WO2006017788A2 - TRANSGENIC HFN4α MICE EXPRESSING THERAPEUTIC TARGET AF1 DOMAINS AND THEIR METHODS OF USE AS MODELS FOR METABOLIC DISORDERS - Google Patents

Abstract

The invention presented herein provides mice genetically modified in the transcription factor HNF4α and methods of using the mice as in vivo models. The α1 and α7 isoforms of HNF4α are distinguished by N-terminal amino acids encoded by the first exons, which comprise the AF-1 transactivation motif. These isoforms result from two alternative promoters. The first, P1, gives rise to HNF4α1, which possesses the AF-1 motif, and the second, P2, gives rise to HNF4α7, which does not. Modifying the HNF4α locus in the mouse genome, using murine embryonic stem cells that were injected in the early mouse embryo, produced mice that express either the αl or the α7 isoform. The mice expressing exclusively HNF4α7 have altered serum triglyceride and cholesterol levels and express low levels of CAR, another transcription factor, which is implicated in xenobiotic detoxification. A reciprocal transgenic mouse expressing HNF4α1 exclusively has an altered pancreatic response to glucose injection, and a diabetes mellitus-like phenotype.

Description

TRANSGENIC HFN4α MICE EXPRESSING THERAPEUTIC TARGET AFl

DOMAINS AND THEIR METHODS OF USE AS MODELS FOR

METABOLIC DISORDERS

PRIORITY CLAIM

[001] This application claims the benefit of provisional application 60/599,065, filed in the United States Patent and Trademark Office on August 6, 2004, the disclosure of which is hereby incorporated by reference in its entirety.

TECHNICAL FIELD

[002] The invention relates to genetically modified animals and molecules used in their generation. Specifically, the invention relates to transgenic mice with genetically modified transcription factor HNF4α, and methods of making and using these mice.

BACKGROUND ART

[003] The transcription factor Hepatocyte Nuclear Factor 4α (HNF4α) gene has been described in several organisms and is detailed in the Genbank database. It is highly enriched in the adult rat liver and bound to specific DNA elements in regulatory regions (promoters) of hepatic genes (Sladek et al., 1990). Different isoforms of FfNF4α arise by alternative splicing. Isoforms expressed in the adult liver include the prototype protein αl (Sladek et al., 1990), and α2, which possesses a ten amino-acid insertion in the C-terminal F domain (Hata et al., 1992; Sladek et al., 2001).

[004] HNF4α, a member of the nuclear receptor superfamily, is highly expressed in the adult liver (Sladek et al., 1990). During mouse development, it is one of the first liver-enriched transcription factors (LETF) to be expressed. HNF4α transcripts are detected at E (embryonic day) 4.5 in primitive endoderm, in the visceral endoderm (E 5.5), and in the liver bud (E 8.5) (Duncan et al., 1994; Taraviras et al., 1994). The crucial role of this factor in the embryo was demonstrated by the inactivation of the HNF4α gene, causing peri-gastrulation lethality (Chen et al., 1994) due to failure to activate visceral endoderm functions (Duncan et al., 1997). When null-embryos were transiently rescued by tetraploid complementation with wild-type morulae-derived visceral endoderm, HNF4α was dispensable for hepatic specification but not for activation of hepatic functions (Li et al., 2000). In addition, liver-specific HNF4α deletion in the embryo (Parviz et al., 2002) led to disorganization of liver architecture (Parviz et al., 2003) and its forced expression in cultured hepatic cells resulted in the reexpression of some liver specific genes (Spath et al., 1997) and restoration of epithelial morphology (Bailly et al., 1998; Spath et al., 1998). HNF4α plays a role in regulating the expression of genes involved in nutrient metabolism and transport (Sladek et al., 2001). Its induced disruption in the adult liver provokes lethal defects in lipid and bile acid metabolism and ureagenesis (Hayhurst et al., 2001; Inoue et al., 2002). Hence, HNF4α is essential both for the induction and the maintenance of hepatic functions. The physiological importance of HNF4α is also reflected in the large number of its putative target genes (Odom et al., 2004).

[005] A mouse HNF4α isoform differing from αl in the N-terminal region of the protein was described by Nakhei et al. in 1998 and designated HNF4α7. These N- terminal amino acids are encoded at the DNA level by distinct first exons. The first exon of FfNF4αl is named IA and the first exon of HNF4α7 is named ID, as shown in Figure IB. The corresponding human sequences are shown in Figure 11.

[006] The HNF4α nuclear receptor contains six domains (A-F) (Figure 2). The N-terminal A/B domain contains the AF-I motif, which is one of the two motifs (AFl and AF2) which confer transactivation potential to HNF4α. The AF-I motif is short (24 amino acids). It is present in HNF4αl and is encoded by exon IA. The AF-I motif is absent from HNF4α7 (Hadzopoulou-Cladaras et al, 1997). In vitro experiments using truncated proteins and transfection assays showed that this motif contributed 40% of the transactivation potential of HNF4αl (Hadzopoulou-Cladaras et al, 1997), whereas no such activity could be detected in HNF4α7 (Torres-Padilla et al., 2001). The AF-I motif is known to interact with some coactivators such as CBP, Gripl, and Srcl. The second transactivation motif, AF-2, is found in the E domain, which is conserved among HNF4αl isoforms. It is also depicted in Figure 2.

[007] The HNF4αl and HNF4α7 isoforms result from alternative promoter usage. The P2 promoter generates the HNF4α7 transcript, and has been reported to be located about 40 kb upstream from Pl, which generates HNF4αl and α2 (Taraviris et al., 1994; Zhong et al., 1994; Nakhei et al., 1998; Thomas et al., 2001). The two promoters differ in their activity profiles. Both isoforms are expressed in mouse intestine. HNF4αl largely predominates in the adult mouse liver and is exclusive in mouse kidney, whereas HNF4α7 predominates in the mouse stomach and both endocrine and exocrine mouse pancreas. HNF4α7 is also slightly expressed during embryogenesis in the mouse liver but is repressed in liver after birth by a mechanism implicating HNF4αl (Briancon et al., 2004). HNF4α7 mice demonstrate enhanced expression from the P2 promoter (Briancon et al., 2004). Roughly equivalent amounts of HNF4α7 and HNF4α7 were found in human islets and exocrine pancreas, as well as in rodent insulinoma cells (Eeckhoute et al., 2003).

[008] HNF4α7 differs from αl only by amino acids encoded by the first exon, which in the case of αl, but not α7, contains a functional AFl, as described above. The differences in activation functions of these isoforms result from differences in cofactor recruitment (Torres-Padilla et al., 2002). In the liver, the α7 isoform is expressed mainly in the embryo and is nearly extinguished in the adult. In accordance with this expression profile, the transactivation capacity of the HNF4α7 isoform is efficient for fetal functions such as α- fetoprotein expression, whereas that of αl is higher than α7 for most adult hepatocyte functions (Torres-Padilla et al., 2002). The FINF4α7 isoform is also expressed in the adult stomach, intestine, and pancreas (Nakhei et al., 1998).

[009] Mutations in human HNF4α coding sequence common to Pl and P2- derived isoforms have been reported to be associated with maturity-onset non-insulin dependent diabetes mellitus of the young (MODY 1) (Ryffel 2001; Yamagata et al., 1996). Since the P2 promoter is predominantly active in the pancreas (Nakhei et al, 1998; Boj et al., 2001; Eeckhoute et al., 2003), mutant HNF4α7 is probably implicated in the insulin-secretion defects of diabetes, whereas the associated hepatic dysfunctions (Lehto et al., 1999; Shih et al., 2000) can likely be ascribed to mutant αl (Sladek et al., 2001).

[010] A deficiency of all FfNF4α proteins in the mouse has been generated by deletion of exons common to all isoforms (Chen et al., 1994). This deficiency is lethal in the early stages of embryogenesis. More recently, an adult liver-specific knock-out has been described (Hayhurst et al., 2001). All isoforms were deleted and the phenotype was lethal in the adult within a few weeks, due to profound defects in bile acids, and in urea and lipid metabolism. These defects could be linked to HNF4αl, as it is the predominant form in the adult liver. No isoform-specific knock¬ out or isoform specific knock-in animal has been described to date.

[011] Mice expressing either HNF4αl or HNF4α7, but not both isoforms, would be useful in determining the in vivo roles of these isoforms. They would provide a tool for discriminating between genes with expression strictly or mainly dependent on the presence of a functional AF-I motif in the HNF4α protein and genes with HNF4α-dependent gene expression which are expressed independently of AF-I. Such mice would also be useful in defining an ensemble of HNF4α target genes which require AF-I for expression in "loss of function" tissues and those misregulated by a "gain of function."

BRIEF DESCRIPTION OF THE TABLE AND DRAWINGS

[012] Table 1 shows the results of a serum chemistry analysis of wild type, αl-only, and α 7-only mice. (^]) indicates that the values are expressed (±) standard deviation and were obtained from a minimum of six mice (9-13 weeks old). Equal numbers of males and females were used. Statistical analyses were performed with GraphPad Instat^® software using a one-way analysis of variance (ANOVA) test followed, when applicable, by the multiple comparison Dunnett's post-test (* = p< 0.05 and ** = p<0.01. ALAT is alanine amino transferase. BFfBA is beta-hydroxy butyrate. ND is not determined.

[013] (²) indicates that the assays were performed on 9-16 week-old mice in groups comprising 4-19 mice. Groups of males and females were tested for each parameter. When sex-specific differences were obtained in a comparison, the results of both sexes are shown, or the sex is specified.

[014] (³) indicates that the results were obtained by testing female mice. Male α7-only mice were not affected.

[015] (⁴) indicates that the assays were performed on 2-10 male mice 14-16 weeks old.

[016] (⁵) indicates that the prothrombin times (PT) and activated partial thromboplastin times (APTT) were obtained on groups of at least eight mice 11-12 weeks old (not separated by sex).

[017] Figure IA shows a schematic drawing of the FfNF4α gene, including the two alternative promoters Pl and P2 (Figure IB), and the exon and intron compositions of the αl and α7 isoforms.

[018] Figure 1 also shows schematic diagrams of genomic loci of the HNF4α gene and plasmid constructs derived from this gene. The loci and constructs are shown both before and after recombination in ES cells. Nucleotide lengths are described in kilobases. Boxes depict exon coding sequences. Black bars depict the probes used for screening ES cells. Restriction endonucleases are abbreviated by .4ZwNI (A), BarnHI (B), EcoKL (E), HmDIII (H), Ncol (N), EcoRV (R), Sphl (S), and Xhol (X). [019] Figure 1C shows the replacement of the HNF4α7 first exon coding sequence (exon ID) by the HNF4αl coding sequence (exon IA). The resulting mice that are homozygous for this allele are referred to as "αl-only."

[020] Figures ID shows the replacement of the HNF4αl first exon coding sequence (exon IA) by the HNF4α7 coding sequence (exon ID). The resulting mice that are homozygous for this allele are referred to as "α7-only."

[021] Figures IE and IF show southern blot screening of recombinant ES cells at the exon ID locus (E) or at the exon IA locus (F). Sizes of targeted (t) and wild-type (+) restriction fragments are given in each case.

[022] Figures IG and IH show PCR genotyping of chimeric males (chim) obtained by blastocyst microinjection of recombinant ES cells, and of their descendants after crossing with wild-type females (wt). Figure IG shows exon ID replacement, while Figure IH shows exon IA replacement. As schematized above both gel pictures, primers (one-branch arrows) were chosen on either side of the first axon coding sequences (gray and black boxes). Using these primers, both PCR assays give a fragment 39 bp longer in the presence of exon IA compared to a fragment containing exon ID. The size of the fragment obtained depends on the presence of a recombinant allele (t) or not (+).

[023] Figure II shows the relative distribution of HNF4αl and HNF4α7 mRNA transcripts determined by quantitative PCR assays performed on cDNA from tissues of 11-16 week old mice. "Intestine" comprises the duodenum and part of the small intestine. Transcript amount was expressed relative to α-actin (± standard deviation; n is greater than or equal to three).

[024] Figure IJ shows the transcript levels of a few genes deregulated in the α7-only livers are not affected in the αl-only livers. Northern blot analysis performed with total RNA from αl-only mouse livers compared to wild-type and α7-only livers. The expression levels of apoAIV, apoCII, apoCIII and SR-Bl in the αl-only livers (right) are, as expected, not altered. GAPDH is used as a loading control.

[025] Figure 2 shows the domain structure of HNF4αl and HNF4α7 isoforms and the location of AF-I and AF-2. The isoforms have different N-terminal domains. Figure 2 depicts differences in the Activation Function 1 (AF-I) motif between the isoforms. It lists functions associated with the structural domains.

[026] Figure 3 shows a schematic diagram of the Hnf4al/a7 reciprocal knock-in replacement and the results of PCR expression of these constructs. [027] Figures 3 A and B show schematic diagrams of the plasmid constructs and the genomic loci before and after homologous recombination in ES cells. Figure 3 A depicts replacement of the HNF4α7 exon ID coding sequence with the HNF4αl coding sequence. Figure 3B shows the reciprocal. "Neo" is the G418-resistance cassette; DT-A is diphtheria toxin.

[028] Figure 3C shows semi-quantitative radioactive PCR performed with total liver RNA using specific forward primers (left) corresponding to the coding sequences of HNF4αl or HNF4α7, or the 5' untranslated region (UTR) at the Pl promoter. The reverse primer was common to both isoforms. The HNF4α7 first exon is 39 base pairs shorter than that of HNF4αl, giving rise to the smaller band visible in the HNF4α7 and heterozygous (α7/+) mouse livers while using the P 1-5 'UTR primer.

[029] Figure 3D shows semi-quantitative RT-PCR showing normal ratios of the C-terminal splicing-derived isoforms in the mutant mouse livers (α2 and α8). The PCR primers frame the 30 nucleotide insertion shown in Figure 1 (Torres-Padilla et al., 2001).

[030] Figure 3E shows a Western blot of mouse liver nuclear extracts of the wild type, αl-only and α7-only genotypes. The extracts were probed with an antibody that recognizes the C-terminus of all HNF4α isoforms. The HNF4α7 protein migrates faster than the HNF4αl isoform (arrowheads). The (*) indicates an HNF4αl degradation product or a non-specific signal. Brain nuclear extracts were used as a negative control for the presence of HNF4αl. TFIIB was used to control the gel loading process.

[031] Figure 3F shows immunohistochemical staining of wild-type and α7- only mouse livers with the C-terminal antibody used in Figure 3E (upper panels), an αl-specific antibody designated Nl-14 (middle panels), or an α7-specific antibody (lower panels). The arrowheads indicate bile duct cells. The scale bar indicates a distance of 250 urn,

[032] Figure 4A shows a comparison of the cholesterol profiles of wild type, αl-only, and α7-only mice. It shows total serum cholesterol (TC), high density lipoprotein (HDL), low density lipoprotein (LDL), and very low density lipoprotein (VLDL). The stars indicate that the cc7-only mice have significantly less TC and HDL (** indicates pO.Ol; * indicates ρ<0.05).

[033] Figure 4B shows a fast pressure liquid chromatography (FPLC) profile comparing the cholesterol profiles of α7-only and wild-type mice. The amount of cholesterol in HDL, LDL, and VLDL fractions was determined on a minimum of three pools of serum from 2-4 males each. The mice were fasted overnight before the serum was collected. The results shown were obtained from one representative pool of wild-type and one representative pool of α7-only mice. The FPLC cholesterol profile of αl-only mice were not significantly different from wild type mice.

[034] Figure 4C shows the accumulation of lipid observed in the livers of cc7- only mice compared to the livers of wild type and αl-only mice. Five month old male mice were either fed ad libitum or fasted for 24 hours prior to sacrifice. Livers were dissected and cryosections obtained and stained with Oil red O using methods known in the art. ( ) indicates a slight lipid accumulation in the wild type livers. The scale bar indicates a distance of 100 um).

[035] Figure 5 A shows the macroscopic appearance of freshly dissected livers. Livers from the α7-only mice are paler than their wild type and αl-only counterparts. This difference becomes more pronounced after fasting.

[036] Figure 5B shows the microscopic appearance of liver cryosections following oil red O staining of wild type, αl-only, and α7-only mice. The oil red O stain demonstrates that the α7-only mouse livers are fat-laden.

[037] Figure 6 shows expression profiles of genes implicated in lipid transport and metabolism in α7-only mouse livers (A-F) and intestine (G) compared to wild-type mice. Figures 6A-D shows the results of Northern blotting performed with liver RNA from 9-12 week old α7-only and wild type mice. In some cases, the results of quantifying the transcript levels are shown in bar graphs to the right of the corresponding Northern blot. Black bars represent wild type and white bars represent α7-only mice. In quantifying the transcript levels, expression was normalized to the expression level of glyceraldehyde 3-phosphate dehydrogenase (GAPDH) and represented as a percentage of the amount of expression by the wild type (n=4-6).

[038] Figures 6A-C show the expression of apolipoprotein genes, which are implicated in blood triglyceride and cholesterol transport (apolipoproteins ApoAI, ApoAII, ApoAIV, ApoA ApoAI, Apo, ApoE and ApoBEC (apolipoprotein B mRNA editing catalytic subunit), VLDL secretion (low density lipoprotein receptor (LDL-R), scavenger receptor class B type 1 receptor (SR-Bl)), and fatty acid metabolism (microsomal triglyceride transfer protein (MTP)), hepatic lipase (HL), lipoprotein lipase (LPL), peroxisome proliferator-activated receptor alpha (PP ARa), 3-hydroxy- 3-methylglutaryl-coenzyme A synthase (HMG-Synt), sterol receptor element binding protein Ic (SREBPIc), S-hydroxy-S-methylglutaryl-coenzyme A reductase (HMG- Red), liver fatty acid binding protein (L-FABP), MCAD, CPT H, AOX, and FAS).

[039] Figure 6D shows genes involved in bile acid biosynthesis, excretion and re-uptake from the blood. Cyp7Al, cholesterol 7α hydroxylase; MDR2, multidrug resistance protein 2; NTCP, sodium taurocholate cotransporter protein; OATPl, and organic anion transporter protein 1.

[040] Furthermore, Figure 6E shows quantitative real time PCR analysis of hepatic apolipoprotein B transcript amounts normalized to β-actin. Results are show as percentage of wild-type values (n>6).

[041] Figure 6F shows semi-quantitative RT-PCR. Shown is hepatic expression of the HNF4α target gene HNF lα and of transcription factors (LXRα/RXRα) known to play an essential role in cholesterol metabolism. No obvious changes in these transcript amounts could be observed in α7-only versus wt livers. LXRα, oxysterol receptor α; RXRα, retinoic acid receptor α.

[042] Figure 6G shows Northern blot analysis performed with total RNA from adult mouse intestines. Quantification as for (A-D) (gray bars, αl-only mice ; n

> 3).

[043] Figure 6H shows that adult α7-only mice express lower amounts of the apolipoproteins A4, C2, and C3 than wild type, when examined by Northern blot analysis of total RNA extracted from their livers.

[044] In Figure 6, the following abbreviations are used: MCAD, medium- chain acyl-coA dehydrogenase; CPT II, carnitoylpalmitoyl transferase II; AOX, acyl- coA oxidase; FAS, fatty acid synthase. (D) Genes involved in bile acid synthesis, excretion and re-uptake from the blood. Cyp7Al, cholesterol 7α hydroxylase; MDR2, multidrug resistance protein 2; NTCP, sodium taurocholate cotransporter protein; OATPl, organic anion transporter protein 1.

[045] Figure 7 shows that αl-only mice present a delay in the response to glucose intraperitoneal injection and α7-only mice show insulin hypersensitivity.

[046] Figure 7 A shows glucose tolerance test performed with at least 15 mice per genotype (8-14 weeks old). No sex-related differences were observed. Blood was collected from the tail prior to (t=0) and after glucose injection, and glycemia was measured. Figure 7 A shows that αl-only mice exhibit a delayed response to an intraperitoneal glucose injection, (* indicates p<0.05; ** indicates pθ.01). [047] Figure 7B shows insulin sensitivity testing. Regular pork insulin was injected intraperitoneally into at least 5 males and 5 females of each genotype and glycemia was measured. Figure 7B shows that male α7-only mice exhibit a hypersensitivity to insulin (** indicates p<0.01).

[048] Figure 7C shows transcript levels of the ATP-dependent potassium channel subunit Kir6.2 in the whole pancreas (left) and in isolated islets (tight) as determined by quantitative real time PCR. Results are normalized to β-actin and shown relative to the wild-type values (n > 3 for whole pancreas; n = 2 for isolated islets).

[049] Figure 8 shows the expression profile of genes implicated in amino acid (A) and glucose metabolism (B, C), of serum protein carriers (D) and hepcidin (E). Except hepcidin (Hepc), these genes are direct targets for the HNF4α or HNF lα transcription factors ((Odom et al., 2004) and references therein). OTC, ornithine transcarbamylase; TAT, tyrosine aminotransferase; GK, glucokinase; Gys2, glycogen synthase; PEPCK, phosphoenolpyruvate carboxykinase; Alb, albumin; TTR, transthyretin; TFN, transferrin. (E) Levels of hepcidin mRNA are increased in the α7- only livers, but not significantly.

[050] Figure 9 shows that α7-only mouse livers express a decreased amount of Constitutive Androstane Receptor (CAR) compared to wild type, when analyzed by semi-quantitative RT-PCR.

[051] Figure 10 shows that expression is strongly diminished in the liver of the α7-only mice. (A) Semi-quantitative RT-PCR revealing a decrease in CAR transcript amounts (3 isoforms) in α7-only livers compared to wild-type, whereas expression of PXR is not altered. (B) Quantitative real time PCR. CAR and PXR transcript amounts normalized to β-actin are represented relative to the wild-type values (n > 3). (C) Northern blot performed with liver RNA of mice that were injected either with TCPOBOP (TC) or vehicle (O). Absence of induction of cyp2blθ was observed in 3 out of 4 α7-only mice; the fourth mouse did express cyp2blθ despite very low levels of CAR transcripts (not shown), CAR expression was quantified by real time PCR, and normalized values are given below the gel for each sample (relative to the non-induced wild-type mice). (D) EMSA. Cos7 cells were transfected with expression vectors for HNF4αl or HNF4αl or HNF4α7 and whole cell extracts prepared. Amounts of extracts were adjusted for HNF4αl and α7 protein amounts, as deduced from titrations in western blots (not shown). Oligonucleotides are named according to their 5' ends relative to the mouse CAR start codon. ApoCIII is a well-known HNF4α binding oligonucleotide, used as a control. Arrows, HNF4αl/ α7 DNA complex. Arrowhead, HNF4αl and α7 supershifts obtained with the antibody recognizing part of the HNF4α C-terminus domain (Ab). *, non specific bands.

[052] Figure 11 shows a schematic diagram of human exon IA and exon ID coding sequences and their corresponding amino acids, including HNF4αl coding sequence (SEQ ID NO: 1); HNF4αl protein sequence (SEQ ID NO: 2); HNF4α7 coding sequence (SEQ ID NO: 3); HNF4α7 protein sequence (SEQ ID NO: 4).

[053] Figure 12 shows a schematic diagram of mouse exon IA and exon ID coding sequences and their corresponding amino acids, including HNF4αl coding sequence (SEQ ID NO: 5); HNF4αl protein sequence (SEQ ID NO: 6); HNF4α7 coding sequence (SEQ ID NO: 7); HNF4α7 protein sequence (SEQ ID NO: 8). MODES FOR CARRYING OUT THE INVENTION

[054] The invention provides a mouse comprising a genetically modified HNF4α gene. This mouse can have an altered lipid profile. It can also have an altered pancreatic response to glucose. In an embodiment, a mouse of the invention can express HNF4α in the liver and/or kidney. In an embodiment, a mouse of the invention can express HNF4α in the stomach and/or pancreas.

[055] The invention also provides a plasmid comprising the nucleic acid sequence of HNF4α wherein exon ID replaces exon IA under the regulatory control of the Pl promoter. The invention further provides a plasmid comprising the nucleic acid sequence of HNF4α wherein exon IA replaces exon ID under the regulatory control of the P2 promoter. The invention provides an embryonic stem cell comprising either of these plasmids. The invention also provides a mouse comprising modified alleles from this embryonic stem cell. It provides a primary or immortalized cell culture derived from such a mouse, hi an embodiment, a mouse comprising modified alleles from the embryonic stem cell expresses HNF4αl and does not express HNF4α7. Such a mouse can have an altered pancreatic response to glucose. In an embodiment, a mouse comprising modified alleles from the embryonic stem cell expresses HNF4α7 and does not express HNF4αl. Such a mouse can have an altered lipid profile. For example, the transcription of apolipoproteins A4, C2, and/or C3 can be altered. [056] In another aspect, the invention provides a method of using a mouse comprising a genetically modified HNF4α gene or an immortalized cell culture derived from a mouse comprising modified alleles from an embryonic stem cell comprising either of the plasmids described above to test the toxic effects of a drug comprising providing a cell culture derived from a mouse comprising the modified alleles, or (1) a mouse comprising the modified alleles, wherein the mouse expresses HNF4 αl and does not express HNF4α7 and (2) a mouse comprising the modified alleles, wherein the mouse expresses HNF4α7 and does not express HNF4αl; contacting the cell culture or the mice of (1) and (2) with the drug; and evaluating the survival of the mouse expressing HNF4α7, but not HNF4αl, compared to the mouse expressing HNF4αl, but not HNF4α7, or evaluating the survival of the cultured cells comprising a plasmid comprising the nucleic acid sequence of HNF4α, wherein exon ID replaces exon IA under the regulatory control of the Pl promoter with the survival of the cultured cells comprising a plasmid comprising the nucleic acid sequence of HNF4α, wherein exon IA replaces exon ID under the regulatory control of the P2 promoter.

[057] The invention also provides a method of determining whether a substance is metabolized by a CAR-dependent activity, comprising providing a mouse as described above, which expresses HNF4α7 and does not express HNF4αl, placing the mouse in contact with the substance, and evaluating whether the mouse resists a dose harmful to a control animal.

[058] The invention further provides a method of identifying in vivo analogues of HNF4α AFl comprising administering a candidate analogue to a mouse comprising modified alleles from an embryonic stem cell, as described above, wherein the mouse expresses HNF4αl and does not express FINF4α7; assaying the expression of AFl -specific target genes in biological specimens; and comparing the results of the assay to that obtained from a biological specimen of an untreated mouse comprising modified alleles from an embryonic stem cell, as described above, wherein the mouse expresses HNF4α7 and does not express HNF4αl . In practicing this method, the specific target gene can be chosen from ApoA4, ApoC2, and CAR.

[059] The invention yet further provides a method of making a mouse genetically modified at the HNF4α locus comprising providing a plasmid comprising at least a fragment of the HNF4α gene, introducing the plasmid into an embryonic stem cell, introducing the stem cell into a mouse embryo, and allowing the embryo to develop into a mouse. This method can be practiced with a mouse that comprises exon IA on both chromosome alleles in place of exon ID. It can also be practiced with a mouse that comprises exon ID on both chromosome alleles in place of exon IA.

[060] In another aspect, the invention provides an antagonist of the AF activity of one or more HNF4α protein, wherein the antagonist is capable of interfering with the activity of AF function in vivo and/or in vitro.

[061] In yet another aspect, the invention provides a method of identifying in vitro analogues of HNF4α AFs comprising treating a primary or immortalized cell culture derived from a mouse comprising modified alleles from the embryonic stem cell comprising a plasmid described above with an antagonist of the AF activity of one or more HNF4α protein, wherein the antagonist is capable of interfering with the activity of AF function in vivo and/or in vitro, or an analogue of the antagonist; and comparing the expression of AF target genes in cell cultures derived from such a mouse, wherein the mouse expresses HNF4αl and does not express HNF4α7, to cell cultures derived from a mouse expressing HNF4α7 and not HNF4αl.

INDUSTRIAL APPLICABILITY

[062] The mice, plasmids, antagonists, and methods provided by the invention are useful for providing information relevant to the understanding of diabetes mellitus, xenobiotic detoxification, iron metabolism, the function of HNF4α activating factor, and the clearance of cholesterol and triglycerides from the blood. DETAILED DESCRIPTION OF THE INVENTION

[063] The invention provides information regarding the necessity and/or redundancy of the HNF4αl and HNF4α7 isoforms. These isoforms differ only by the AF-I motif. Specifically, the invention relies on information regarding the role of the AF-I motif in the adult liver where αl is predominant. It also relies on information regarding the role of the AF-I motif in the pancreas, where α7 is predominant. HNF4αl and HNF4α7 Expression

[064] Two plasmid constructs were designed to enable homologous recombination at the exon IA or at the exon ID locus, permitting the simultaneous deletion of the endogenous exon and its replacement by its counterpart. (Figures 1C and D.) In one construct, exon IA replaced exon ID at the P2 promoter and was still present at the Pl promoter; in the other construct, exon ID replaced exon IA at the Pl promoter and was still retained at the P2 promoter. Thus, the modified locus expressed only αl or only α7, but the expression was driven from both promoters (see below) so that the total amount of HNF4α expressed was approximately the same as in the wild type mouse. A mouse of the invention, designated the " α7 only" mouse, expressed α7 in the liver and the kidney in place of αl in the normal mouse. Another mouse of the invention, designated the "αl only" mouse, expressed αl in the pancreas and the stomach in place of α7.

[065] Because up to nine isoforms of HNF4α have been described or proposed (Sladek et al, 2001), HNF4α7 is used herein to include not only α7 but also the constitutively produced splicing-derived isoform, α8 and α9. Likewise, HNF4αl is used herein to encompass at least the α2 to α4 isoforms.

[066] The tissue distributions of mouse HNF4αl and mouse HNF4α7 were compared by quantitative real time PCR. The results are shown in Figure IE. Liver and kidney expressed HNF4αl mRNA nearly exclusively. The intestine expressed both isoforms. The ratio of HNF4αl mRNA to HNF4α7 mRNA was approximately 2.4: 1. The spleen provided a negative control for HNF4α expression.

[067] The profile of gene expression in the livers of the HNF4αl-only and the HNF4α7-only mice reflected the isoform replacement. As shown in Figure 3 C, HNF4α7 was strongly expressed from the Pl promoter and HNF4αl transcripts were not detectable. No HNF4α7 transcripts were detected in the HNF4αl-only mouse liver.

[068] The ratio between the C-terminal splicing-derived isoforms, α2 and α8, compared to αl and α7, was maintained in the mutant mice, as shown in Figure 3D. This confirms that the exon replacement did not affect downstream spicing events.

[069] As shown in Figure 3E, Western blot analysis of the proteins in liver extracts showed that HNF4α proteins of α7-only mice migrated faster than those in wild-type and αl-only samples, consistent with the expected smaller size of the protein. The total amount of expressed protein was approximately the same in the different genotypes.

[070] As shown in Figure 3F, immunohistochemistry staining of α7-only and wild-type mouse liver cryosections demonstrated that HNF4α expression was restricted to hepatocytes and excluded from bile duct cells.

[071] Replacing the αl isoform by α7 in a tissue expressing predominantly or exclusively αl in the wild-type diminished the physiological function for which the αl -specific AF-I motif is required. Reciprocally, expressing HNF4α7 target genes in tissues expressing predominantly HNF4α7 can be affected by ectopically adding the AF-I motif. Lipid Metabolism

[072] Plasma lipoproteins influence the development of atherosclerosis and their concentrations are associated with the risk of coronary heart disease. The invention provides mice expressing only one isoform of the essential transcription factor HNF4α, the α7 isoform, which was forced to be expressed in the adult liver by genetic modification. In the normal mouse, the predominant HNF4α isoform present in adult liver is αl . The αl and α7 isoforms differ in their N-terminal amino acids comprising the AF-I transactivation domain. Thus, phenotypic differences observed between mice expressing only α7 or only αl are likely due to the expression of this short protein motif, which contributes to the global activity of the HNF4α factor.

[073] The α7-only mice showed an obvious decrease in triglycerides and cholesterol/lipoprotein blood content compared to wild-type, and, with a compensatory slight lipid accumulation in hepatocytes, the liver cells responsible for fatty acid metabolism. Hence, the invention provides that these α7-only and αl-only mice are useful for decreasing global blood lipoprotein content. Inhibiting the AF-I domain of the HNF4α factors present in normal liver is predicted to diminish blood lipoprotein levels, based on the information provided herein. The invention also provides the αl-only and the α7-only mice as models for pharmacological testing.

[074] An interesting phenotype was revealed by serum chemistry assays, as shown in Figure 4 and Table 1. The αl-only mice were normal for all the parameters tested. However, the α7-only mice exhibited a significant decrease in triglyceride and cholesterol serum contents in all lipoprotein fractions. This was associated with an increase in ketone body levels, which are the degradation products of fatty acids. Serum cholesterol and triglycerides were lower in α7-only mice compared to wild type. Serum cholesterol decreased 41% compared to wild-type and triglycerides decreased 53%. Non-esterified fatty acids were decreased by 24% and ketone bodies (β-hydroxybutyrate) were increased by 82%.

[075] Correspondingly, the liver, which metabolizes fatty acids, accumulated lipid. It has been reported that the lethal liver-specific HNF4α total knock-out provokes a strong lipid accumulation in the livers of the non-surviving mice. The results shown in Figure 4 A demonstrate a role of the HNF4α AF-I motif in lipid metabolism, since its absence led to lowered serum lipids.

[076] Figure 5 A shows that livers from five month old α7-only mice show hepatic steatosis, manifest by a white-gray aspect and are paler than wild type controls, in a manner characteristic of a fat-laden tissue. This difference is exacerbated following a 24h fast, which stresses the animals and produces a greater lipid accumulation in the liver. During prolonged fasting, adipose tissue releases fats into the blood circulation, which in turn is absorbed mostly by the liver.

[077] These macroscopic observations were confirmed by the microscopic observation of lipid accumulation on liver sections using the classical method of oil red O staining. Males fed ad libitum or fasted for 24 hours were sacrificed and their livers were frozen in embedding medium. Cryosections were fixed in 4% paraformaldehyde and stained with oil red O, following classical protocols. Oil red O stains neutral lipids red.

[078] Figure 5B shows stained cryosections of fed mice. A homogeneously dispersed lipid accumulation was observed in the α7-only livers. Lipid droplets were homogeneously distributed within the parenchyma. The hepatocytes of the α7-only mice were easily distinguishable from wild-type, which have rare, localized fatty hepatocytes. The size of the lipid droplets was much larger in the α7-only than the wild-type mice. Additionally, heterogeneity was observed among the αl-only mice.

[079] Steatosis was observed, but was subtle in nine-week old mice and was chronically amplified with age. A 24 hour fast, which stimulated the release of fatty acids from adipose tissue and accumulation and catabolism in hepatocytes, exacerbated the accumulation of lipid in the α7-only mice compared to the wild-type.

[080] The expression of numerous genes known to be implicated in lipid metabolism help elucidate the molecular mechanisms associated with the decrease in serum lipid content and the lipid accumulation in the livers of the α7-only mice. As described in further detail below, a VLDL secretion deficiency and increased lipid uptake from the blood into the liver, but not a higher rate of lipid syntheses or a defect in β-oxidation or ketogenesis contribute to this α7-only mouse phenotype. The liver transcript levels of apolipoproteins A4, and to a lesser extent C2 and C3, were diminished in α7-only mice compared to wild-type (Figure 6H). The apolipropoteins constitute the protein fraction of the circulating lipoproteins (HDL, LDL, VLDL) carrying triglycerides, phospholipids, and cholesterol in the blood. Figure 6H shows that adult α7-only mice express lower amounts of the apolipoproteins A4, C2, and C3 than wild type, when examined by Northern blot analysis of total RNA extracted from their livers.

[081] As shown in Figure 6 A, few of the genes known to be associated with liver transport and metabolism were strongly deregulated in the livers of the α7-only mice. The HDL component apoAIV and the VLDL components apoCII and apoCIII were specifically down-regulated in the α7-only mice. As shown in the left panel of Figure 6A, transcripts of apoAIV were nearly undetectable. Expression of apoCII was reduced by 64% and expression of apoCIII was reduced by 19%. Thus, the αl- specific AF-I motif is required for the full expression of these genes. ApoAII expression was altered in the lethal liver-specific HNF4α knockout (Chen et al., 1994; Hayhurst et al., 2001) but was unaltered in the α7-only mice.

[082] Disrupting the mouse apoAIV gene has been reported to decrease VLDL and HDL levels, and to decrease triglyceride levels as a result of altered apoCIII gene expression (Maeda et al., 1994; Weinstock et al., 1997). Combined diminished expression of the apoAIV and apoCIII genes could contribute to the low serum cholesterol and triglyceride levels of the α7-only mice.

[083] ApoAIV and apoCIII are expressed both in the liver and in the intestine. Their down-regulation was specific to the liver. No decline in apoAIV or apoCIII was observed in the intestine. Thus, expression of the apoATV gene is dependent on HNF4α in both the liver and intestine, but AF-I is involved in its expression in the liver and not in the intestine. The regulatory region of the gene recruits distinct cofactors in hepatocytes and intestinal cells and requires a functional AF-I only in the liver.

[084] MTP and apoB, two gene products involved in VLDL secretion, were also down-regulated in the α7-only mice. As shown in Figures 6A and 6E, MTB was down regulated by 25% (left panel) and apoB was also down-regulated by 25% (bar graph in right panel). MTP and apoB are both involved in lipid assembly. Their disruption has been reported to be associated with reduced serum cholesterol and triglyceride and increased hepatic accumulation. Furthermore, a VLDL secretion defect can be expected to result in the accumulation of lipid by hepatocytes. Thus, the MTP and apoB down-regulation observed in the α7-only mice may contribute to their low serum triglyceride and cholesterol levels and hepatic lipid accumulation. [085] The top portion of the middle panel of Figure 6B shows the expression of certain enzymes involved in the control of hepatic fatty acid and cholesterol metabolism. LPL, the rate-limiting enzyme for the hydrolysis of VLDL and chylomicron triglyceride, is induced in the liver of the α7-only mice. Hepatic LPL has been reported to mediate the uptake of HDL cholesterol (Rinninger et al., 1997), an observation consistent with the low HDL cholesterol levels observed in the α7- only mice. Also consistent with the low HDL levels of the α7-only mice is the observation herein that the HDL receptor gene SR-Bl was induced in the α7-only mice (Figure 6A, left panel). LPL gene expression has been reported to be mediated by cytokines (Merkel et al., 1998). Thus, a mechanism for LPL up-regulation is mediation by inflammatory signals present in the α7-only mouse liver.

[086] As mentioned above, neither a higher rate of lipid synthesis nor a defect in β-oxidation or ketogenesis can account for the α7-only mouse phenotype. These catabolic pathways are, however, chronically activated by the accumulated lipid in the livers of the α7-only mice. Fatty acid synthase (FAS) transcription, an indicator of fatty acid synthesis, is decreased in the α7-only mouse liver. This decrease becomes more pronounced with age (Figure 6C, middle panel). This inverse correlation between lipid and FAS transcript levels indicates a negative feedback loop.

[087] As shown in the bottom of the middle panel of Figure 6C, the transcript levels of MCAD and AOX, enzymes involved in the rate limiting steps of mitochondrial and peroxisomal β-oxidation pathways, respectively, were either increased or unchanged. Transcription of the mitochondrial enzyme HMG coA- synthase (HMG-Synt, shown in the upper portion of the middle panel), which is the rate-limiting enzyme in ketogenesis, was increased by 77%, consistent with the high levels of ketone bodies observed in α7-only mouse serum.

[088] MCAD, AOX, and HMG-Synt are targets for the lipid metabolic regulator PP ARa. Expression of PPARa was not affected in α7-only mice fed ad libitum, but PP ARa expression was down-regulated following fasting. This down- regulation was detectable in the livers of nine- week old α7-only mice and was pronounced in five-month old α7-only mice with steatosis (Figure 6C).

[089] Bile acid metabolism was not affected in the α7-only mice. Transcription of the genes encoding Cyp7Al, MDR2, OATPl, and GAPDH was not affected in the α7-only mice compared to wild-type, as shown in the top portion of the right panel of Figure 6D. GAPDH transcript levels have been reported to remain stable in the absence of HNF4α (Wiwi et al., 2004). Transcription of NTCP was slightly elevated.

[090] As shown in Figure 6G, the αl -only mice exhibited a normal expression pattern of genes known to be associated with liver transport and metabolism, including apoArV, apoCII, apoCIII, SRBl, and GAPDH. Kidney Function

[091] Wild type kidney expresses the HNF4αl but not the HNF4α7 isoform. It is expressed in the renal proximal tubules and is absent from the glomerulus (Chabardes-Garonne et al., 2003). However, glucosuria, an indicator of a glucose reabsorption defect was not observed in any of the eight α7-only mice examined. Serum creatinine and urea levels were also unchanged in α7-only mice. Potassium concentration was reduced by 16% in α7-only females but not males, indication a possible gender-dependent defect in renal reabsorption. Alternatively, this hypokalemia is due to intestinal absorption defects. Carbohydrate and Amino Acid Metabolism

Unaltered Gene Expression

[092] HNF4α has been reported to regulate carbohydrate and amino acid metabolism. As shown in Figure 8, these enzymes were not affected in the α7-only mouse liver. For example, ornithine decarboxylase (OTC), a reported target gene of HNF4α is unaltered. OTC deregulation has been reported to be responsible for ureagenesis defects reported in Hnf4a-mύ\ livers by Inoue et al., 2002. PEPCK expression was normal in the α7-only mice and was induced by fasting to the same extent as in wild-type mice. The transcript levels of certain carriers of nutrients and hormones in the blood was also unaltered in the α7-only mouse liver (Figure 8).

Diabetes Mellitus

[093] In humans, diabetes mellitus includes type I diabetes, which is a genetic auto-immune disorder with Langerhans islet destruction, and type II diabetes, which results from defects in pancreatic insulin secretion. Maturity Onset Diabetes of the Young (MODYl), a form of type II diabetes, co-segregates with loss-of-function mutations in the HNF4α gene (Sladek et al., 2001). The αl-only mice demonstrated a diabetic-like phenotype following glucose injection. This phenotype was not observed in α7-only mice. Thus, not only can a loss of HNF4α function in the pancreas lead to diabetes, but, as provided herein, a gain of function leads to diabetes. Hence, the αl-only and α7-only mice determine molecular mechanisms underlying specific types of diabetes.

[094] Human polymorphisms that cause alterations in HNF4α gene activity have been associated with MODYl (Yamagata et al., 1996). Mice with HNF4α deletions in their pancreatic beta cells are slightly hyperinsulinemic and have an impaired response to a glucose tolerance test associated with an insulin secretion defect that has been reported to correlate with low transcript levels of the ATP- dependent potassium channel subunit Kir6.2 (Gupta et al., 2005).

[095] A glucose tolerance test, for example, performed by an intraperitoneal glucose injection, can detect an abnormally slow recovery to a normal glycemic level. Reasons for such a delay include an effect on pancreatic insulin secretion, and an effect on peripheral tissues, such as liver, adipose tissue, or muscles which are slightly insulin-resistant and do not normally absorb a glucose surplus. Following an intraperitoneal glucose injection, the αl-only mice showed a delay in the recovery to a normal glycemic level, indicating a significant resistance to a glucose injection (Figure 7). The HNF4αl mice had normal insulin and glucose levels (Table 1). The HNF4α7 mice displayed slight resistance to glucose at 15 min. post-injection but quickly.

[096] Insulin sensitivity tests can distinguish between a defect in insulin secretion and insulin resistance in peripheral tissues as the reason for the delay in the recovery of the αl-only mice to a normal glycemic level. The test involves an insulin injection. If the glycemia does not decrease in response, peripheral tissues are insulin-resistant. An insulin sensitivity test determined that αl-only mice were insulin resistant, since no differences were observed as compared to wild-type mice (Figure 7A). An islet defect in the absence of peripheral resistance would be expected to cause significant variations in glycemia in response to an insulin injection, which were not observed in the αl-only mice. Surprisingly, the α7-only male mice, but not their female counterparts, were markedly hypersensitive to insulin, due to hepatic defects in glucose transport and metabolism (Figure 7B).

[097] The αl-only mice presented only slight pancreatic defects, characteristic of a subtle form of type II diabetes mellitus. In contrast, the α7-only mice displayed an obvious insulin hypersensitivity of the peripheral tissues (Figure 7). This phenotype is associated with liver defects in glucose metabolism. The αl-only mice displayed no decrease in Kir6.2 expression in the pancreas or isolated pancreatic islets (Figure 7C). Xenobiotic Detoxification

[098] Blood detoxification is a function of the liver which is mediated largely by products of the cytochrome P450 gene. Constitutive Androstane Receptor (CAR) is a factor that switches on cytochrome P450 gene expression, and is thus, in part responsible for xenobiotic metabolism. CAR activity was shown to be potentiated by phenobarbital. Loss of CAR increases sensitivity to zoxazolamine- induced paralysis while decreasing sensitivity to cocaine-induced acute hepatic response in the mouse, indicating a high degree of complexity in xenobiotic clearance and suggesting that studies of pharmacological responses in the HNF4α mouse model will yield valuable information (Wei et al., 2000). Furthermore, CAR null mice are resistant to acetaminophen (paracetamol) hepatotoxicity (Zhang et al., 2002).

[099] CAR is one of several factors implicated in the activation of the large cytochrome P450 gene family essential in xenobiotic metabolism. CAR expression was nearly abolished in α7-only mice. Thus, these mice provide a pharmacological model for xenobiotic metabolism, in addition to the models for lipid metabolism and diabetes mellitus described above. They provide a rapid test to determine whether pharmaceuticals are metabolized by a CAR-dependent activity. In this test, α7-only mice are selectively resistant to a dose harmful or lethal to wild-type or αl-only mice.

[0100] CAR was weakly expressed in α7-only mice compared to the expression levels observed in the wild-type mice (Figures 9 and 10A). Thus, one can predict that these mice would show either resistance or hypersensitivity depending on which xenobiotics are injected or ingested. Hence, the α7-only mice have pharmacological interest, since suppressing activity may confer advantages or hypersensitivity in tolerance to pharmaceuticals, via regulation of CAR expression. Furthermore, the CAR or apoA4 transcripts can be used to monitor AF-I activity.

[0101] Several nuclear receptors, including CAR and the pregnane-X receptor (PXR) induce expression of cytochrome P450 genes. As shown in the left panel of Figure 1OB, the expression of PXR was not altered in the α7-only mice compared to wild-type. CAR transcripts were reduced approximately 10-fold in the α7-only mice. This expression pattern is similar to that observed in the lethal HNF4α knock-outs (Chen et al., 1994; Hayhurst et al., 2001; Tirona et al., 2003). The CAR-specific ligand TCPOBOP of the CAR-inducible gene cyp2blθ did not induce cyp2blθ expression in the α7-only mice (Figure 10C).

[0102] The invention provides that the mouse CAR promoter comprises eight potential binding sites, identified by the Matlnspector program, in the 10 kb region upstream of the initiating ATG codon. In vitro studies demonstrated that four of these binding sites bound HTNF4α oligonucleotides -1341, -3624, -7598, and -6979. Oligos -1341 and -3624 bound more strongly and oligos -7598 and -6979 bound less strongly. (Figure 1OD.) All four sites bound equivalent levels of HNFαl and HNFα7 homodimers, consistent with in vitro studies and deletion construct studies of Sladek et al., 1999. Thus, the drastic reduction in CAR expression shown in Figure 9 and 1OA is mediated by differences other than the affinities of the different HTNF4α isoforms for the CAR promoter. This is consistent with a mechanism wherein HNF4 αl directly regulates CAR expression via one or more of the herein identified binding sites and the diminished capacity of HNF4 αl to regulate CAR gene expression is caused by differences in cofactor recruitment capacity of the HNF4α7 isoform (Torres-Padilla et al., 2002).

[0103] In addition to the defects in lipid metabolism, serum bilirubin levels were higher in α7-only mice compared to wild-type in females but not in males. This could reflect a hepatic defect and is consistent with low levels of CAR, which regulates bilirubin clearance (Huang et al., 2003). Iron Metabolism

[0104] The invention provides that blood iron levels were decreased by 24% in α7-only mice. This is consistent with elevated hepatic expression of hepcidin, a peptide hormone involved in body iron homeostasis and secreted by the liver upon iron overload and inflammation (Nicolas et al., 2002). This suggests that the α7-only mice may have a hepatic iron sensing defect. The hepatic function indicators albumin, ALAT, and coagulation tests were unchanged.

[0105] Thus, these mice can also be used as models in iron pathologies, such as hereditary hemochromatosis, a defect linked to excessive intestinal iron absorption, leading to iron deposition in several tissues. Since the normal equilibrium between HNF4 αl and α7 isoforms is disrupted in the transgenic mice, the AF-I domain could be implicated in such defects. AF-I Functional Assay

[0106] The HNF4αl and HNF4α7 mice of the invention be used to identify HNF4α target genes that require AF-I for expression. Array analysis of tissues from wild type and mutant mice that express FlNF4α using techniques known in the art can identify the full spectrum of HNF4α target genes. This information permits analysis, including clustering, of the targets and the identification of common regulatory elements. It will also facilitate the identification of cofactors mediating AF-I dependency.

[0107] The αl and α7-only genetically modified mice can reveal pathologies under conditions that produce stress to an organ that does not express the appropriate isoform. For example, a high fat diet could present such a stress to the liver of the α7- only mice. Other potential target organs include the pancreas of αl -only mice and the kidney of α7-only mice. Any substance, e.g., a small molecule that interferes with AF-I function could be monitored by quantitative analysis of transcripts such as CAR or apoA4 that are reduced or abolished in α7-only mice.

[0108] AF analogues include any chemical or biological molecules which show the same activity of an AF, or an antagonist or agonist activity. As an example, analogues can be comprised of a chemical family derived from AFl peptide sequences. Addition of one or more amino acids or replacement of at least one amino acid of the sequence MDMAD YS AALDP AYTTLEFENVQV or the sequence MDMADYSAALDPAYTTLEFENVQVLTMGN. (Figures 11 and 12.) Analogues can be characterized by their capacity to compete for or inhibit the natural activity of one or more HNF4α AFs.

[0109] It must be noted that, as used herein and in the appended claims, the singular forms "a," "or," and "the" include plural referents unless the context clearly dictates otherwise. Thus, for example, reference to "a plasmid construct" includes a plurality of such constructs and reference to "the agent" includes reference to one or more agents and equivalents thereof known to those skilled in the art, and so forth.

[0110] Further, all numbers expressing quantities of ingredients, reaction conditions, % purity, polypeptide and polynucleotide lengths, and so forth, used in the specification and claims, are modified by the term "about," unless otherwise indicated. Accordingly, the numerical parameters set forth in the specification and claims are approximations that may vary depending upon the desired properties of the present invention. At the very least, and not as an attempt to limit the application of the doctrine of equivalents to the scope of the claims, each numerical parameter should at least be construed in light of the number of reported significant digits, applying ordinary rounding techniques. Nonetheless, the numerical values set forth in the specific examples are reported as precisely as possible. Any numerical value, however, inherently contains certain errors from the standard deviation of its experimental measurement.

[0111] With respect to ranges of values, the invention encompasses each intervening value between the upper and lower limits of the range to at least a tenth of the lower limit's unit, unless the context clearly indicates otherwise. Further, the invention encompasses any other stated intervening values. Moreover, the invention also encompasses ranges excluding either or both of the upper and lower limits of the range, unless specifically excluded from the stated range. Examples

[0112] The examples, which are intended to be purely exemplary of the invention and should therefore not be considered to limit the invention in any way, also describe and detail aspects and embodiments of the invention discussed above. The examples are not intended to represent that the experiments below are all or the only experiments performed. Efforts have been made to ensure accuracy with respect to numbers used (e.g., amounts, temperature, etc.) but some experimental errors and deviations should be accounted for. Unless indicated otherwise, parts are parts by weight, molecular weight is weight average molecular weight, temperature is in degrees Centigrade, and pressure is at or near atmospheric.

Example 1 : Plasmid Constructs for Homologous Recombination at the

HNF4α Locus

[0113] A PCR fragment containing the αl-specific exon IA was obtained by amplification from a mouse genomic 129Sv BAC clone and inserted at the Kpnl-Smal sites present in the commercial pBluescript plasmid. Another PCR fragment containing approximately 200 bp downstream of the α7-specific exon ID was obtained from another BAC clone called m21 (Incyte, Wilmington, DE). This fragment was inserted 3' to the previous fragment so as to reconstitute accurately the end of exon IA and the beginning of intron ID. An HinDUI-Kpnl restriction fragment (~ 4.1 kb) obtained from the m21 BAC clone was inserted in front of the cloned exon IA so as to reconstitute the endogenous α7 5 'untranslated region and promoter sequences. The entire mosaic fragment containing 4.1 kb of α7 promoting sequences, exonlA, and approximately 200 bp of intron ID, was then cloned into pPGK-neo-PGK-DTA plasmid upstream of the pgk-neo (neomycin) cassette framed by loxP sites (floxed). Another restriction fragment AlwNl-AlwNl (~ 3 kb) obtained from intron ID sequence was cloned 3' to the neo cassette and 5' to pgk-DTA (diphtheria toxin A). The DTA cassette was used for counter-selection of false positive clones which are neomycin resistant and derive from a non-homologous recombination event in random genomic sites.

[0114] A PCR fragment containing exon ID obtained from the m21 BAC clone was subcloned in pBluescript plasmid, followed by approximately 230 bp subcloned from the αl -specific intron IA sequence. Approximately 1.6 kb of αl promoting element, containing the 5 'untranslated region, was grafted in front of exon ID. The whole mosaic construct was, as for the previous construct, integrated 5' to the pgk-neo cassette of the pPGK-neo-PGK-DTA plasmid and an approximately 4.2 kb intron IA fragment was inserted 3' to the pgk-neo cassette and 5' to the pgk-DTA cassette. Both constructs were linearized and purified prior to transfection into murine embryonic stem (ES) cells.

[0115] To replace the coding sequence of the α7-specific exon ID by that of the αl-specific exon IA, a plasmid construct was prepared with exon IA coding sequence closed 3¹ of the promoter sequences and the 5'UTR of exon ID, and 5' of a neomycin-resistance cassette, intron ID sequences, and DT-A cassette. The reciprocal construct was created for exon IA replacement.

[0116] Two mouse 129/Sv genomic DNA bacterial artificial chromosome (BAC) clones (Briancon et al., 2004), were used as templates for constructs and probe syntheses (Figure 1C and D). To replace the coding sequence of the oc7-specific first exon (exon ID) by that of the αl-specific first exon (exon IA), a PCR fragment containing the 88 bp exon IA coding sequence, preceded by 15 bp of the α7 5' untranslated region (5' UTR) was obtained by using an α7/ αl mosaic forward primer. The α7 sequences abut the ATG start codon and carry a Kpnl site present in the endogenous α7 5' UTR. The fragment was cloned into the Bluescript^® II KS plasmid (Stratagene) within the Kpnl-Smal sites. A Ncol-Notl 152 bp PCR fragment, containing the exon IA 3' end grafted to the 5' end of the α7-specific first intron, was then ligated between the Ncol site, present in the previously subcloned exon IA, and the Notl site present in the plasmid. Thus, the exon IA splice donor site and the 5' end of the α7 first intron were accurately re-created. PCR assays were performed with the Expand™ High Fidelity PCR system (Boehringer Mannheim, Germany). A 4.1 kb HinDIII-Kpnl restriction fragment containing the sequences located upstream of the endogenous α7 ATG was then subcloned in front of the exon IA using the 5'UTR Kpnl site (see above). The whole mosaic construct was ligated into the pPGKNEOLox2PPGKDTA plasmid, upstream of the neomycin-resistance cassette flanked by loxP sites ('floxed'). A blunted 2.9 kb AIwNI- AIwNI restriction fragment present in the endogenous α7 first intron was then inserted 3' of the neo cassette and 5' of the DT-A cassette (diphtheria toxin- A). This DT-A cassette was used to eliminate potential non-homologous recombinant ES clones.

Example 2: ES Cell Screening and Mouse Breeding

[0117] The plasmid constructs of Example 1 were introduced into murine ES stem cells by electroporation. Recombinant ES cell colonies were selected by the addition of 300 ug/ml G418 to the culture medium. The Neo cassette incorporated into the plasmid constructs enabled G418 resistance. G418 is toxic to unmodified ES cells.

[0118] Six positive ES clones for the construct αl-> α7 and 2 clones for α7-> αl were obtained and stored under liquid nitrogen. Cells from two clones of each construct were injected into C57B1/6N mouse blastocyst embryos 4.5 days post- coϊtum by microinjection. These chimeric embryos were transplanted into a pseudopregnant B6CBA female uterus and developed normally. When adult, founder chimeric males of each construct were crossed with wild-type C57B1/6N females to test for allelic transmission. Heterozygous descendants were then crossed with pgk- cre mice (Lallemand et al., 1998) in order to delete the floxed Neo cassette, which interferes with HNF4α gene expression. A subsequent cross with wild type C57B16 mice eliminated the ere transgene. Neo-deleted heterozygous mice were crossed to generate homozygous mice.

[0119] Insertion of the floxed phosphoribosyltransferase gene (neo) in the HNF4α7-specific intron ID was not lethal, suggesting that a knock-out of transcripts derived from P2 would be viable. The same neo cassette was embryonic lethal in intron IA, mimicking the HNF4α null allele.

[0120] The mice possessing exon IA on both chromosome alleles in place of exon ID were called "αl-only" mice and the reciprocal genotype was called "α7- only." Serum chemistry studies were performed at the Mouse Clinical Institute (Illkirch, France). Mice 9-16 weeks old were fasted overnight before retro-orbital blood collection. Statistical analyses were performed using a one-way analysis of variance test (ANOVA) followed when applicable by the multiple comparison Dunnett's post-test (* indicates p<0.05 and ** indicates p<0.01).

[0121] The neo gene was inserted into the first introns within each targeting construct for the selection of recombinant ES clones. G418 resistant ES clones were analyzed by Southern blotting using 5' and 3' probes external to the targeting construct (Figures 1C and D, probes a, b, e,f). hi addition, a Ncol site present in the exon IA coding sequence and absent from the exon ID enabled confirmation of the exon replacement (Figures 1C and D, probes d, e). The absence of any other non¬ homologous integration of the targeting constructs in the ES cell genome was verified with a probe specific for the neo cassette (Figures 1C and D, probe c). Two clones of each knock-in construct were microinjected into blastocysts and chimeras were mated with wild-type females for transmission of the targeted allele (Figures IG and H).

[0122] hi a second step, the foxed neo cassette was deleted by mating with pgk-cre mice. The importance of this step is implied by the fact that the presence of the selection cassette on both chromosomes was embryonic lethal in the case of the exon IA replacement. Indeed, from neo-positive heterozygous mouse intercrosses, heterozygotes and wild-type were identified in non-mendelian proportions close to 2/3 and 113 respectively (53 and 29 mice respectively out of 82 descendants), which was coherent with a prenatal lethality of the homozygous embryos. In the case of the total HNF4α gene disruption, null embryos recovered at El 0.5 consisted primarily of extra¬ embryonic tissues and the embryonic parts had degenerated (Chen et al., 1994). hi this work, at ElO.5, we identified by genotyping neo-positive homozygous embryos for the exon IA replacement that were much smaller than their littermates and for which the rostral part of the neural tube was not closed, thus presenting a delay in their development (not shown). This indicates that ectopic integration of the neo cassette in the first intron of the HNF4α gene mimicks the total gene knock-out, as was expected since the neo cassette carries a polyadenylation site. Interestingly, no such lethality was observed for the exon ID replacement since neo-positive αl-only mice were identified in mendelian proportions (35 αl-only and 42 wt out of 185 descendants) this suggests that the knock-down or knock-out of P2-derived transcripts is not lethal.

[0123] The mice comprise two discrete "knock-in" mouse lines. The first expresses only the αl isoform and its splice-derived variants, under the control of both the Pl and the P2 promoter. The second expresses only the α7 isoform and its splice-derived variants, under the control of both the Pl and the P2 promoter. The αl- only and the α7-only mice were viable and fertile. They presented no phenotype obvious in the absence of testing. The reciprocal partial gene replacement does not affect their longevity. Thus, contrary to the lethality of the HNF4α total knock-out phenotypes, both isoforms are sufficiently redundant to enable the animals to survive. The αl-only mice have impaired glucose tolerance and the α7-only mice are dyslipidemic. These "knock-in" mice provide the first direct tests for HNF4α AF-I function in vivo. They have been used to identify AF-I dependent target genes.

[0124] The mice were genotyped using PCR-based methods. Genomic DNA was extracted from mouse tails and subjected to PCR analysis using primers framing exon ID (forward, 5'-TCACTGCCTTCCTGGTGGACTGGCTCCCGG-S' (SEQ ID NO: 9); reverse, 5'-CCAGCCGTCTCCCAGCCCCAGATATTGGCC-S' (SEQ ID NO: 10)) or exon IA (forward, 5'-GGAGAATGCGACTCTCTAAAACCCT-S' (SEQ ID NO: 11); reverse, 5'-TCTGGCCACAGTA CGACGAAGGC-3' (SEQ ID NO: 12)). In PCR assays, the amplified band differed by 39 bp because exon IA is 39 bp longer than exon ID. Primers specific for the neo cassette confirmed deletion of the selection marker after crossing with pgk-cre mice. Cre-recombinase specific primers have been described previously (Hayhurst et al., 2001). Alpha only and α7-only mice were born in agreement with a mendelian inheritance pattern from heterozygous crosses: 49 αl-only and 31 wild-type in 165 births (chi² = 0.13), and 33 α7-only and 47 wild-type in 170 births (chi² = 0.24).

[0125] Total RNA was extracted from mouse tissues with TRIzol reagent (Invitrogen™ Life Technologies), or isolated through a C₅Cl cushion (pancreas; (Chirgwin et al., 1979)) or with the Ambion RNAqueous^®-Micro kit (islets). Pancreatic islets were hand-picked after standard digestion with collagenase P (Roche). For Northern blots, RNA (20 μg per lane) was separated by electrophoresis in denaturing agarose gels, transferred to Nylon membranes and hybridized with ³²P- labeled probes obtained by random priming (Amersham). Probe templates were synthesized by PCR from liver reverse transcriptase products or extracted from plasmids (Hayhurst et al., 2001). Signals were analyzed with a Storm 860 apparatus (Molecular Dynamics) and the ImageQuant software. Conditions for RT-PCR and sequences of the primers specific for the αl and α7 exon 1 coding sequences have been described (Briancon et al., 2004). Quantitative real-time PCR assays were performed with SYBR Green Master Mix (Applied Biosystems) and analyzed following either the absolute standard curve method to take into account the amplification efficiency of primers (for HNF4αl/ α7, see details in (Briancon et al., 2004), or the comparative Cγ method when amplification efficiency of the primer pair used is equivalent to that of the β-actin primers.

[0126] Nuclear protein extracts were prepared for Western blotting from adult mouse tissues following Dounce homogenization in 15 mM HEPES, 15 mM KCl, 2.0 M sucrose, 1.0 mM EDTA, 0.5 mM DTT, 0.5 mM spermine, 0.5 mM spermidine, 0.5 mM pefaboc (Roche), and a cocktail of protease inhibitors (Complete, Roche). Proteins were measured with the Bradford assay (Bio-Rad), separated (20 μg/lane) on a 4-12% polyacrylamide Bis-Tris NuP AGE™ gel (Invitrogen) and transferred to a nitrocellulose membrane. The membrane was probed with HNF4α C-terminal peptide antibody (sc-6556; Santa Cruz Biotechnology), and reprobed with TFIIB antibody (sc-225; Santa Cruz Biotechnology). Bound antibody was revealed by peroxidase- conjugated secondary antibody (Caltag, DakoCytomation) detected with the ECL Plus reagent (Amersham Biosciences) using Hyperfilm™ autoradiographic films (Amersham Biosciences). Western Blot results are shown in Figure 3E.

[0127] For immunohistochemistry, 10 μm liver cryosections were fixed in 4% paraformaldehyde prior to antibody addition. The HNF4αl (Nl-14) and α7-specific primary antisera (Sladek et al., 1999) and revealed as for TFIB. Immune complexes were detected with 3,3'-diaminobenzidine (DakoCytomation). Sections were counterstained with Mayer's Hematoxylin. Immunohistochemistry results are shown in Figure 3F.

Example 3: Lipid Profiles

[0128] Oil Red O Staining. Liver cryosections from mice fed ad libitum or fasted for 24h were fixed in 4% paraformaldehyde and stained for 10 minutes in a solution of oil red O (Sigma) at 3 g/L in 60% isopropanol.

Example 4: Glucose Tolerance and Insulin Sensitivity Tests

[0129] Mice (8-15 weeks old) were fasted overnight prior to intraperitoneal injection with a solution of 20% glucose in sterile saline at a dose of 2 g/kg body weight or with regular pork insulin at a dose of 0.2-0.5 IU/kg. Blood was collected from the tail prior to and after injection and glycemia was measured (Figure 7). Most of the experiments were carried out at the Mouse Clinical Institute (Illkirch, France). [0130] Blood was also collected retro-orbitally for biochemical and endocrinological assays and by intracardiac punction for coagulation tests. Cholesterol in lipoprotein fractions was determined by fast-protein liquid chromatography (F. P. L. C). Glucose tolerance tests were performed by injecting a solution of 20% glucose in 0.9% NaCl intraperitoneally at a dose of 2g/kg body weight, and blood was collected from tails prior to and 15, 30, 45, 60, 90, 120, 150 and 180 minutes after injection. Glucose concentrations were measured by an Accu- Chek Active blood glucose sensor, using Accu-Chek active sticks (Roche Diagnostics). For the insulin sensitivity test, mice were injected intraperitoneally with regular pork insulin (0.2-0.5 IU/kg body weight) and blood glucose concentrations were measured as above until 90 minutes after injection. The presence of glucose in urine was investigated with Clinistix ® strips (Bayer Diagnostics).

Example 5: Electrophoretic mobility shift assays (E. M. S. A.)

[0131] Cos7 cells were grown in Dulbecco's modified Eagle's medium containing 10% fetal calf serum under humidified 7% CO₂, and transfected by phosphate calcium co-precipitation with pCB6 HNF4αltag (Spath and Weiss, 1997), CMV-HNF4α7-VSV (Torres-Padilla et al., 2001) or the empty vector pCB6. Whole cell extracts were prepared (Jacquemin et al., 1999) and EMSA were adapted from reference (Cereghini et al., 1988). The P³²-labeled double-stranded oligonucleotides (37 bp long) were designed in the 10 kb region upstream of the mouse CAR start codon (GenBank contig NT_078306.1) and carry the HNF4α binding sites determined in silico using the Matlnspector program (Quandt et al., 1995). ApoCIII is a well- known HNF4α binding oligonucleotide (Mietus-Snyder et al., 1992), used as a control. The C-terminal HNF4α antibody was used to confirm that the complexes observed were due to HNF4α proteins. Dried gels (6% polyacrylamide) were exposed in a Phosphorlmager cassette.

Example 6: Ligand-Induced Activity of Constitutive Androstane Receptor

[0132] Five-month old mice were injected intraperitoneally with the CAR agonist l,4-bis[2-(3,5- dichloroρyridyloxy)]benzene (TCPOBOP) at a dose of 3 mg/kg body weight, or vehicle (5% DMSO in sunflower oil), and sacrificed 6 hours later (Wei et al., 2000). References

[0133] The specification is most thoroughly understood in light of the cited references, all of which are hereby incorporated by reference in their entireties. The disclosures of the patents, applications, and other references cited above are also herein incorporated by reference in their entireties. The publications discussed herein are provided solely for their disclosure prior to the filing date of the present application. Nothing herein is to be construed as an admission that the present invention is not entitled to antedate such publication by virtue of prior invention. Further, the dates of publication provided may be different from the actual publication dates which may need to be independently confirmed.

[0134] Aoyama, T, et al. (1998) J. Biol. Chem. 273, 56-78-84.

[0135] Bailly, A., et al. (199S) J CeIl Sd Hl(Pt 16), 2411-2421.

[0136] Boj, S. F., et al. (2001) Proc Natl Acad Sci USA 98, 14,481-14,486.

[0137] Briancon N; et al. (2004) J Biol Chem 279, 33,398-33,408.

[0138] Chabardes-Garonne, D., et al. (2003) Proc Natl Acad Sd 100, 13,710-13,7

[0139] Chen, W. S., et al. (1994) Genes Dev 8(20), 2466-2477.

[0140] Chirgwin, J.M., et al. (1979) Biochemistry 18, 5294-9.

[0141] Duncan, S. A., et al. (1994) Proc Natl Acad Sd 91, 7598-7602.

[0142] Duncan, S. A., et al. (1997) Development 124(2), 279-287.

[0143] Eeckhoute, J., et al. (2003) Endocrinology 144(5), 1686-1694.

[0144] Elshourbagy, N.A. et al. (1985) Proc. Nat 'I Acad. Sci. USA 82, 8242- 46.

[0145] Farese, R.V. et al. (1995) Proc. Nat'lAcad. Sd. USA 92, 1774-78.

[0146] Geier, A. et al. (2003) Hepatology 38, 345-54.

[0147] Goldberg I. J. (1998) J. CHn. Invest. 102, 893-901.

[0148] Green, VJ. et al. (1998) J. Biol. Chem. 273 29950-57.

[0149] Gulick, T., et al. (1994) Proc. Nat'lAcad. Sci. USA 91, 11012-16.

[0150] Gupta, R. K., et al. (2005) J Clin Inves 115, 1006-1015.

[0151] Hadzopoulou-Cladaras, M., et al. (1997). J Biol Chem 272, 539-550.

[0152] Hashimoto, T., et al. (2000) J. Biol. Chem. 275, 28918-28.

[0153] Hashimoto, T. et al. (1999) J. Biol. Chem. 274, 19228-36.

[0154] Hata, S., et al. (1992) Biochim Biophys Acta 1131(2), 211-213.

[0155] Hayhurst, et al. (2001) MoI Cell Biol 21(4), 1393-1403.

[0156] Huang, L.S. et al. (1995) J.Clin. Invest. 96, 2152-61. [0157] Huang, W. et al. (2003) Proc. Nat'lAcad. ScL 100, 4156-61.

[0158] Inoue, Y., et al. (2002) J Biol Chem 277(28), 25,257-25,265.

[0159] Ktistaki E., et al. (1994) Nucl. Acid. Res. 22, 4689-96.

[0160] Lallemand Y, et al. (2002) Transgenic Res 7, 105-112.

[0161] Leung, G.K. et al. (2000) J Biol. Chem. 275, 7515-20.

[0162] Li, J., et al. (2000) Genes Develop 14(4), 464-474.

[0163] Lehto, M., et al. (1999) Diabetes 48(2), 423-425.

[0164] Maeda, N., et al. (1994) J Biol Chem 269, 23,610-23,616.

[0165] Malerod, L. et al., (2003) Biochem. BioPhys. Res. Commun. 305, 557- 65.

[0166] Malik, S. et al. (2002) MoI. Cell. Biol. 22, 5626-37.

[0167] Mascrez, B. et al. (2001) Development 128, 2049-62.

[0168] McEwen, IJ. (2004) Endocr. Relat. Cancer 11, 281-93.

[0169] Merkel, M., et al. (1998) J Clin Invest 102, 893-901.

[0170] Nakhei, H., et al. (1998) Nucleic Acids Res 26(2), 497-504.

[0171] Nicholas, G., et al. (2002) Proc. Natl. Acad. Sd. 99, 4596-4601.

[0172] Odom, D. T. et al. (2004) Science 303, 1378-1381.

[0173] Parviz, F., et al. (2002) Genesis 32(2), 130-133.

[0174] Parviz, F., et al. (2003) Nat Genet 34(3), 292-296.

[0175] Rhee, J. et al. (2003) Proc. Nat'lAcad. Sd. USA 100, 4012-17.

[0176] Rimiinger, F., et al. (1997) J Lipid Res 39, 1335-1348.

[0177] Rodriguez, J.C. et al. (1994) J. Biol.Chem. 269, 18767-72.

[0178] Rodriguez, J.C. et al. (1998) Biochem. Biophys. Res. Commun. 242, 692-6.

[0179] Ryffel, G. U., et al. (2001) J MoI Endocrinol 27(1), 11-29.

[0180] Sauvaget D. et al. (2002) J. Biol. Chem. Ill, 34540-48.

[0181] Shih, D. Q., et al. (2000) Diabetes 49(5), 832-837.

[0182] Sladek F. M., et al. (1990) Genes Dev 4, 2353-2365.

[0183] Sladek F. M. et al. (1999) MoI. Cell. Biol. 19, 6509-22.

[0184] Sladek, F. M., et al. (2001) in Nuclear Receptors and Genetic Disease, pp. 309-361, Academic Press

[0185] Spath, G. F., et al. (1997) MoI Cell Biol 17(4), 1913-1922.

[0186] Spath, G. F., et al. (1998) J Cell Biol 140(4), 935-946.

[0187] Taraviras, S., et al. (1994) Mech Dev 48(2), 67-79. [0188] Thomas, H., et al. (2001) Hum MoI Genet 10(19), 2089-2097. [0189] Tirana, R.G. et al. (2003) Nat. Med. 9, 220-24. [0190] Torres-Padilla, M. E., et al. (2001) Mech Dev 109(2), 183-193. [0191] Toires-Padilla, M. E., et al. (2002) J Biol Chem 277(47), 44,677-

[0192] Tirana, R. G., et al. (2003) Nat Med 9, 220-224.

[0193] Wang, H; et al., (2000) J Biol Chem 275, 35,953-35,959.

[0194] Wang, J. C, et al. (1998) J. Biol. Chem. 273, 30847-50.

[0195] Wammark, A. et al. (2003) MoI. Endocrinol. 17, 1901-09.

[0196] Wei, P., et al. (2000) Nature 407, 920-923.

[0197] Weinstock, P. H., et al. (1997) J Lipid Res 38, 1782-1794.

[0198] Wetteran, J. R. et al. (1992) Science 258, 999-1001.

[0199] Wiwi, C. A., et al. (2004) MoI Endocrinol 18, 1977-1987.

[0200] Yamagata, K., et al. (1996) Nature 384, 458-460.

[0201] Yoon, J. C, et al. (2001) Nature 413, 131-18.

[0202] Zhang, J., et al. (2002) Science 298, 422-424.

[0203] Zhong, W., et al. (1994) MoI Cell Biol 14(11), 7276-7284.

Table 1

Claims

WHAT IS CLAIMED IS:

1. A mouse comprising a genetically modified HNF4α gene.

2. The mouse of claim 1, wherein the mouse has an altered lipid profile.

3. The mouse of claim 1, wherein the mouse has an altered pancreatic response to glucose.

4. The mouse of claim 1 , wherein the mouse expresses HNF4α7 in the liver and/or kidney.

5. The mouse of claim 1, wherein the mouse expresses HNF4αl in the stomach and/or pancreas.

6. A plasmid comprising the nucleic acid sequence of HNF4α wherein exon ID replaces exon IA under the regulatory control of the Pl promoter.

7. A plasma comprising the nucleic acid sequence of HNF4α. wherein exon IA replaces exon ID under the regulatory control of the P2 promoter.

8. An embryonic stem cell comprising the plasmid of claim 6 or the plasmid of claim 7.

9. A mouse comprising modified alleles from the embryonic stem cell of claim 8.

10. A primary or immortalized cell culture derived from the mouse of claim 9.

11. The mouse of claim 9, wherein the mouse expresses HNF4αl and does not express HNF4α7.

12. The mouse of claim 11, wherein the animal has an altered pancreatic response to glucose.

13. The mouse of claim 9, wherein the animal expresses HNF4α7 and does not express HNF4αl.

14. The mouse of claim 13, wherein the animal has an altered lipid profile.

15. The mouse of claim 14, wherein the transcription of apolipoproteins A4, C2, and/or C3 are altered.

16. A method of using the mouse of claim 1 or the cell culture of claim 10 to test the toxic effects of a drug comprising:

(a) providing the mouse of claim 11 and the mouse of claim 13, or providing the cell culture of claim 10;

(b) contacting said mice Or₁ said cultured cells with the drug; and

(c) evaluating the survival of the mouse of claim 13 compared to the mouse of claim 11, or evaluating the survival of the cells of claim 10 comprising the plasmid of claim 6 with the survival of the cells of claim 10 comprising the plasmid of claim 7.

16. A method of determining whether a substance is metabolized by a CAR- dependent activity, comprising

(a) providing the mouse of claim 13 ;

(b) placing the mouse in contact with the substance; and

(c) evaluating whether the mouse resists a dose harmful to a control animal.

17. A method of identifying in vivo analogues of HNF4α AFl comprising:

(a) administering a candidate analogue to the mouse of claim 11 ;

(b) assaying the expression of AFl -specific target genes in biological specimens; and

(c) comparing the results of the assay to that obtained from a biological specimen of an untreated mouse of claim 13.

18. The method of claim 17, wherein the specific target gene is chosen from ApoA4, AρoC2, and CAR.

19. A method of making a mouse genetically modified at the HNF4α locus comprising:

(a) providing a plasmid comprising at least a fragment of the HNF4α gene;

(b) introducing the plasmid into an embryonic stem cell;

(c) introducing the stem cell into a mouse embryo; and

(d) allowing the embryo to develop into a mouse.

20. The method of 19, wherein the mouse comprises exon IA on both chromosome alleles in place of exon ID.

21. The method of 19, wherein the mouse comprises exon ID on both chromosome alleles in place of exon IA.

22. An antagonist of the AF activity of one or more HNF4α protein, wherein the antagonist is capable of interfering with the activity of AF function in vivo and/or in vitro.

23. A method of identifying in vitro analogues of HNF4α AFs comprising:

(a) treating the cell culture of claim 10 with the antagonist of claim 22, or an analogue thereof; and

(b) comparing the expression of AF target genes in cell cultures derived from the mice of claim 11 to cell cultures derived from the mice of claim 13.