Lox and FRT recognition sites are known, for example as set forth in Branda et al., Developmental Cell(2004) 6: 7-28, as are other recombinases, and these may also be used in the invention. The term "recombinase mouse" refers to a mouse that has a transgene for a recombinase. Recombinase mice are known in the art, and are commercially available. For example, The Jackson Laboratory, offers for sale numerous mice strains which express Cre or FLP in different tissues and under the control of different promoters. The term "tissue specific expression" refers to expression of the recombinase in a subsets of the tissues, as opposed to expression across all tissues in the mouse. Tissue specific expression does require expression in only one tissue.

METHOD OF MAKING THE MOUSE MODEL OF THE INVENTION The invention described herein utilizes refinements of recombinase-mediated gene targeting strategies to produce a murine model of glucocerebrosidase deficiency. This can be controlled in a tissue specific and development specific manner. By introducing the recombinase (for example Cre) into the system in a spacio-temporally controlled manner, the lethal phenotypes observed in other attempted GBA knockout models are avoided. The mouse model of the invention is homozygous for recombinase site insertions flanking part of the glucocerebrosidase (GBA) gene. Thus, the mouse model gives researchers the ability to control temporal GBA expression and tissue specific GBA expression by exposing said animal, or specific tissues within the animal, to recombinases such as Cre or Flpe. The recombinase will remove part of the GBA gene, thus eliminating or attenuating GBA activity and allowing the phenotype of Gaucher disease and/or other phenotypic features of GBA deficiency to develop within the animal. However, the lethal development aspects that occur from standard or prior art embryonic stem cell GBA knockout strategies will be prevented, and non-lethal knock out or attenuation will occur only when investigator directed recombinase exposure occurs. Figs. 1 A-D show a GBA locus and targeting construct in accordance with the invention, and are illustrative of the method and recombinant genome of the mice of the invention . Fig. 1A shows the native exon structure of GBA is depicted in black and the MTX (metaxin) locus in grey. The nucleotide numbering is in reference to Genbank Accession # AY115108 (Seq. ID No. : 3) and the Sea I cutsites and external probe used for Southern blot analyses are also noted. Fig. IB shows te targeting construct including a neomycin resistance marker (Neo¹), Cre recognition sites (loxP) and Flpe recognition sites (FRT). The Sea I site introduced by the Neo^r was used as a diagnostic marker for the Southern analysis. Fig. 1C shows the targeted locus following Flpe-mediated removal of the neomycin resistance marker. This sequence differs from the native locus only by the addition of one FRT and two loxP sites. Fig. ID shows the final recombined, non-functional, GBA knockout allele following Cre recombination and excision of exons 9-11. In accordance with the method of the invention, a sequence spanning at least a portion of the GBA gene and at least a portion of the MTX gene is replaced with a construct such as that shown in Fig. IB. The construct comprises, from the 5' to the 3' end, a first portion of the GBA gene, a first recombinase recognition site, a second portion of the GBA gene, a second recombinase recognition site, recognized by the same recombinase as the first recombinase recognition site, and optionally a third portion of the GBA gene. In Fig. IB, there is no third portion of the GBA gene because the excised portion of the gene is the 3' terminal exons (9-11). However, interior exons could also be excised. In Fig. IB, the first and second recombinase recognition sites are loxP recognition sites. However, any pair of recombinase recognition sites oriented for excision (deletion) of the intervening sequence could be utilized. Continuing along the targeting construct, there is a third recombinase recognition site, that is targeted by a different recombinase than the first and second recognition sites, a marker gene, and a fourth recombinase recognition site of the same type as the third recombinase recognition site. In Fig. IB, the third and fourth recombinase recognition sites are FRT sites, but again all that is important is the difference. The marker shown is Neo^r, but any resistance or other selectable marker could be used. A construct as shown in the Fig. IB is introduced into totipotent mouse embryonic stem cells for homologous recombination at the GBA locus. The selection marker (e.g. Neo¹) is used for selection of transformed cells. The selected cells are then treated to introduce an recombinase expression vector (an Flpe recombinase in the case of the construct in Fig. IB) to remove the selection marker, leaving embryonic stem cells with a structure as shown in Fig. lC. These cells were screened to select for sensitive cells (e.g. Neo sensitive cells). These cells are the introduced into blastocyst to produce recombinant mice of the invention. These mice will produce a functional glucocerebrosidase product until such time as the appropriate recombinase (Cre in the case of the construct in Fig. 1C) is present.

METHODS OF EXPOSING THE MOUSE MODEL TO RECOMBINASE Recombinase exposure can occur by crossing the animal described herein with a recombinase mouse with known recombinase expression patterns, or exposing the animal to viral vectors capable of infecting the animals cells and tissues and delivering a functional recombinase gene. Other methods for gene delivery include liposome delivery of recombinase genes, or other transfection techniques described in the art. Expression of recombinase can be controlled through various means. Various means to express CRE recombinase in an inducible manner to provide temporal, spatial and cell-type specific control of Cre-mediated DNA recombination in transgenic mice are described in Utomo, et al., Nature Biotechnol.(l 999), 17, 1091-1096, which is incorporated herein by reference. These options for control provide great diversity for the model of the invention. See also, Branda et al., Developmental Cell (2004). 6(l):7-28, which is incorporated herein by reference. Mechanisms for delivery of recombinase proteins per se to target cells also exist. A review of such methods is provided in Cleland et al. Current Opinion in Biotechnology 2001, 12:212-219, which is incorporated herein by reference. For example, electroporation can be used. Although this method was first used to introduce foreign DNA into mammalian, plant and bacterial cells, it will also permit the introduction of proteins. Cells suspended in a buffered solution of the purified protein of interest are placed in a pulsed electrical field. Brief, high-voltage electric pulses result in the formation of small (nanometer-sized) pores in the cell membrane. Proteins enter the cell via these small pores or during the process of membrane reorganization as the pores close and the cell returns to its normal state. The efficiency of delivery is dependent upon the strength of the applied electrical field, the length of the pulses, temperature and the composition of the buffered medium. Electroporation is successful with a variety of cell types, even some cell lines that are resistant to other delivery methods, although the overall efficiency is often quite low. The electroporation process can be damaging to the cells, frequently killing the majority of the cells used. As a result, considerable optimization is required to obtain efficient delivery and nώiimize cell death. Also, some primary lymphocyte cell lines remain refractory even to electroporation unless partially activated, eliminating the possibility of any downstream studies relevant to cell activation or cell cycle control. Microinjection may also be used to introduce recombinase proteins. Microinjection was first used to introduce femtoliter volumes of DNA directly into the nucleus of a cell, where it can be integrated directly into the host cell genome, thus creating an established cell line bearing the sequence of interest. Proteins such as antibodies and mutant proteins can also be directly delivered into cells via microinjection to determine their effects on cellular processes first hand. Microinjection has the advantage of introducing macromolecules directly into the cell, thereby bypassing exposure to potentially undesirable cellular compartments such as low-pH endosomes. However, the art of microinjection requires extensive training to master and specialized equipment including micropipette pullers, microinjectors and inverted microscopes equipped with micromanipulators. Viral protein fusions can be used to introduce recombinases. Several proteins and small peptides have the ability to transduce or travel through biological membranes independent of classical receptor- or endocytosis-mediated pathways. Examples of these proteins include the HIV-1 TAT protein, the herpes simplex vims 1 (HSV-1) DNA-binding protein VP22, and the Drosophila Antennapedia (Antp) homeotic transcription factor. The small protein transduction domains (PTDs) from these proteins can be fused to other macromolecules, peptides or proteins to successfully transport them into a cell. Sequence alignments of the transduction domains from these proteins show a high basic amino acid content (Lys and Arg) which may facilitate interaction of these regions with negatively charged lipids in the membrane. Secondary structure analyses show no consistent structure between all three domains. The advantages of using fusions of these transduction domains is that protein entry is rapid, concentration-dependent and appears to work with difficult cell types. However, each of the three commonly used PTDs has its own unique considerations. While fusion proteins of >1,000 residues is possible with the TAT and VP22 approaches, the size of the protein being transduced as an Antp fusion is limited to <100 residues. A major disadvantage of creating a TAT fusion protein is that the transduced protein is denatured or inactivated as it passes through the membrane. Thus, upon entry into the cell, the protein needs to be properly refolded to regain its biological activity. Currently, the VP22 strategy is used as an indirect method in that the vector bearing the fusion construct is transfected into cells where the fusion protein is made and the resulting protein then transduces into surrounding cells. A drawback to all of the PTD-mediated protein delivery systems is that the transduction domain must be covalently attached to the protein being delivered, either by creating a DNA construct in a specially designed vector or by chemically cross-linking the protein and PTD via functional groups on each molecule. (See, US 6,348,185; U.S. Patent Publication US 2003/0229202, and PCT publication WO 00/62067). Liposomes have been rigorously investigated as vehicles to deliver oligonucleotides, DNA (gene) constructs and small drug molecules into cells. Certain lipids, when placed in an aqueous solution and sonicated, form closed vesicles consisting of a circularized lipid bilayer surrounding an aqueous compartment. These vesicles or liposomes can be formed in a solution containing the molecule to be delivered. In addition to encapsulating DNA in an aqueous solution, canonic liposomes can spontaneously and efficiently fonn complexes with DNA, with the positively charged head groups on the lipids interacting with the negatively charged backbone of the DNA. The exact composition and/or mixture of cationic lipids used can be altered, depending upon the macromolecule of interest and the cell type used. The cationic liposome strategy has also been applied successfully to protein delivery. Because proteins are more heterogeneous than DNA, the physical characteristics of the protein such as its charge and hydrophobicity will influence the extent of its interaction with the cationic lipids. See, US 6376248 and US2004/0023391A1. Pro-Ject™ Protein Transfection Reagent (Pierce, Product # 89850) which utilizes a cationic lipid formulation that is noncytotoxic and is capable of delivering a variety of proteins into numerous cell types. The protein being studied is mixed with the liposome reagent and is overlayed onto cultured cells. The liposome:protein complex fuses with the cell membrane or is internalized via an endosome. The protein or macromolecule of interest is released from the complex into the cytoplasm free of lipids and escaping lysosomal degradation. The noncovalent nature of these complexes is a major advantage of the liposome strategy as the delivered protein is not modified and therefore maintains its activity. As with all direct protein delivery systems, the time saved is significant over the indirect DNA transfection procedures.

METHOD OF USING THE MICE OF THE INVENTION The animal described herein can be used for evaluating therapeutic agents for use in treating Gaucher disease or other disease states caused by GBA deficiency. In one embodiment of the invention, the screening protocol involves therapies to provide glucocerebrosidase activity to those tissues, or whole animals, in which GBA activity has been removed by exposure of the GBA gene to recombinases capable of recognizing the appropriate sites flanking the GBA gene. Thus, the ability of the therapy to reconstitute GBA activity can be assessed. In another embodiment of the invention, the mice of the invention are used to monitor the influx of lipid laden macrophages to the bone marrow of GBA compromised recombinant mice. Thus, potential treatments for chronic bone disease and/or chronic bone pain can also be evaluated using the recombinant mouse described herein. An effective therapy would be marked by lower bone migration of macrophages and/or marked lessening of the macrophages laden with lipid. In addition to monitoring bone crisis within the recombinant animal for the purposes of screening therapeutics, other therapeutics for pathologies associated with Gaucher such as hepatomegaly (abdominal pain), splenomegaly (enlarged spleen), cardiac alterations, pulmonary alterations, and neurologic manifestations could be screened. The recombinant animal described herein can also be used to evaluate the ability of a targeting system to deliver a therapeutic agent to selected tissues or organs which have been compromised by selective recombinase removal of the GBA gene. In this way a therapeutics' potential to cleave glucosylcera ide or alter the glucosylcera ide pathway, can be monitored and evaluated, whether the therapeutic is an in vivo or ex vivo gene therapy technique, enzyme replacement therapy, or some form of compound capable of cleaving, or enhancing the cleavage or clearance of glucosylceramide, stabilizing a mutated endogenous GBA protein, or other means by which glucosylceramide levels can be altered. Other important uses of the mice of this invention relate to studies directed to the understanding of the role that GBA its substrates and products play, in metabolism in general. Important aspects would include, but not be restricted to, the role of GBA in maintenance of skin permeability and neuronal function. The invention will now be further illustrated with reference to the following non-limiting examples.

Example 1 A targeting construct was designed to utilize the pFLRT3 vector containing both Cre and Flpe recombinase recognition sites for these engineering steps. This allows for the removal of the Neomycin selectable marker using Flpe-recombinase, leaving the GBA locus intact with only the addition of two CRE recombinase sites, and a single FRT site. The targeting construct pFLRTGBA (Fig. 1A) was designed such that a 1.2kb PCR fragment of the GBA gene obtained using Seq. ID Nos. 3 and 4 as primers, encoding the active site of the protein (exons 9-11) would be flanked by loxP sites and excised in the presence of the Cre recombinase. Upstream of the first loxP site, a 6.7kb fragment encoding the 5'UTR and exons 1-8 of GBA was obtained using primers of Seq. ID Nos. 5 and 6, and cloned-in using PCR-based compatible linkers at the extremities and a large subcloned fragment for the remainder of this long arm. Downstream of the second FRT site, a 2.9kb fragment containing exons 8-4 of the Metaxin gene was introduced using a PCR-based approach and primers of Seq ID Nos: 7 and 8. MXT is contiguous (on the reverse strand) with GBA on mouse chromosome 3 and its function is required for viability. All cloning steps were performed using proofreading polymerases and the construct designed such that all cloning junctions occurred in intronic or intergenic regions. All coding and flanking intronic regions of the final construct were sequenced to confirm cloning and PCR fidelity. Cre and Flp expressing strains of E. coli (294-Cre and 294-Flp) were utilized to corifirm functionality of the loxP and FRT sites for recombination. Transformation of each of these E. coli strains with the final pFLRTGBA construct resulted in excision of the appropriate fragment of the plasmid as confirmed by PCR analysis. Example 2 The development of a knockout mouse utilizes the introduction of the targeting vector to totipotent mouse ES cells for homologous recombination of the transgene at the locus of interest. The inclusion of a Neomycin resistance marker in the targeting construct allows for selection of recombinants following transfection of ES cells. Mouse embryonic stems cells were expanded from a pass 13 stock following estabUshed ES cell culture protocols. The targeting plasmid was linearized and electroporated into the ES cells under a range of DNA concentrations (10-40ug per electroporation). Following 14 days of selection on 150ug/ml G418 antibiotic, a total of 480 colonies were picked into 96-well plates for expansion, freezing, and genetic analysis. Clones were screened by PCR and Southern blotting for correct targeted integration of the recombinant allele. The PCR screening strategy was designed to amplify a fragment spanning the short arm of the construct (Metaxin exons 8-4) with one primer in vector sequence and the second primer in the mouse genomic DNA downstream of the short arm and thus outside of the construct. Only correctly integrated plasmid would be amplified and two PCR-positive clones were selected for confiπnation by Southern blotting. Using a genomic DNA probe 3' to the construct (Seq. ID Nos. 10 and 11) and a restriction digest exploiting an internal Sea I cutsite introduced by the neomycin selectable marker, the two correctly targeted clones were confirmed. The two confirmed clones were then subjected to a second round of electroporation and selection to remove the neomycin marker in preparation for blastocyst injection. The recombinant ES cells were electroporated with the Flpe recombinase expression plasmid, pCAGGS-Flpe, and transiently selected with lug ml puromycin for 48 hours. Five days after replating at low density, 600 colonies were picked and split to pairs of 96-well plates. One plate of each pair was selected with 200ug/ml G418 to select for neomycin sensitive (Flpe recombined) clones. Of the original 600 clones, 24 showed sensitivity to the G418 selection and removal of the neomycin marker in all of these clones was identified by a PCR screen and finally confirmed by Southern blotting. PCR confirming the removal of the neomycin marker from Flpe-recombined clones utilized primers of Seq. ID Nos: 12 and 14 to flank the neomycin marker, yielding a 2.2kb fragment in its absence and a 4.5kb fragment in its presence. An internal neomycin forward primer (Seq. ID No.: 13) in combination with primer Seq ID No. 14 was used to test for the presence of the marker. Example 3 The final ES cell clones, containing one GBA locus altered to include loxP sites flanking exons 8-1 l(FloxGBA), were used to create chimeras by blastocyst injection. Using estabUshed techniques, the FloxGBA Rl ES cells were injected into the inner cell mass of day 3.5 blastocysts harvested from superovulated females. As the Rl ES ceU line originated from (129Sv x 129/J)F1 hybrid mice displaying an agouti coat, C57BL/6J (black) blastocysts were utilized for the injection. Once injected and implanted into pseudopregnant host mothers, the blastocysts developed into chimeric mice displaying a patchwork mix of agouti and black coat colours. Three chimeras displaying a high degree of agouti coat (high contribution from the FloxGBA ES cells) were bread to C57BL/6J mice to produce animals 100% hetero:zygous for the FloxGBA allele. These mice possess the conditional glucocerebrosidase knockout in the presence of Cre-recombinase when bred to homozygosity.

Example 4 The possible variations of GBA knockout using this system are limited only by the diversity of Cre-expressing mouse strains as well as other methods to introduce Cre to the mouse. As a specific example of the use of a breeding strategy, mice with a combination of ubiquitous and hematopoietic cell-specific knockout are obtained are obtained utilizing the Tie2-Cre (B6.Cg-Tg(Tekcre) 12Flv) mouse expressing Cre under control of the receptor tyrosine kinase Tek promoter/enhancer (Koni et al. 2001). Cre expression in this animal is limited to endotheUal and hematopoietic cells and the female germline, aUowing for tissue specific knockout of the GBA locus in the first generation and ubiquitous recombination when the FloxGBA aUele is transmitted through the maternal lineage to the second generation. In the presence of the Cre-recombinase, the FloxGBA allele wiU recombine to produce a non-functional GBA aUele only in those tissues expressing Cre (Fig. ID). A hematopoietic ceU-specific knockout leads to glucocerebroside storage in circulating macrophages and subsequent deposition of these "Gaucher cells" in the Uver, spleen, and bone marrow of these mice. In this way, the model recapitulates the major visceral symptoms associated with human Gaucher disease while avoiding the lethal epidermal abnormaUties observed in traditional knockout animals. The abiUty to generate ubiquitous knockout mice through the maternal lineage also serves as a control to show that total knockout of the locus is, in fact, lethal in this model system. Numerous other Cre expression strategies targeting other tissues, or under temporal control, could be utilized to explore the complexities of sphingoUpid metaboUsm in the mouse. This would include the use of Uver specific CRE expression, CNS specific CRE expression, epidermal specific CRE expression and partial ubiquitous CRE expression. Branda, supra, Nagy, A. and Mar, L. (2001) Creation and Use of a Cre Recombinase Transgenic Database. In Methods in Molecular Biology, vol. 158:95-106.: Gene Knockout Protocols Edited by: M. J. Tymms and I. Kola, Humana Press Inc., Totowa, NJ; Nagy, A. (2000) Cre recombinase: the universal reagent for genome tailoring. Genesis 26(2): 99-109.

Example 5 Initial chimeric mice and the hetero and homozygotes produced from further matings are genotyped using a combination of PCR and Southern blot analyses. Our starting point for Cre expression studies was a homozygous FloxGBA mouse. Through a combination of histological and biochemical studies of affected tissues (liver, spleen, bone marrow, brain, etc.) normal tissue pathology is estabUshed as well as activity levels of GBA and other marker enzymes in these and control mice. An important detail in this portion of the work is the estabUshment of the extent and distribution of the Cre-recombinase mediated excision events in these mice. While PCR and Southern blotting may identify the presence of genetic mosaicism, the patterns of this mosaic will only be estabUshed through in situ antibody or activity staining for GBA. Although no anti-mouse GBA antibody is pubUcly available, an in situ fluorescent activity assay for GBA has been developed. Mouse tissue sections, Ughtly fixed in paraformaldehyde, can be investigated by confocal microscopy for localized activity on the artificial GBA substrate, 4MUGP.

Claims

CLAΓJVIS

1. A recombinant mouse, wherein the genome of the mouse is modified by introduction of paired recombinase recognition sites integrated in the glucocerebrosidase gene in the genome of the transgenic mouse, said recombinase recognition sites flanking a portion of the glucocerebrosidase gene including at least one exon of the glucocerebrosidase gene and being oriented such that the portion of the glucocerebrosidase gene is excised in the presence of a recombinase that recognizes the paired recombinase recognition site.

2. The recombinant mouse of claim 1, wherein the recombinase recognition sites are recognized by Flpe recombinase.

3. The recombinant mouse of claim 1, wherein the recombinase recognition sites are recognized by Cre recombinase.

4. The recombinant mouse of any of claims 1 to 3, wherein the genome of the mouse is further modified by introduction of a recombinase transgene encoding the recombinase that recognizes the paired recombinase recognition site.

5. The recombinant mouse of claim 4, wherein the expression of the recombinase is under the control of tissue-specific promoter.

6. The recombinant mouse of claim 5, wherein the tissue-specific promoter is specific to hematopoietic ceUs.

7. The recombinant mouse of claim 5, wherein the tissue-specific promoter is specific to CNS ceUs.

8. A method for making a recombinant mouse comprising the steps of: (a) transforrning murine embryonic stem ceUs with a targeting construct to fonn recombinant embryonic stem cells comprising a transgene encoding glucocerebrosidase, wherein the transgene encoding glucocerebrosidase comprises paired first recombinase recognition sites integrated in the glucocerebrosidase gene in the genome of the embryonic stem ceUs, said pair of first recombinase recognition sites flanking a portion of the glucocerebrosidase gene including at least one exon of the glucocerebrosidase gene and being oriented such that the portion of the glucocerebrosidase gene is excised in the presence of a first recombinase that recognizes the paired recombinase recognition site and a selection marker flanked by a pair of second recombinase recognition sites, said pair of second recombinase recognition sites being oriented such that the selection marker is excised in the presence of a second recombinase different from the first recombinase; (b) selecting transformed cells using the selection marker; (c) exposing the selected transformed ceUs to the second recombinase to excise the selection gene; (d) selecting marker-sensitive cells from the cells exposed to the second recombinase; (e) introducing selected marker sensitive ceUs into mouse blastocysts; and (f) introducing the injected blastocysts into a host mother and aUowing the blastocyst to develop into a recombinant mouse.

9. The method of claim 8, wherein the second recombinase is Flpe recombinase.

10. The method of claim 8 or 9, wherein the first recombinase is Cre recombinase.

11. The method of any of claims 8 to 10, further comprising the step of breeding the recombinant mouse produced in step (f) with a mouse expressing the first recombinase to produce a progeny recombinant mouse.

12. The method of claim 11 , wherein the mouse expressing the first recombinase expresses the first recombinase in a tissue specific manner.

13. The recombinant mouse of claim 11 , wherein the mouse expressing the first recombinase expresses the first recombinase in hematopoietic ceUs.

14. The recombinant mouse of claim 11, wherein the mouse expressing the first recombinase expresses the first recombinase in CNS cells.

15. A method for evaluating a candidate therapeutic agent for treatment of glucocerebrosidase deficiency, comprising the step of introducing the candidate therapeutic agent to a recombinant mouse according to any of claims 1 to 7 or made in accordance with any of claims 8 to 14 in which a portion of the glucocerebrosidase gene has been excised by exposure to the recombinase, and monitoring the extent of phenotypes associated with glucocerebrosidase deficiency.

16. A targeting nucleic acid construct comprising, in sequence, from 5' to 3', (a) a first portion of a mouse glucocerebrosidase (GBA) gene, (b) a first recombinase recognition site, (c) a second portion of the GBA gene, (d) a second recombinase recognition site, recognized by the same recombinase as the first recombinase recognition site, said first and second recombinase selection sites being oriented for excision of the second portion of the GBA gene; (e) a third recombinase recognition site, that is targeted by a different recombinase than the first and second recombinase recognition sites; (f) a sequence encoding a selection marker, and (g) a fourth recombinase recognition site of the same type as the third recombinase recognition site, said third and fourth recombinase selection sites being oriented for excision of the second portion of the sequence encoding a selection marker.

17. The construct of claim 16, wherein the recombinase recognition sites are loxP or FRT sites.

18. The construct of claim 16 or 17, wherein the second portion of the GBA gene comprises exons 9 through 11.

19. The construct of any of claims 16 to 18, further comprising at least a portion of the mouse metaxin gene located 3' from the fourth recombinase recognition site.

20. The construct of any of claims 16 to 19, wherein the selection marker is an antibiotic resistance marker.