WO1993019166A1

WO1993019166A1 - Small animal model for studying cholesterol metabolism

Info

Publication number: WO1993019166A1
Application number: PCT/US1993/002542
Authority: WO
Inventors: Nobuyo Maeda; Oliver Smithies
Original assignee: The University Of North Carolina At Chapel Hill
Priority date: 1992-03-23
Filing date: 1993-03-22
Publication date: 1993-09-30
Also published as: AU3814393A

Abstract

Mouse animal models are provided for studying lipid metabolism defects and the indications resulting therefrom. Specific genes associated with lipid metabolism are targeted for inactivation by homologous recombination, particularly apoE, apoAI, apoB, apoCIII and hepatic lipase, and homozygous mice are produced, competent to retain the defect in the germ line. The resulting mice may be used for the study of lipid metabolic disorders, as well as the effect of drugs on the disorders and the genetic defect.

Description

SMALL ANIMAL MODEL FOR STUDYING CHOLESTEROL METABOLISM

INTRODUCTION Technical Field The field of this invention is the development of a small animal model for use in studying cholesterol and lipid metabolism and testing drugs designed to alter cholesterol and lipid levels.

Background The etiology of atherosclerosis is complex. Its development is affected by many different genes, including those related to lipid metabolism, and is greatly influenced by environmental factors. In addition, the cause of the disease can only be reproduced in a^' living animal, there are no adequate in vi tro systems available. For studying this type of complex multifactorial disease, gene targeting in embryonic stem cells offers an attractive approach. Success with this approach, allows for the generation of animals with changes in specific genes, and thence the study of the effects of these changes on the physiological state of the animals. It is therefore of interest to attempt to introduce genetic changes in small animals by modifying embryonic stem cells and establishing homozygous hosts with the appropriate phenotype. In this manner, this approach may allow for the dissecting of some of the molecular details of atherogenesis and the generation of animals useful for testing new drugs for the treatment or prevention of atherosclerosis. Relevant Literature

The biological role of apolipoprotein E is described by Shore and Shore (1973) Biochem. 12, 502-507; Mahley (1986) Clin. Invest. Med. 9, 304-308; Hui et al . (1981) J. Biol. Chem. 256, 5646-5655; Driscoll and Getz (1984) J. Lipid Res. 25, 1368-1379. Apo E, besides its involvement in cholesterol and lipid metabolism, is implicated in regulation of the immune response (Hui et al . (1980) J. Biol. Chem. 255, 11775-11781) ; in nerve regeneration (Boyles et al . (1989) J. Clin. Ivest. 83, 1015-1031; LeBlanc and Podulso (1990) J. Neuro. Sci. Res. 25, 162-171) and in muscle differentiation (Majec et al . (1988) J. Cell. Biol. 107, 1207-1213; Millis et al. (1986) J. Cell. Phvsiol. 127, 366-372) . Homologous integration resulting in head-to-tail concat?- .^;.rs has been reported by Thomas and Capecchi (199) Nature .5, 847-850 and Hasty et al . (1991) Mol. Cell. Biol. 11, ? -4517. Mice are reported to use LDL as a major cholesterol carrier, as distinct from humans who use LDL (Camus et al . (1983) J. Lipid Res. 24, 1210-1218) . Inactivation of specific genes in the genome of ES cells by homologous recombination followed by generation of animals carrying the mutated gene has been described by Capecchi (1989) -Sci. 244. 1288-1292; Merlino (1991) FASEB. J. 5, 2996-3001; Roller and Smithies (1992) Annu. Rev. Immunol. 10, 705-730; McMahon and Bradley (1990) Cell 62, 1073-1085; Mucenski et al . (1991) Cell 65, 677-689; Zimmer and Gruss (1989) Nature 338, 150-153 and Chisaka and Capecchi (1991) Nature 350, 473-479 as illustrative. ApoAI and its physiological role is described by Scanu et al. (1969) Biochem. 8, 3309-3316; Fielding et al . (1972) Biochem. Biophys. Res. Comm. 46, 1493-1498; Clift et al. (1991) Am. J. Hum. Genet. 49, 8192 and Walsh et al . (1989) J. Biol. Chem. 246, 6488-6494.

SUMMARY OF THE INVENTION Mouse models are provided having genetic defects in cholesterol and lipid metabolism by substantially inactivating one or more target genes by homologous recombination in embryonic stem cells. Particularly, individual genes and coding proteins associated with transport and metabolism of cholesterol and lipid are inactivated in embryonic stem cells. Techniques providing double knock-outs may be employed which may result in homozygotes. Alternatively, a single knockout is employed where heterozygous mice may be produced, which may then be mated to provide the homozygotes. The animals may be used for studying the etiology of lipid-related diseases and to test agents as to their effect on lipid-related diseases.

SUMMARY OF THE INVENTION Mouse strains are provided which are defective as to at least one gene encoding a protein related to lipid metabolism, particularly an apolipoprotein. The genetically-defective mice may be produced by first knocking out the target gene in an embryonic stem cell, where the resulting embryonic stem cell may be hetero- or homozygous for the genetic defect. The mutated embryonic stem cell may then be introduced into a blastocyst and the embryos grown to neonate chimeras. As appropriate, the chimeras may be mated to provide for homozygous strains having a genetic defect in lipid metabolism. The resulting mice may then be used to study lipid-associated diseases, e . g. atherosclerosis, or for the testing of drugs.

DESCRIPTION OF THE SPECIFIC EMBODIMENTS

A small animal model, particularly a mouse, is provided which is genetically mutated as to a gene encoding a protein associated with lipid metabolism, particularly a gene encoding an apolipoprotein or enzyme. The mice can be produced by knocking out the target gene in an embryonic stem cell ("ES cell") . The knock-out can be as a result of homologous recombination with a construct which results in the lack of a functional product. The embryonic stem cells may then be further manipulated to provide for homozygosity or may be used directly. The modified embryonic stem cells are introduced into a blastocyst and the blastocyst carried by a pseudopregnant mouse mother. After coming to term, the rest-.--.ting pups are screened for germ line competence. As appropriate, the resulting animals may be bred for homozygosity and then continuously bred for production of the genetically-defective mice. Other small animals which may find use include golden hamsters.

Apolipoproteins associated with the lipid metabolism pathway include apolipoprotein B (apoB) ; apolipoprotein E

(apoE) ; apolipoprotein Al (apoAI) and apolipoprotein CIII (apo CIII) . Enzymes of interest include hepatic lipase and lipoprotein lipase. Of particular interest are apoAI and apoE.

Any of a wide variety of mouse strains may be employed, which may be substantially equivalent to a naturally- occurring mice or have one or more genetic defects. Genetic defects include immunocompromised mice, such as nude and scid/scid_r diabetes susceptible mice, mice with various genetic defects in hematopoiesis, atherosclerosis susceptible mice, etc. Mice with one or more of the lipid- related defects may be employed and may be bred to mice with one or more other genetic def cts. Mice which find common utilization for their availability and genetic background include C57BL/6J and 129.

In order to produce the subject mice, embryonic stem cells are employed which can be grown in culture and expanded. Embryonic stem cells from the 129 mouse strain is available and may be employed and there will be a continuous expansion of the number of embryonic stem cells which are available. The ES cell lines are grown in an appropriate nutrient medium, e.gr. fetal bovine serum enhanced DMEM, and may be co-cultured with fibroblast feeder layers, and/or grown in a medium containing leukemia inhibitory factor ("LIF") .

A construct is employed which has homology to the target gene locus and may provide for one or a combination of insertion, deletion, substitution, inversion, or the like, so as to provide for the host being at least partially non-functional as to the target gene. It is not necessary that there be no product produced, it is only necessary that if a product is produced, the product is non-functional in at least one respect.

The target construct will usually have at least about 50 bases of homology with the target locus, generally at least about 20, more usually at least about 50 bases at each flanking region, more usually at least 100 bases at least one flanking region, and may have a total of 1 kb or more, usually not more than about 20 kb of homology. Single or double cross-over recombination may be employed. Conveniently, the flanking homologous regions may flank a marker gene which allows for positive selection. Negative selection may also be employed. Thus the insertion of the marker gene will serve to disrupt and inactivate the target gene. Genetic markers which may be used for selection include neomycin resistance, which may be selected with

G418, HPRT, which may be selected with thioguanine

(negatively) or HAT (positively) , HSV-thymidine kinase, which may be selected with nucleoside analogs, e . g. acyclovir or ganciclovir, etc.

A number of different promoters have been found to be useful with the marker gene, either constitutive or inducible. Promoters of interest include the _/S-actin promoter, SV40 early and late promoters, phosphoglycerate kinase promoter, immunoglobulin gene promoter, thymidine kinase promoter, hCMV promoter, Friend spleen focus-forming promoter, etc. Alternatively, the homologous sequence can provide that the marker gene is under the transcriptional and translational regulation of the 5' promoter region of the target gene. The promoters may or may not be naturally associated with enhancers, where the enhancers may be naturally associated or from a different promoter.

Depending upon the target gene, various sites at the locus of the target gene may be selected for disruption. Thus, with the apoE gene, the region involving the second and third exons may be selected for insertion and/or deletion. With apoAI, the region associated with the second exon may be selected for insertion and/or deletion. Various techniques are employed which substantially enhance the efficiency of selection of homologous recombination. Thus, one may provide a construct where a second marker gene which allows for negative selection is placed up- or downstream from a flanking region. Thus the construct would comprise a homologous region, a positive first selection marker, a homologous region, and a second marker in the downstream direction. A double crossover event results in the loss of the second marker. Thus, by selecting for those target cells which lack the second marker, but include the first marker, one will greatly enrich for cells having homologous recombination at the target site. While thymidine kinase has found exemplification as the second marker, other selective genes may also be employed. The markers may be reversed in order, or used in more than one copy.

Further, constructs may be employed and conditions which allow for in-out targeting. By this technique, one can obtain target loci lacking the first marker gene, so as to substantially avoid having heterologous DNA in the cell.

The constructs may be produced in accordance with conventional ways, conveniently by joining the homologous sequences to the appropriate marker sequences in the proper order. The various sequences may be carried on prokaryotic vectors, so that at each step, one can analyze whether the desired product has been obtained. For homologous recombination, the entire vector may be employed or only that portion of the vector intended to be integrated, or varying degrees in between. If desired, small changes may be made in the regions of homology to provide for convenience in construction of the construct, introduction of restriction sites to simplify analysis, or for other reasons unassociated with the primary purpose of inactivation of the target gene. The region of homology may include the 5' -non-coding region associated with transcriptional regulation, the coding exons or non-coding introns, where the insertion or deletion will result in the disruption of the correct transcription and expression of the target gene.

Once the construct has been prepared, it may now be used for homologous recombination. The construct may be introduced into the ES cell by any convenient means. Techniques include electroporation, bacterial protoplast fusion with intact cells, calcium phosphate DNA coprecipitates, microinjection of DNA into the cell, lipofection, transfection, and the like. The DNA may be single- or double-stranded, linear or circular, super coiled or relaxed. For a general description, see Keown et al .

(1990) Methods in Immunol. 185, 527-537. The particular manner in which one introduces the construct is not critical to this invention, but will be selected for convenience and efficiency.

After carrying out the introduction of the construct, the ES cells are expanded in a selective medium, as described previously and then analyzed. The selective medium will provide for substantial enrichment of the ES cells carrying the construct. Conveniently, the polymerase chain reaction ("PCR") may be employed for analysis, using probes distinctive for target loci having undergone homologous recombination. The PCR-positive clones may then be further analyzed by restriction analysis and gel electrophoresis, or the like. It was noted that, interestingly, targeted colonies which grew more slowly had a higher efficiency of targeted disruption. It was further noted that, in some instances, the restriction' analysis indicated the presence of concatamers, where the incoming construct integrated in a head-to-tail fashion producing a multiplicity, about 4-5, tandem repeats.

The resulting positive ES cells may then be expanded by growth in culture and used to produce chimeric neonates. The ES cells, generally from about 1 to 30 cells/injection, are injected into the blastocoel of the blastocyst from an appropriate mouse strain. The blastocysts may be obtained from 4 to 6 week old superovulated females by flushing the uterus 3.5 days after mating. The ES cells may then be trypsinized and the modified cells microinjected into the blastocoel.

The blastocysts comprising the modified ES cells are then introduced into the uterus of a pseudopregnant maternal host. At least 1 and up to 15 of the blastocysts are returned to each uterine horn of pseudopregnant females. Females are then allowed to go to term, with birth 16-18 days after introduction. The resulting pups may then be screened for the presence of the disrupted target gene. By providing for a different phenotype of the blastocysts and ES cells, chimeric progeny can be readily detected, e . g. coat color.

Homozygosity may be achieved in two different ways. One way is to grow the embryonic stem cell containing the construct at an elevated concentration of the selective agent. The heterozygote cells are expanded in an appropriate culture medium over an extended period of time, usually at least one to six weeks or more. The cells may then be plated and, if required, individual colonies selected at varying concentrations of the drug to which the marker gene imparts resistance. The gene imparting resistance will be a non-a plifiable gene, that is, the gene does not result in tandem repeats upon exposure to in varying amounts of the drug. With neomycin resistance, for example, the concentration of G418 may vary from about 0.5-4 mg/ml, more usually from about 0.5-2.5 mg/ml. One may then screen the surviving ES cells for the existence of homozygosity, by using the polymerase chain reaction and identifying a single band for the target locus. Alternatively, one may identify germ line competent chimeric heterozygotes, first mate these chimeric heterozygotes with wild-type mates, and then mate non-chimeric heterozygotes from the expanded number of pups to achieve homozygotes. The pups may be screened for the presence of the homozygous locus.

The subject animals may be used in a wide variety of ways to establish the etiology of lipid metabolism diseases, particularly atherosclerosis. In addition, drugs which affect the various diseases may also be screened as to their effect on the phenotype resulting from the defective target locus or the treatment of indications resulting from the defective target locus. The following examples are offered by way of illustration and not by way of limitation.

EXPERIMENTAL ApoE Targeting and ApoE^" Mice

Cloning of the ao-oE gene: Mouse genomic DNA, isolated from STO cells (Ware and Axelrand (1972) Virology 50, 339-348) and fully digested with EcoRI, was used to make a library in lambda phage, Charon 32 (Loenan and Blattner (1983) Gene 26, 171-179) . The library was screened with an apoE-specific probe (a Sacl/Bglll fragment) isolated from a mouse apoE cDNA clone (Rajavashisth et al. (1985) Proc. Natl. Acad. Sci. USA 82, 8085-8089) which was kindly provided by Dr. A.J. Lusis (University of California, Los Angeles) . From one of the strongly hybridizing phage clones obtained in this screening, a 7.8 kb EcoRI fragment was isolated. Comparison of the restriction map of this fragment with the nucleotide sequence of the mouse apoE gene (Horiuchi et al . (1989) J. Biochem. 106, 98-103) showed that it contained the complete apoE gene in addition to 5.7 kb of its upstream sequence. Cell culture and electroporation: An ES cell line, E14TG2a, obtained from Dr. Martin Hooper (Hooper et al. (1987) Nature 326, 292-295) was used for all experiments. Cells were maintained in Dulbecco's modified Eagle's medium (DMEM), supplemented with 15% fetal bovine serum, 0.1 mM jβ-mercaptoethanol, 2 mM glutamine (ES medium) . Penicillin (100 IU/ml) and streptomycin (100 μg/ml) were used in some experiments. ES cells were maintained on embryonic fibroblast feeder layers prepared as previously described (Doetschman et al . (1985) J. Embrvol. Exp. Morph. 87, 22-45) .

Cells were elctroporated essentially as described by Reid et al . (1991) Mol. Cell. Biol. 11, 2769-2777, using linearized DNA at a final concentration of 2-5 nM. Electroporated cells were plated at a density of 1-2 x 10⁶ cells per 10 cm plate. Twenty-four hours after electroporation, the ES medium was replaced with medium containing 150-200 μg/ml G418 with or without 2 x 10^"s M ganciclovir.

Screening for homologous recombinants: ES cells carrying an apoE gene disrupted by homologous recombination were identified by polymerase chain reaction (PCR) and/or by genomic Southerns. Briefly, 10-14 days after electroporation and selection by G418, individual colonies were picked and divided into two parts. One portion was used for further growth and the other was combined with portions from 3-5 other colonies to form pools of 4-6 colonies per pool. DNA was isolated from pooled cells and subjected to PCR as described (Reid et al . (1991) Mol. Cell. Biol. 11, 2769-2777; Kim and Smithies (1988) Nuc. Acids Res. 16, 8887-8903), except that 0.5 μg of each primer and 1.0 U Tag polymerase were used in the reaction, and that dimethyl sulfoxide was omitted. The mixture was amplified for 55 cycles with 60 second denaturation at 92°C followed by 10 minutes of annealing and extension at 65°C. The primers used were: apoE-specific, (SEQ ID NO:l) 5' -TGTCTTCCACTATTG- GCTCG-3'; Neo-specific, (SEQ ID NO:2) 5' -TGGCGGACCGCTATCAG- GAC-3' . After PCR, samples were electrophoresed in 1.5% agarose, and transferred to nylon membranes (Hybond, Amersham, IL) . Filters were hybridized with the ³²P-labeled apoE-specific probe described above to detect a 1.2 kb diagnostic band. When employing plasmid pNMC109 as the targeting plasmid, Southern blot analysis was used to screen for homologous recombinants. Individual colonies were expanded and genomic DNA was prepared, most conveniently by the "salting out" procedure of Miller et al. ((1988) Nuc. Acids Res. 16, 1215) .

Collection of germ line chimeras: Chimeras were generated as previously described by Roller et al . ((1989) Proc. Natl. Acad. Sci. USA 86, 8927-8931) . Animals classified as chimeric by coat color were mated to strain 129 and/or C57BL/6J mice. DNA samples were isolated from tails of ES cell-derived animals by the "salting out" procedure and were analyzed for the presence of a disrupted apoE gene by genomic Southern blotting.

Double immunodiffusion test: Double immunodiffusion precipitation (Ouchterlony (1967) in Progress in Allergy, Vol. 6, eds. Kallos and Waksman, Karger, Basle pp. 30-154) on glass slides was carried out for apoE detection in 1.0% agarose in PBS containing 3% polyethylene glycol (8000) . Rabbit anti-rat-apoE antiserum was kindly supplied by Dr. John Taylor (University of California, San Francisco) . Blood samples were collected in EDTA-coated capillaries from the tails of animals between 2 weeks and 3 weeks of age, and plasma was isolated by centrifugation. Plasma (5-8 μl) in the peripheral wells was used with 8 μl of antiserum in the central well. The slides were incubated in a wet chamber for 16 h at 37°C, washed in PBS for two days and stained with Coomassie Brilliant Blue G250.

Results

Construction of targeting plasmids: Two different targeting plasmids were constructed from parts of the 7.8 kb EcoRI fragment. Plasmid pJPB69 was constructed by removing a 1.1 kb Sacl fragment containing part of intron 1, through parts of exon 3, and replacing it with the neo gene driven by a thymidine kinase (TK) promoter and a mutated polyoma enhancer (pMCneopolyA, Thomas and Capecchi (1987) Cell 51, 503-512) . In addition, a 400 bp BstEII fragment corresponding to the 3' end of exon 4 and 3' flanking region was deleted in order to facilitate the choice of a PCR primer sequence unique to the target locus. The plasmid contains 4.9 kb of uninterrupted sequence in the 5' region and 700 bp in the 3' region; prior to electroporation it was digested with Nsil and Clal. The second plasmid, pNMC109, was designed so as to be able to use positive-negative selection as described by Mansour ((1988) Nature 336, 348-352) . PNMC109 was constructed from the 7.8 kb EcoRI fragment by removing an XhoI/BamHI fragment containing part of exon 3 and part of intron 3 and replacing it by the neo gene. In addition, a copy of the herpes simplex virus TK gene was placed at the 3' end of the construct. Both of these selectable marker genes are inserted in the same transcriptional orientation as the ApoE gene. The plasmid contained 6.4 kb and 1.2 kb of uninterrupted sequences in the 5' and 3' region, respectively. Prior to electroporation, the plasmid was linearized by digestion at a NotI site that occurs in the plasmid vector 5' to the end of the neo insert.

Targeting with PJPB69: Four electroporations were carried out using plasmid pJPB69. The number of cells treated ranged from 2 to 8 x 10⁷.

1) 2 x 10⁷ 420 36(4) t 2 1

2) 4 x 10⁷ 660 32(6) 1 1

3) 8 x 10⁷ 3216 60(4) 3 2

4) 4 x 10⁷ 1808 72(1) 3 1

G418^r, G418 resistant.

All electroporations were carried out at 200 μF and 300 V in ES cell medium.

*The number of colonies isolated and confirmed by Southern blots . tThe number of colonies in each pool is in parenthesis .

For the first 3 electroporations , PCR analysis was carried out in pools of 4-6 colonies per pool . Six of 120 pools analyzed gave a positive PCR signal as judged by amplification of the 1.2 kb fragment diagnostic of targeting. From these 6 pools , four individual PCR positive clones were identified. For the fourth electroporation of the series, PCR analysis was carried out on individual colonies instead of on pools. From 72 colonies analyzed, 3 were PCR positive, although two of them were lost due to contamination prior to genomic analysis (Table 1) .

To confirm the occurrence of the targeted disruption of the apoE gene, PCR positive clones were expanded and DNA isolated from them was digested with EcoRI, Ncol or Bglll. Introduction of the neo gene in the planned manner creates a new EcoRI site in the locus, and a 2.4 kb fragment is thereby generated. An undisrupted apoE gene generates a 7.8 kb fragment. In a Southern blot analysis the parental cell line (lane P) gives only the 7.8 kb fragment, and the targeted cell line gives both the 7.8 kb fragment and the new 2.4 kb fragment, thus confirming the presence of one normal and one disrupted apoE gene. Similarly, Ncol and BgJ-.II digestions of the targeted line showed fragments of 1.6 kb, and 3.0 kb respectively, indicative of a disrupted apoE gene, in addition to the 7.0 kb and 1.8 kb bands indicative of an undisrupted apoE gene, respectively.

When the Southern blot was rehybridized to a probe specific for the neo gene, only bands predicted from the structure at the apoE locus after homologous recombination were detected, showing that no insertions of the targeting plasmid has occurred at non-homologous sites. Thus all 3 enzyme digests confirm the PCR results by demonstrating that the cell lines analyzed carry a disrupted apoE gene.

The combined targeting frequency for all electroporations was 0.77% (5/648) of the G418 resistant colonies with a range of 0.5% (1/192) to 1.3% (1/72) (Table 1) . The overall targeting frequency at the mouse apoE locus using pJPB63 was 2.6 in 10⁷ treated cells.

Targeting with PNMC109: Five electroporations were carried out as summarized in Table 2. A combined total of 1132 colonies doubly resistant to G418 and Ganciclovir were obtained, and 177 of these colonies were expanded. If homologous recombination takes place between the endogenous apoE locus and the plasmid Table II. Homologous recombination at the apoE locus with pNMC109.

Cells G418^r G418^r + GC^r G418^r + GC^r Targeted Exp. Treated Colonies* Colonies Analyzed Colonies

G418^r, G418 resistant; GC^r, Ganciclovir resistant.

Electroporation (1) thrbugh (3) were carried out at 200 μF/300 V in ES medium and (4) and (5) were carried out at 250 μF/350 V in PBS.

♦Determined on a known fraction of the cells selected with G418 alone.

pNMC109, the size of the Hindi11 fragment hybridizing to the apoE probe increases from 6.5 kb to 7.5 kb. Thus the presence of a 7.5 kb band in the Southern blot analysis was used to diagnose homologous recombination. During this screening, it was noted that a higher population of slow grow in colonies (picked after 12 days of double selection) were targeted when compared to fast growing colonies (picked after ten days of selection) . Only 1 of 15 fast growing colonies was targeted (6.6%) compared to 6 of 16 slow growing colonies (37%) . Altogether 39 targeted colonies were identified in 177 of the doubly resistant colonies (a combined frequency of 22%, Table 2) . The frequency in individual electroporations ranged from 8.3% (2/24) to 45.8% (11/24) . The higher frequency obtained in the fifth electroporation was partly because in this experiment only the slower growing colonies were analyzed. Excluding the last electroporation, the overall targeting frequency using pNMC109 was estimated to be one in 10⁶ treated cells. The existence of the desired modification of the apoE locus in the isolated cell lines was confirmed by Southern blot analysis of DNA digested with various enzymes. 37 of 39 cell lines isolated in the pNMC109 experiments showed a simple pattern of two hybridizing bands of equal intensity, one corresponding to the endogenous apoE locus, and the other to the modified locus.

The two remaining clones, T-88 and T-89, showed a third, strongly hybridizing fragment in each digest in addition to the two bands indicative of a targeted cell. Digests of the DNA from T-88 gave always patterns identical to that from T-89, and it is likely that they are of identical origin (they were picked one after the other from a single dish) . The third hybridizing band in T-89 DNA digested with Xmnl is 7.9 kb in length, approximately four to five times as intense as the other two bands, and also hybridizes the plasmid vector DNA. Similarly, extra bands of 12 kb and 4.3 kb were seen in Xbal and Kpnl digests, respectively.

Analysis of T-89 DNA with several other restriction enzymes (Apal, BamHI, Hindlll, NotI, and Sacl) using both the apoE probe and a plasmid vector probe failed to reveal any bands suggesting that the extra copies are integrated at non-homologous locations in the genome. These various results are consistent with the explanation that four to five identical copies of incoming plasmid have integrated in a head to tail, concatameric orientation at the apoE locus. The sizes of fragments obtained with these digests suggest that the incoming linear DNA lost about 1.6 kb of DNA prior to its concatamerization.

Production of chimeras and transmission of the modified gene: Six different targeted ES cell lines were tested for the production of chimeras after injection into C57BL/6J blastocysts; one (A13-D) had been targeted with the plasmid pJPB69, the other five (T-76, T-84, T-89, SZ-38, and SZ91) with the PNMC109 plasmid. About 50 embryos were used to test each line. All size generated chimeras ranging from very weak (less than 10% contribution of ES cells to coat color) for chimeras obtained from cell line T-84 to very strong (greater than 80%) for chimeras obtained with cell line T-89. In all, a total of 95 embryos were injected with T-89 cells to obtain 6 male and 4 females chimeras with 60 to 100% ES cell contribution. Two r les and one female were found to be germ line competent. srmline transmission of the E14TG2a ES cell gemone thr gh. female chimera has previously been reported by Roller et al. ((1990) Science 248, 1227-1230.)) One male chimera, JH126.1, transmitted the ES cell genome to 100% of its offspring. Southern blot of tail DNA from his pups revealed 14 of 30 pups (the expected Mendelian proportion) have inherited the disrupted copy of the apoE gene. C57xl29 FI animals heterozygous for the disrupted apoE gene are healthy at age 3 months and are breeding normally. From two sets of heterozygous FI matings, litters of 8 and 6 pups were obtained to date. Three of these 14 animals are homozygous for the modified apoE locus (the expected Mendelian proportion) , as identified by Southern blot of their tail DNA. All appear healthy at age 4 to 6 weeks.

Plasma was obtained from the offspring of a mating between heterozygotes, and the presence of apoE protein was investigated by the Ouchterlony double immunodiffusion method using a rabbit anti-rat-apoE antiserum. Plasma from the animal homozygous for the modified apoE gene gave no detectable immunoprecipitate with anti-apoE antiserum, although precipitation is clearly visible with plasma from animals heterozygous or homozygous for the normal gene. The plasma level of apoE in heterozygotes appears to be less than that in the normal animal as judged by the positions where the precipitates were formed.

ApoAI Targeting and ApoAI^~ Mice Materials and Methods Cloning of the mouse apoAI locus and construction of the targeting plasmid: A 9 kb BamHI fragment of mouse genomic DNA containing the apoAI gene and a part of ApoCIII gene was isolated from mouse ES cells (Strain 129/Ola) . The nucleotide sequence of about 5 kb of the fragment was determined to establish its validity; the sequence is approximately 89% identical to the equivalent region of the rat sequence (Hadded et al . (1986) J. Biol. Chem. 261, 13268-13277) . A targeting construct designed to disrupt the endogenous apoAI locus after the homologous recombination was made by replacing the 400 bp EcoRV-EcoRI fragment containing exon 2 of the gene with the positively selectable neomycin resistance gene, (PmclneopolyA, Stratagene) . A Herpes simplex virus thymidine kinase (TK) gene was placed either at the 5' or 3' end of the construct to make 5'TK or 3'TR constructs, respectively. The construct contains a HindiII and a Ball site at the 3' end of the Neo gene that are not present in the native locus.

Gene targeting and screening of the homologous recombinants: ES cell (E14TG2a) culture and electroporations were carried out as described (Roller et al . (1991) Proc. Natl. Acad. Sci. USA 88, 10730-10734) . Prior to electroporation, the targeting plasmids were linearized at the NotI site in the vector. Ten to twelve days after electroporation, colonies resistant to G418

(200 μg/ml, Sigma) and 2 μM ganciclovir were picked and passaged in clonal fashion. DNA was isolated for Southern analysis from 60 mm dishes of nearly confluent ES cells by standard methods, or from 20 mm dishes by a salting-out method (Miller et al . (1988) Nucl . Acids Res. 16, 1215) . Three μg DNA were digested with restriction enzymes and Southern blots were made using conventional procedures. Generation of germline competent chimeras : Approximately 10 ES cells were injected into the blastocoele cavity of C57BL/6J embryos. Surviving blastocysts were transferred to the uteri of pseudopregnant CD-I or C57BL x CBA FI females. An average of two to three transfers were made per cell line. Animals chimeric by coat color were bred to C57BL/6J animals to determine their germline competency.

Measuring total and HDL cholesterol: Total plasma cholesterol levels were measured enzymatically using one or two μl plasma and 100 μl of commercially-available reagents (Sigma) . HDL cholesterol levels were measured after removing apolipoprotein B-containing particles by precipitation with polyethylene glycol (Rubin et al . (1991) Nature 353, 256-263) . Measurements were made in duplicate and averaged for samples collected from eight-week-old homozygotes and heterozygotes derived from two FI heterozygote matings. Values for seven normal animals of approximately the same age but from other FI x FI matings are given for comparison. Before fasting overnight, the animals had been on breeding chow (ProLab 2000 Formula) which contains about 9% crude fat.

Results Targeting constructs were made using a 9 kb BamHI fragment isolated from strain 129 mouse genomic DNA that contains the complete apoAI gene plus 5' and 3' flanking region sequences (including a part of the apoC-III gene) . Exon 2 of the apoAI gene in the fragment was replaced by sequences from pMClneopolA that confer resistance to G418 in mammalian cells. These sequences are in the same transcriptional orientation as the apoAI gene. A herpes simplex virus thymidine kinase (TK) gene was placed either at the 5' end or at the 3' end of the construct to allow the use of a positive\negative selection strategy with G418 and ganciclovir. Double crossovers would be expected (in the direction of transcription of the ApoAI gene 3' of the 5' TK gene and in the intervening region of exons 4 and 3 of the ApoCIII gene. Probes which were employed were complementary to the region of the construct 5' to the apoAI gene (Probe 1) , the region 5' and 3' of exon 3 of the apoAI gene (Probe 2) and the region of exons 4 and 3 of the ApoCIII gene (Probe 3) .

The targeting DNA was introduced into ES cells (E14TG2a) by electroporation, and DNA samples from ES cell colonies resistant to G418 and ganciclovir were assayed for targeting by Southern blotting. Probe 1 hybridizes to a single 12.0 kb HindiII fragment in DNA from parental cells, but also to this fragment and to an additional 4.6 kb fragment of equal intensity in DNA from targeted cells. DNA from cells that have incorporated the targeting DNA elsewhere in the genome, by a nonhomologous event, give Hindlll fragments of unpredictable sizes, as was exemplified by one lane. Targeting was confirmed by the presence of an 8.5 kb HindiII fragment after hybridization to probe 2, probe 3, or to a probe specific for the neo gene, and by Southern blots of digests with other restriction enzymes. New Nhel and Bell fragments of the expected sizes in DNA from two of the targeted cells in addition to fragments of the same size as those from parental cells were observed; hybridization was to probe 1. Table 3 summarizes the targeting data from five experiments, the first two with the 5' TR construct and the remaining with the 3' TK construct.

As the data show, the frequency of targeting the mouse apoAI locus was very high, ranging from 82% (experiment 1) of colonies that survived double selection when using the 5' TK construct to 46% (experiment 5) with the 3' TK construct. Homologous recombination occurred under the conditions approximately once per 10⁵ treated cells. Ganciclovir selection gave a 2 to 4 fold enrichment over than seen with G418 alone, as demonstrated in experiment 3 in which 19% (8/42) of G418 resistance cells were targeted.

When injected into blastocysts, the targeted cell lines varied in the extent of chimerism produced in resulting offsprings. Two out of 9 lines used for injection were germline competent; one of them gave three male chimeras able to transmit the ES cell genome to their offspring, another gave one germline competent female chimera. Southern blot analysis of DNA isolated from the tails of ES cell-derived offspring of a male chimera mated to C57BL/6J females showed that approximately 50% of them inherited the inactivated apoAI allele.

Animals homozygous for the modified apoAI gene were generated from matings of heterozygotes. Southern blot analysis of DNA isolated from the tails of 9 pups from two litters was performed. In four of these samples (pup numbers 27-29 and 32) , probe 3 hybridized to a single 8.5 kb HindiII fragment; this establishes that these animals are homozygous for the modified apoAI gene. Pups number 30, 31,

GANC; ganciclovir

34, and 35 were heterozygotes (8.5 kb and 12 kb bands) , and number 33 had no modified genes (12 kb band only) . Animals homozygous for the disrupted apoAI appear healthy at their present age (2 to 3 months) . Mice homozygous for the disrupted apoAI gene have no apoAI in their plasma detectable by Ouchterlony double immunodiffusion tests against rabbit anti-mouse apoAI antiserum, although strong precipitation is clearly visible with plasma from animals heterozygous or homozygous for the normal gene. This establishes that at the protein level the planned modification eliminated apoAI production, as was expected from the design of the targeting construct.

At the lipid level, it is found (Table 1) that the total cholesterol in plasma from homozygous mutant at 8 weeks of age is approximately 33% that of normal animals of similar age (p<0.0005). HDL cholesterol levels in the animals show an even greater reduction (17%, p<0.0005) . Heterozygous animals have _. total cholesterol and HDL cholesterol levels 54% and 40% respectively of the levels in normal animals.

It is evident from the above results, that the subject invention provides for a useful animal model for determining the etiology of lipid metabolism-associated diseases, as well as screening compounds for their effect on the genetic defect, as well as the indications resulting from the genetic defect. In this manner, one can better understand how to treat various lipid metabolism-related diseases, such as atherosclerosis, as well as develop drugs for such treatment. The mice animal models are stable, retain the genetic defect through multiple generations and can be readily bred and raised in large numbers. Individual or combinations of genetic defects may be employed to provide models of varying genetic diseases in lipid metabolism.

All publications and patent applications cited in this specification are herein incorporated by reference as if each individual publication or patent application were specifically and individually indicated to be incorporated by reference. Although the foregoing invention has been described in some detail by way of illustration and example for purposes of clarity of understanding, it will be readily apparent to those of ordinary skill in the art in light of the teachings of this invention that certain changes and modifications may be made thereto without departing from the spirit or scope of the appended claims.

SEQUENCE LISTING

(1 ) GENERAL INFORMATION:

(i) APPLICANT: The University of North Carolina at Chapel Hill

(ii) TITLE OF INVENTION: Small animal model for studying cholesterol metabolism

(iii) NUMBER OF SEQUENCES: 2

(iv) CORRESPONDENCE ADDRESS: (A) ADDRESSEE: Bertram I. Rowland

(B) STREET: 4 Embarcadero Center, Suite 3400

(C) CITY: San Francisco

(D) STATE: California

(E) COUNTRY: USA (F) ZIP: CA 941 1 1

(v) COMPUTER READABLE FORM:

(A) MEDIUM TYPE: Floppy disk

(B) COMPUTER: IBM PC compatible

(C) OPERATING SYSTEM: PC-DOS/MS-DOS (D) SOFTWARE: Patentln Release #1.0, Version #1.25

(vi) CURRENT APPLICATION DATA:

(A) APPLICATION NUMBER: PCT/US93/

(B) FILING DATE: March 22, 1993

(C) CLASSIFICATION: 800 (vii) PRIORITY APPLICATION DATA:

(A) APPLICATION NUMBER: US 07/855,290

(B) FILING DATE: March 23, 1992

(C) CLASSIFICATION: 800

(viii) ATTORNEY/AGENT INFORMATION: (A) NAME: Rowland, Bertram I.

(B) REGISTRATION NUMBER: 20,015

(C) REFERENCE/DOCKET NUMBER: A-56241 /BIR

(ix) TELECOMMUNICATION INFORMATION: (A) TELEPHONE: (415) 781-1989 (B) TELEFAX: (415) 398-3249

(2) INFORMATION FOR SEQ ID NO: 1 : (i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 20 base pairs

(B) TYPE: nucleic acid

(C) STRANDEDNESS: single (D) TOPOLOGY: linear

(ii) MOLECULE TYPE: cDNA

(xi) SEQUENCE DESCRIPTION: SEQ ID NO: 1 :

TGTCTTCCAC TATTGGCTCG 20

(2) INFORMATION FOR SEQ ID NO:2:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 20 base pairs

(B) TYPE: nucleic acid

(C) STRANDEDNESS: single

(D) TOPOLOGY: linear

(ii) MOLECULE TYPE: cDNA

(xi) SEQUENCE DESCRIPTION: SEQ ID NO:2:

TGGCGGACCG CTATCAGGAC 20

Claims

WHAT IS CLAIMED IS:

1. A mouse embryonic stem cell heterozygous or homozygous for a genetic defect in a gene in the lipid metabolism pathway as a result of homologous recombination with a construct comprising at least one flanking region homologous for said gene and a marker gene for selection.

2. A mouse embryonic stem cell according to Claim 1, wherein said gene is the apolipoprotein E gene.

3. A mouse embryonic stem cell according to Claim 1, wherein said gene is the apolipoprotein Al gene.

4. A mouse embryonic stem cell according to Claim 1, wherein said gene is the apolipoprotein B gene.

5. A mouse embryonic stem cell according to Claim 1, wherein said gene is the apolipoprotein CIII gene.

6. A mouse embryonic stem cell according to Claim 1, wherein said gene is the hepatic lipase gene.

7. A mouse embryonic stem cell according to Claim 1, wherein said stem cell is produced by mating heterozygotes for said genetic defect to produce a homozygous embryonic stem cell.

8. A mouse embryonic stem cell according to Claim 1, wherein a neomycin resistance gene is present at the locus of said gene.

9. A mouse germ line cell heterozygous or homozygous for a genetic defect in a gene in the lipid metabolism pathway as a result of homologous recombination with a construct comprising at least one flanking region homologous for said gene and a marker gene for selection.

10. A mouse germ line cell according to Claim 9, wherein said gene is the apolipoprotein E gene.

11. A mouse germ line cell according to Claim 9, wherein said gene is the apolipoprotein Al gene.

12. A mouse germ line cell according to Claim 9, wherein said gene is the apolipoprotein B gene.

13. A mouse germ line cell according to Claim 9, wherein said gene is the apolipoprotein CIII gene.

14. A mouse germ line cell according to Claim 9, wherein said gene is the hepatic lipase genes.

15. A mouse germ line cell according to Claim 9, wherein said cell is produced by mating heterozygotes for said genetic defect to produce a homozygous germ line cell.

16. A mouse germ line cell according to Claim 9, wherein a neomycin resistance gene is present at the locus of said gene.

17. A mouse heterozygous or homozygous for a genetic def ct in a gene in the lipid metabolism pathway as a result of homologous recombination with a construct comprising at least one flanking region homologous for said gene and a marker gene for selection.

18. A mouse according to Claim 17, wherein said gene is the apolipoprotein E gene.

19. A mouse according to Claim 17, wherein said gene is the apolipoprotein Al gene.

20. A mouse according to Claim 17, wherein said gene is the apolipoprotein B gene.

21. A mouse according to Claim 17, wherein said gene is the apolipoprotein CIII gene.

22. A mouse according to Claim 17, wherein said gene is the hepatic lipase genes.

23. A method for producing a mouse homozygous for a genetic defect in a gene in the lipid metabolism pathway, said method comprising: introducing a construct into a mouse embryonic stem cell, said construct having at least 50 bases of homology to the locus of said gene, and a marker gene having a promoter region functional in said mouse embryonic stem cell and flanked by homologous sequences, whereby said construct becomes integrated into the genome of said embryonic stem cell; selecting for embryonic cells having undergone homologous recombination with inactivation of said gene; optionally submitting said embryonic cells having undergone homologous recombination, where said marker gene is a non-amplifiable gene, to a level of drug to which said marker gene imparts resistance resulting in embryonic stem cells homozygous for said homologous recombination; introducing said embryonic stem cell having undergone homologous recombination into a mouse blastocyst to provide a chimeric blastocyst; transferring said chimeric blastocyst to the uterus of a pseudopregnant female mouse and allowing said female mouse to go to term, whereby mouse pups are produced; screening said pups for germ line competence for transmitting said defect; and growing and mating the grown-up pups having germ line competence for transmitting said defect; whereby mice are obtained homozygous for said genetic defect.

24. A method according to Claim 23, wherein said gene is the apolipoprotein E gene.

C . A method according to Claim 23, wherein said gene is th apolipoprotein Al gene.

26. A method according to Claim 23, wherein said gene is the apolipoprotein B gene.

27. A method according to Claim 23, wherein said gene is the apolipoprotein CIII gene.

28. A method according to Claim 23, wherein said gene is the hepatic lipase genes.

29. A method for screening a drug for efficacy in the treatment of an indication as a result of a genetic defect in the lipid metabolism pathway, said method comprising: administering a drug to a mouse heterozygous or homozygous for a genetic defect in at least one gene in the lipid metabolism pathway as a result of homologous recombination, wherein said drug is being screened for its effectiveness in treating said indication; and analyzing the effect of said drug on said indication.

30. A method according to Claim 29, wherein said gene is at least one of apolipoprotein Al, apolipoprotein E, apolipoprotein B, apolipoprotein CIII and hepatic lipase.