WO2013142571A2 - Assays for the identification of compounds that modulate lipid homeostasis - Google Patents

Abstract

The invention describes the role of regulated IRE 1 -dependent decay (RIDD) in modulating lipid homeostasis by degradation of particular lipid metabolism genes and the role of XBPl in modulating lipid homeostasis by regulating the transcriptional activation of particular lipogenic target genes. Methods for identifying modulators of lipid homeostasis based on modulation of IRE 1 -mediated RIDD of lipid metabolism genes or based on modulation of XBPl -mediated transcriptional regulation of lipogenic target genes are provided.

Description

ASSAYS FOR THE IDENTIFICATION OF COMPOUNDS

THAT MODULATE LIPID HOMEOSTASIS

Government Support

This invention was made with government support under AI032412, DK082448, and DK089211 awarded by National Institutes of Health. The government has certain rights in the invention.

Background of the Invention

In mammals, the liver is the principal organ that controls energy homeostasis through regulating carbohydrate and fatty acid metabolism. During starvation, the liver produces glucose to maintain circulating glucose levels by breaking down glycogen stores or by synthesizing glucose through gluconeogenesis. In contrast, ingestion of dietary carbohydrates promotes lipid synthesis in the liver to convert carbohydrate s to triglyceride (TG) for long-term energy storage (see e.g., Towle, H.C. et al. (1997) Annu. Rev. Nutri. 17:405-433). The incidence of metabolic syndrome, a condition characterized by the constellation of central obesity, dyslipidemia, hepatic steatosis, elevated blood glucose, and hypertension, continues to rise in industrialized nations (see e.g., Mensah, G.A. et al. (2004) Cardiol. Clin. 22:485-504). Dyslipidemia, manifested by elevated levels of plasma TG and low density lipoprotein (LDL) cholesterol and low levels of high density lipoprotein (HDL) cholesterol, is a risk factor for coronary artery disease (see e.g., Ginsberg, H.N. and Zhang, Y.L. (2006) Obesity (Silver Spring) 14 (Suppl. 1):41S-49S). Control of dyslipidemia in patients with coronary artery disease with statins and with triglyceride- lowering agents has resulted in measurable improvements in cardiovascular morbidity and mortality, suggesting clinical benefits of lowering hepatic de novo lipid synthesis (see e.g., Brunzell, J.D. (2007) N. Engl. J. Med. 357: 1009-1017).

The transcription factor XBP1 has been shown to be involved in the regulation of hepatic lipogenesis (for a review, see Glimcher, L.H. and Lee, A-H. (2009) Ann. N. Y. Acad. Sci.

l_173(Suppl. 1):E2-E9). Deficiency of XBP1 in the liver in mice led to profound decreases in serum TG, cholesterol and free fatty acids without causing hepatic steatosis, and these low plasma lipid levels were primarily due to decreased de novo synthesis of lipids in the liver. Additionally, XBP1 protein expression is induced in the liver by a high carbohydrate diet and directly controls the induction of critical genes involved in fatty acid synthesis (Lee, A-H. et al. (2008) Science 320: 1492-1496). The mechanism by which XBP1 deficiency leads to

hypolipidemia, however, remains unclear.

XBP1 is activated by a post-transcriptional modification of its mRNA by inositol requiring enzyme 1 (IREl), a serine-threonine protein kinase and endoribonuclease (Calfon, M et al. (2002) Nature 415:92-96: Yoshida, H. et al. (2001) Cell 102:881-891). IREl (IRElcc and IREIP) is an endoplasmic reticulum (ER) transmembrane protein that is involved in the unfolded protein response (UPR) that occurs in response to ER stress (see e.g., Lee, A-H and Glimcher L.H. (2009) Cell. Mol. Life Sci. 66:2835-2850). It also has been shown that activated IREl degrades a subset of mRNAs in a process referred to as regulated IREl -dependent decay (RIDD) (see e.g., Hollier, J. et al. (2009) J. Cell. Biol. 186:323-331 : Hur, K.Y. et al. (2012) J. Exp. Med. 209:307-318). The precise relationship between XBP1 and IREl and lipid homeostasis, however, remains unclear.

In view of the foregoing, screening assays based on elucidation of biological mechanisms involved in lipid homeostasis, in particular those involving the critical regulator protein XBP1, would be very useful and are still needed to facilitate identification of compounds that may be useful in the treatment of various disorders associated with dysregulation of lipid homeostasis.

Summary of the Invention

This invention provides screening assays based on elucidation of biological mechanisms involving XBP1 and IREl that are important in regulating lipid homeostasis. In particular, the invention described herein discloses the discovery that RIDD is a crucial control mechanism of lipid homeostasis. Suppression of RIDD by RNA interference or genetic ablation of

IRE la reversed hypolipidemia in XBP1 deficient mice. Moreover, comprehensive microarray analysis of XBP1 and/or IREl a deficient liver identified genes involved in lipogenesis and lipoprotein metabolism as RIDD substrates, which may contribute to the suppression of plasma lipid levels by activated IRE la. Comprehensive microarray analysis also identified XBP1 dependent genes that represent lipogenic target genes whose transcription is directly regulated by XBP1. Additionally, ablation of XBP1 ameliorated hepatosteatosis, liver damage and hypercholesterolemia in dyslipidemic animal models, suggesting that direct targeting of either IREl a or XBP1 may be a feasible strategy to treat dyslipidemias. Accordingly, in one aspect, the invention provides a method of identifying candidate modulators of lipid homeostasis comprising, a) determining the ability of a test compound to modulate regulated IRE 1 -dependent decay (RIDD) of a lipid metabolism gene, wherein mRNA of the lipid metabolism gene is degraded by IRE1; and b) selecting said test compound as a compound of interest when said compound modulates RIDD of the lipid metabolism gene, to thereby identify candidate modulators of lipid homeostasis.

In another aspect, the invention provides a method of identifying candidate modulators of lipid homeostasis comprising, a) determining the ability of a test compound to modulate XBP1- mediated transcription of a lipogenic target gene; and selecting said test compound as a compound of interest when said compound modulates XBPl -mediated transcription of the lipogenic target gene, to thereby identify candidate modulators of lipid homeostasis.

In one embodiment, modulation of lipid homeostasis comprises modulation of lipid metabolism. In another embodiment, modulation of lipid metabolism comprises induction of lipid clearance from blood. In another embodiment, modulation of lipid homeostasis comprises modulation of lipogenesis. In another embodiment, modulation of lipogenesis comprises induction of lipogenesis.

In one embodiment, the compound inhibits RIDD of mRNAs of lipid metabolism genes. In another embodiment, the compound stimulates RIDD of mRNAs of lipid metabolism genes. In another embodiment, the compound stimulates XBPl -mediated transcription of the lipogenic target gene. In another embodiment, the compound inhibits XBPl -mediated transcription of the lipogenic target gene.

In a preferred embodiment, the lipid metabolism gene or lipogenic target gene is a hepatic gene.

In one embodiment, the lipogenic target gene is selected from the group consisting of Sdf2ll, Cflar, Dnajc3, Dnajb9, Gale, Nans, Edeml, Dnaljbll, Tmem39a, Erolb, Gmppa, Hyoul, Usol, Sec24d, Slc35bl, Uggtl, Ssr3, Ubxn4, Txndc5, Stt3a, Ssrl, Serpl, Ddost, Sec61al, Surf4, 0610007 LOlRik, Rrbpl, Ostc, Odd, Cnpy3, Cdk5rap3, Fdps, Spcs3, P4hb, Spcs2,

2810482107Rik, Sec61b, Lman2, Krtcap2, Secllc, Sec61g, Ssr4, Sec63, HI 3, D17Ws l04e, Chidl, Sndl, Ikbke, Traml, Cbaral, Slc33al, Pexllc, Mlec, Sale, Cyp51, Tnfaip8ll, Pmvk, Fdtfl, Hsdl7b7, Anubll, Lifr, Sc4mol, Mvk, and Idil In one embodiment, the lipid metabolism gene is selected from the group consisting of Paqr7, Ptprf, Ces3, AU018778, Cesl, Ceslc, Cesld, Cesle, Ceslf, Ceslg, Lrpl, Bloclsl, Dhcr7, Limd2, Ml, 221041 lKHRik, Es22, Apon, Glrx, Fads2, He, Oaf, Mvk, Gjbl, Sc4mol, Plala, F7, Cpn2, Pxmp2, Abca3, Gm2a, Kcnn2, Nsdhl, Stim2, Dnase2b, Vnn3, Hsdl7b7, Hgsnat, Coll8al, Aldoc, Fadsl, Fnl, Smarcc2, Dera, Spp2, Angptl3, Atrnll, Tppl, Faml08a, Tmprss6, Sidt2, Esl, Dbt, Ropnll, Tex264, Tapbp, Selk, Btbdl, Cyp2c68, Ndufa9, Pmvk, Blvra, lgf2r, Proc, Abca2, Taldol, Bola3, Cypla2, Retsat, Lampl, Rnfl30, Tex2, Afin, Jkamp,

0610007 C21Rik, Plbd2, Cyp51, Tm7sf2, Cl , Hgfac, 3110073H01Rik, Furin, C920025E04Rik, Serpingl, Lipc, Ttcl3, Coll8al, Slc27a5, Pcsk6, Tmem208, PmlOdl, Sqle, Man2a2, Pplrl5b, Azgpl, Ces6, Pla2gl2b, llvbl, Adam9, Mlec, Pdapl, F2, Commd9, Itih4, Dgat2, Cyb561d2, Slc35f5, Algl, Gnptg, Rarresl, Yifla, Slc25al7, Mxd4, Fdps, Papola, Erp44 and Rpnl. In a preferred embodiment, the lipid metabolism gene is selected from the group consisting of Ceslc, Cesld, Cesle, Ceslf, Ceslg and Angplt3.

In one embodiment, the test compound is comprised in a combination of test compounds, such as a library of test compounds.

In one embodiment, determining the ability of a test compound to modulate RIDD of the mRNA of the lipid metabolism gene comprises contacting an indicator composition {e.g., test cell, cell extract or assay composition) with the test compound and determining a level of lipid metabolism gene mRNA in the presence of the test compound. Typically, the indicator composition comprises IRE1 and mRNA of at least one lipid metabolism gene.

In another embodiment, determining the ability of a test compound to modulate REDD of the lipid metabolism gene comprises contacting an indicator composition {e.g., test cell, cell extract or assay composition) with the test compound and determining a level of lipid metabolism gene-encoded protein in the presence of the test compound. Typically, the indicator composition comprises IRE1 and mRNA of at least one lipid metabolism gene.

In another embodiment, determining the ability of a test compound to modulate XBP1- mediated transcription of the lipogenic target gene comprises contacting an indicator composition {e.g., test cell, cell extract or assay composition) with the test compound and determining a level of lipogenic target gene mRNA in the presence of the test compound.

Typically, the indicator composition comprises XBPl and at least one gene encoding at least one lipogenic target gene. In another embodiment, determining the ability of a test compound to modulate XBP1- mediated transcription of the lipogenic target gene comprises contacting an indicator composition (e.g., test cell, cell extract or assay composition) with the test compound and determining a level of a reporter of lipogenic target gene transcription in the presence of the test compound. Typically, the indicator composition comprises XBPl and at least one reporter of a lipogenic target gene.

In another embodiment, the test compound is further determined to modulate XBP1- mediated transcription of the lipogenic target gene independent of unfolded protein response (UPR) activation.

In another aspect, the invention provides a method for validating a compound as a compound useful in modulating lipid homeostasis in vivo comprising, selecting a candidate compound identified according to one of the screening methods described hereinbefore, and testing said candidate compound for modulation of lipid metabolism in an animal model of dyslipidemia. In various embodiments, the animal is tested for modulation of one or more of: plasma cholesterol levels, plasma and/or hepatic triglyceride levels, plasma and/or hepatic lipid levels, steatosis, and lipoprotein metabolism.

In yet another aspect, the invention provides a method for validating a compound as a compound useful in modulating lipid homeostasis in vivo comprising, selecting a candidate compound identified according to one of the screening methods described hereinbefore, and testing said candidate compound for modulation of lipid metabolism gene activity in an animal model deficient in XBPl and/or IRE la.

Kits for carrying out the methods described herein are also encompassed by the invention.

Description of the Figures

Figures 1A-1E show results from experiments demonstrating partial restoration of plasma lipid levels by IRE la silencing in XBPl deficient mice. Figure 1 A shows plasma TG levels of male mice with the indicated genotypes measured at fed state. Figure IB shows cholesterol levels of male mice with the indicated genotypes measured at fed state. Figure 1C shows results from experiments in which Xbpl^KLO mice were i.v. injected with siRNAs targeting luciferase or IRE la mRNA and plasma TG levels were measured at indicated time points, n = 8-9 per group. Figure ID shows results from experiments in which Xbpl ^L0 mice were i.v. injected with siRNAs targeting luciferase or IRElcc mRNA and plasma cholesterol levels were measured at indicated time points, n = 8-9 per group. Figure IE shows results from experiments in which Xbpl mice were sacrificed 2 or 8 days after siRNA or PBS injection and hepatic mRNA levels were measured by qRT-PCR. Values represent mean ± s.e.m. n = 2 (PBS), or 3-6 (siRNA). *i° <0.05, **P <0.01, ***/> <0.001.

Figures 2A-2D show results from experiments demonstrating plasma lipid levels in IRE la, and IREla XBPl double deficient mice. Figure 2A shows plasma TG levels and Figure 2B shows plasma cholesterol levels that were measured before and 3 wks after poly (I:C) injection of male Ernl^{f f} and Ernl^f/f;Mx l-cre mice. Bloods were drawn at fed state. Figure 2C shows plasma TG

f/f f/f f/f f /f levels and Figure 2D shows plasma cholesterol levels of Ernl ;Xbpl and Ernl ;Xbpl ;Mx l- cre mice before and after poly(I:C) injection.

Figures 3A-3C show results of experiments to determine the identification of direct XBPl targets and RIDD substrates by microarray analysis. Figure 3A shows comparison of gene expression profiles between WT and XbplA; WT and IrelA; untreated (NT) and tunicamycin-treated; siLuc and silrel injection to Xbpl^LK°. Shown are data for 237 genes that were decreased in XbplA liver. Figure 3B shows mRNA levels of 64 genes that were suppressed in both XbplA and IrelA liver in individual liver samples. Figure 3C shows mRNA levels of 1 12 genes that were induced by IRE la siRNA in Xbpl^LKO liver. Genes involved in protein folding processes in the ER are shown in blue. Lipid metabolism genes are highlighted in red.

Figures 4A-4I show the results of experiments to determine the identification of Cesl and Angptl3 mRNAs as RIDD substrates. Figure 4A shows microarray signals for Cesl genes.

Figure 4B shows Western blot analysis of liver lysates of WT and Xbpl^LKO mice using TGH (Cesld) antibody which also detects Es-X {Ceslg). Figure 4C shows Angptl3, -4, and -6 mRNA levels that were measured in the liver of WT and XBPl deleted male mice, n = 4 per group. Figure 4D shows plasma Angptl3 protein levels in WT and XbplA mice measured by ELISA. Figure 4E shows hepatic Angptl3 mRNA levels measured 8 days after siRNA injection of Xbpl mice, n = 3-5 per group. Figure 4F shows plasma AngptB protein levels measured 4 or 8 days after siRNA injection to Xbpl^{L 0} mice, n = 3-5 per group. ***p <0.001 compared with the siLuc group. Figure 4G shows results from in vitro cleavage experiments in which in vitro transcribed AngptB mRNA was incubated with recombinant IRE la and then resolved on an agarose gel. Angptl3mut was generated by changing the two G residues to C shown in bold face in Figure 4H. Figure 4H shows the predicted secondary structure of AngptB mRNA with a potential IREloc cleavage site that is depicted by an arrow. Figure 41 shows results from in vivo cleavage experiments in which WT and mutant AngptB constructs were transfected into 293T cells together with WT or mutant IRE la constructs. EGFP plasmid was also included in the transfection cocktail and served as a normalization control. AngptB mRNA levels in the transfected cells were measured by qRT-PCR.

Figures 5A-C show the results of experiments demonstrating the effects of XBP1 ablation in dyslipidemic mouse models. Figure 5A shows results of experiments in which metabolic parameters of 8- week old Xbpl^{f /f};ob/ob and Xbpl^{L O};ob/ob mice measured after a 4-h fasting, n = 4-6 female mice per group. Figure 5b shows hepatic mRNA levels that were measured by qRT-PCR. Figure 5C shows results from experiments in which WT and XbplA mice were fed a high fat diet for 5 months, and then sacrificed to measure metabolic parameters, n = 9-10 male mice per group. **P <0.01, ***P <0.001.

Figures 6A-6E show the results of experiments demonstrating ablation of XBP1 in apoE knock out mice reduces plasma lipids. Figures 6A-6B show results from experiments in which Apoe^"'^" ;Xbpl^{f /f} and Apoe^^'jXbp^ ^f;Mx-cre mice were injected with poly(I:C) on day 0, day 4, week 12 and week 20. Bloods were drawn at fed state to measure plasma TG (Figure 6A) and plasma cholesterol levels (Figure 6B). n = 8-9 per group. **P <0.01, <0.001 compared with the control group (Apoe^{_ "};Xbpl^{f f} mice). Figure 6C shows results from experiments in which plasma collected at weeks 8 and 17 were pooled and subjected to FPLC analysis. Figures 6D-6E show results from experiments in which descending aorta (Figure 6D) and aortic arch (Figure 6E) were stained en face with oil red O (ORO) and ORO positive areas were quantified. Figures 7A-7I show results from experiments involving silencing of XBPl mRNA in vivo. Figure 7A shows results from experiments in which female C57BL/6 mice were i.v. injected with PBS, luciferase siRNA (5 mg/kg), or varying doses of XBPl siRNA. Mice were sacrificed 48 hrs later. Hepatic XBPl mRNA levels were measured by qRT-PCR. n = 4 mice per group. ***p <0.001 compared with the control groups (PBS and luciferase siRNA groups). Figure 7B shows results from a Western blot to measure XBPls protein levels in pooled nuclear extracts. *non-specific band. Figure 7C shows results from experiments in which mice were injected with siRNAs at 7.5 mg/kg, and sacrificed after indicated days to measure hepatic XBPls protein levels by western blot. SP1 serves as a loading control, n = 3-4 mice per group. Figure 7D shows results of qRT-PCR analysis of hepatic gene expression, n = 3 per group. Figure 7E shows results of Western blot analysis of IREla in liver lysates. The panel shows IREla

phosphorylation measured by Phos-Tag western blot. Figures 7F-7I show the results of experiments in which Male C57B1J6 mice (Figures 7F-7G), or Apoe^"'^" mice (Figures 7Η-7Γ) were i.v. injected with siRNAs (7.5 mg/kg), and bled to measure plasma lipid levels, n = 5 -8 per group. *P <0.05, ***P <0.001.

Figure 8 shows the expression of lipogenic genes in Xbpl^LKO mice determined by qRT-PCR. Values represent mean ± s.e.m. n = 3-6. *P <0.05, **P <0.01, ***P <0.001.

Figure 9 A shows a schematic representation of Cesl gene cluster.

Figure 9B shows an alignment of predicted amino acid sequences of mouse Cesl proteins. Conserved sequences are highlighted in red.

Figures 10A-10F shows results from experiments demonstrating effects of XBPl ablation in dyslipidemic mouse models. Figures 10A-10B show H&E staining of liver sections. Scale bar = 200 μπι. Figure 10C-10D shows results from experiments in which WT and XbplA mice fed high fat diet for 4 months were subjected to glucose tolerance test (Figure 10C), and Insulin tolerance test (Figure 10D). Figure 10E shows the results of Western blot analysis and Figure 10F shows the results of qRT-PCR analysis of UPR markers in ob/ob mouse liver. *P <0.05, **P <0.01, ***P <0.001. Detailed Description of the Invention

The present invention is based, at least in part, on the elucidation of biological mechanisms involving XBPl and IRE1 that are important in regulating lipid homeostasis. In particular, the invention described herein discloses the discovery that RIDD is a crucial control mechanism of lipid homeostasis. Suppression of RIDD by RNA interference or genetic ablation of IRE la reversed hypolipidemia in XBPl deficient mice. Moreover, comprehensive microarray analysis of XBPl and/or IRE la deficient liver identified genes involved in lipogenesis and lipoprotein metabolism as REDD substrates, which may contribute to the suppression of plasma lipid levels by activated IREla. Comprehensive microarray analysis also identified XBPl dependent genes that represent lipogenic target genes whose transcription is directly regulated by XBPl Additionally, abalation of XBPl ameliorated hepatosteatosis, liver damage and hypercholesterolemia in dyslipidemic animal models, suggesting that direct targeting of either IREla or XBPl may be a feasible strategy to treat dyslipidemias.

Various aspects of the invention are described in further detail in the following subsections: I. Definitions

As used herein, the term "XBPl" refers to a X-box binding human protein that is a DNA binding protein and has an amino acid sequence as described in, for example, Liou, H-C. et. al. (1990) Science 247:1581-1584 and Yoshimura, T. et al. (1990) EMBO J. 9:2537-2542, and other mammalian homologs thereof, such as described in Kishimoto T. et al., (1996) Biochem. Biophys. Res. Commun. 223:746-751 (rat homologue). Exemplary proteins intended to be encompassed by the term "XBPl " include those having amino acid sequences disclosed in GenBank with accession numbers A36299 [gi: 105867], NP.sub.--005071 [gi:4827058], P17861 [gi:139787], CAA39149 [gi:287645], and BAA82600 [gi:5596360] or e.g., encoded by nucleic acid molecules such as those disclosed in GenBank with accession numbers AF027963 [gi: 13752783]; NM.sub.-013842 [gi: 13775155]; or M31627 [gi: 184485]. XBPl is also referred to in the art as TREB5 or HTF (Yoshimura et al. 1990. EMBO Journal. 9:2537; Matsuzaki et al. 1995. J. Biochem. 117:303).

XBPl is a basic region leucine zipper (b-zip) transcription factor isolated independently by its ability to bind to a cyclic AMP response element (CRE)-like sequence in the mouse class II MHC Aoc gene or the CRE-like site in the HTLV-1 21 base pair enhancer, and subsequently shown to regulate transcription of both the DRa and HTLV-1 ltr gene. Like other members of the b-zip family, XBPl has a basic region that mediates DNA-binding and an adjacent leucine zipper structure that mediates protein dimerization. Deletional and mutational analysis identified transactivation domains in the C-terminus of XBPl in regions rich in acidic residues, glutamine, serine/threonine and proline/glutamine.

There are two forms of XBPl protein, unspliced and spliced, which differ markedly in their sequence and activity. Unless the form is referred to explicitly herein, the term "XBPl " as used herein includes both the spliced and unspliced forms. Spliced XBPl protein directly controls the activation of the unfolded protein response (UPR), while unspliced XBPl functions in this pathway only due to its ability to negatively regulate spliced XBPl.

As used herein, the term "spliced XBPl" refers to the spliced, processed form of the mammalian XBPl mRNA or the corresponding protein. Human and murine XBPl mRNA contain an open reading frame (ORF1) encoding bZIP proteins of 261 and 267 amino acids, respectively. Both mRNA's also contain another ORF, ORF2, partially overlapping but not in frame with ORF1. ORF2 encodes 222 amino acids in both human and murine cells. Human and murine ORF1 and ORF2 in the XBPl mRNA share 75% and 89% identity respectively. In response to ER stress, XBPl mRNA is processed by the ER transmembrane endoribonuclease and kinase IREl which excises an intron from XBPl mRNA. In murine and human cells, a 26 nucleotide intron is excised. The boundaries of the excised introns are encompassed in an RNA structure that includes two loops of seven residues held in place by short stems. The RNA sequences 5' to 3' to the boundaries of the excised introns form extensive base-pair interactions. Splicing out of 26 nucleotides in murine and human cells results in a frame shift at amino acid 165 (the numbering of XBPl amino acids herein is based on GenBank accession number NM.sub.--013842 [gi: 13775155] and one of skill in the art can determine corresponding amino acid numbers for XBPl from other organisms, e.g., by performing a simple alignment). This causes removal of the C-terminal 97 amino acids from the first open reading frame (ORF1) and addition of the 212 amino from ORF2 to the N-terminal 164 amino acids of ORFl containing the b-ZIP domain. In mammalian cells, this splicing event results in the conversion of a 267 amino acid unspliced XBP1 protein to a 371 amino acid spliced XBP1 protein. The spliced XBP1 then translocates into the nucleus where it binds to its target sequences to induce their transcription. The nucleic acid and amino acid sequence of the spliced form of murine XBP1 are also shown in FIG. 8C and 8D, respectively, of US Publication No. 20040170622.

As used herein, the term "unspliced XBP1" refers to the unprocessed XBP1 mRNA or the corresponding protein. As set forth above, unspliced murineXBPl is 267 amino acids in length and spliced murine XBP1 is 371 amino acids in length. The sequence of unspliced XBP1 is known in the art and can be found, e.g., Liou, H-C. et. al. (1990) Science 247:1581-1584 and Yoshimura, T. et al. (1990) EMBO J. 9:2537-2542, or at GenBank accession numbers NM.sub.- 005080 [gi: 14110394] or NM.sub.-013842 [gi:13775155]. The nucleic acid and amino acid sequence of the unspliced form of murine XBP1 are also shown in FIG. 8A and 8B, respectively, of US Publication No. 20040170622.

As used herein, the term "Unfolded Protein Response" (UPR) or the "Unfolded Protein Response pathway" refers to an adaptive response to the accumulation of unfolded proteins in the ER and includes the transcriptional activation of genes encoding chaperones and folding catalysts and protein degrading complexes as well as translational attenuation to limit further accumulation of unfolded proteins. Both surface and secreted proteins are synthesized in the endoplasmic reticulum (ER) where they need to fold and assemble prior to being transported.

As used herein, the term "IREl " refers to an ER transmembrane endoribonuclease and kinase called "inositol requiring enzyme 1 " which oligomerizes and is activated by

autophosphorylation upon sensing the presence of unfolded proteins* see, e.g., Shamu et al., (1996) EMBO J. 15: 3028-3039. In Saccharomyces cerevisiae, the UPR is controlled by IREp. In the mammalian genome, there are two homologs of IREl, IREla and Κ,ΕΙ β. IREla is expressed in all cells and tissue whereas ΠΙΕΙβ is primarily expressed in intestinal tissue. The endoribonucleases of either IREla and ΠΙΕΙβ are sufficient to activate the UPR. Accordingly, as used herein, the term "IREl" includes, e.g., IREla, Π¾Ε1β and IREp. In a preferred embodiment, IREl refers to IREla.

IREl is a large protein having a transmembrane segment anchoring the protein to the ER membrane. A segment of the IREl protein has homology to protein kinases and the C-terminal has some homology to RNAses. Over-expression of the IREl gene leads to constitutive activation of the UPR. Phosphorylation of the IREl protein occurs at specific serine or threonine residues in the protein. IREl senses the overabundance of unfolded proteins in the lumen of the ER. The oligomerization of this kinase leads to the activation of a C-terminal endoribonuclease by trans-autophosphorylation of its cytoplasmic domains. IREl uses its endoribonuclease activity to excise an intron from XBP1 mRNA. Cleavage and removal of a small intron is followed by re-ligation of the 5' and 3' fragments to produce a processed mRNA that is translated more efficiently and encodes a more stable protein (Calfon et al. (2002) Nature 415(3): 92-95). The nucleotide specificity of the cleavage reaction for splicing XBP1 is well documented and closely resembles that for IREp mediated cleavage of HACl mRNA (Yoshida et al. (2001) Cell 107:881-891). In particular, IREl mediated cleavage of murine XBP1 cDNA occurs at nucleotides 506 and 532 and results in the excision of a 26 base pair fragment for mouse XBP1. IREl mediated cleavage of XBP1 derived from other species, including humans, occurs at nucleotides corresponding to nucleotides 506 and 532 of murine XBP1 cDNA, for example, between nucleotides 502 and 503 and 528 and 529 of human XBP1.

As used herein, the term "lipid homeostasis" refers to the physiological processes and biological mechanisms and pathways involved in the maintenance of an internal metabolic equilibrium or steady- state of lipid within an organism or cell.

As used herein, the term "lipid metabolism" refers to the physiologic and metabolic processes involved in the assimilation of dietary lipids and the biosynthesis (anabolism) and degradation (catabolism) of lipids.

As used herein, the term "lipogenesis" refers to the process by which lipids (fats) are formed in the body, typically from glucose and other substrates, in particular the formation of fatty acids from acetyl coenzyme A. Lipogenesis encompasses the processes of fatty acid synthesis and subsequent triglyceride synthesis (when fatty acids are esterified with glycerol to form fats).

As used herein, the term "dyslipidemia" refers to an abnormal amount of lipids {e.g., cholesterol, triglycerides, lipoproteins and/or other fats) in the blood circulation {e.g., plasma levels). The term "hyperlipidemia: refers to a disorder of lipid metabolism that results in abnormally high levels of cholesterol, triglycerides, lipoproteins and/or other fats in the blood circulation {e.g., plasma levels). Hyperlipidemia is a key contributor to various dyslipidemia disorders including, but not limited, to atherosclerosis, coronary artery disease, peripheral vascular disease, obesity, metabolic syndrome, type II diabetes and pancreatitis.

As used herein, the term "lipid metabolism gene" refers to a gene, and its corresponding gene product, that is directly or indirectly involved in lipid metabolism.

As used herein, the term "lipogenic gene" refers to a gene, and its corresponding gene product, that is directly or indirectly involved in lipogenesis.

As used herein, the term "lipogenic target gene" refers to a gene, and its corresponding gene product, that is directly or indirectly involved in lipogenesis and that is a target for transcriptional regulation by XBP1.

As used herein, the term "regulated IRE 1 -dependent decay" or "RIDD" refers to a biological process of mRNA degradation that is mediated by IRE1 , in particular the

endonuclease activity of activated IRE1, as described in, for example, Hollier, J. et al. (2009) J. Cell. Biol. 186:323-331 : Hur, K.Y. et al. (2012) J. Exp. Med. 209:307-318.

The various forms of the term "modulate" include stimulation (e.g., increasing or upregulating a particular response or activity) and inhibition (e.g., decreasing or downregulating a particular response or activity).

As used herein, the term "contacting" (e.g., contacting a cell with a compound) is intended to include incubating the compound and the cell together in vitro (e.g., adding the compound to cells in culture) or administering the compound to a subject such that the compound and cells of the subject are contacted in vivo. The term "contacting" is not intended to include exposure of cells to a lipid homeostasis modulator that may occur naturally in a subject (i.e., exposure that may occur as a result of a natural physiological process).

As used herein, the term "compound of interest" or "test compound" or "candidate modulator" includes a compound that has not previously been identified as, or recognized to be, a modulator of lipid homeostasis.

The term "library of test compounds" is intended to refer to a panel comprising a multiplicity of test compounds.

In one embodiment, small molecules can be used as test compounds. The term "small molecule" is a term of the art and includes molecules that are less than about 7500, less than about 5000, less than about 1000 molecular weight or less than about 500 molecular weight. In one embodiment, small molecules do not exclusively comprise peptide bonds. In another embodiment, small molecules are not oligomeric. Exemplary small molecule compounds which can be screened for activity include, but are not limited to, peptides, peptidomimetics, nucleic acids, carbohydrates, small organic molecules (e.g., Cane et al. 1998. Science 282:63), and natural product extract libraries. In another embodiment, the compounds are small, organic non- peptidic compounds. In a further embodiment, a small molecule is not biosynthetic. For example, a small molecule is preferably not itself the product of transcription or translation.

As used herein, the term "indicator composition" refers to a composition that includes a protein of interest (e.g., IRE1 , such as IRElcc, or XBP1), for example, a cell that naturally expresses the protein, a cell that has been engineered to express the protein by introducing an expression vector encoding the protein into the cell, or a cell free composition that contains the protein (e.g., purified naturally-occurring protein or recombinantly-engineered protein).

As used herein, the term "cell free composition" refers to an isolated composition, which does not contain intact cells. Examples of cell free compositions include cell extracts and compositions containing isolated proteins.

As used herein, the term "reporter gene" refers to any gene that expresses a detectable gene product, e.g., RNA or protein. Preferred reporter genes are those that are readily detectable. The reporter gene can also be included in a construct in the form of a fusion gene with a gene that includes desired transcriptional regulatory sequences or exhibits other desirable properties. Examples of reporter genes include, but are not limited to CAT (chloramphenicol acetyl transferase) (Alton and Vapnek (1979), Nature 282: 864-869) luciferase, and other enzyme detection systems, such as beta-galactosidase; firefly luciferase (deWet et al. (1987), Mol. Cell. Biol. 7:725-737); bacterial luciferase (Engebrecht and Silverman (1984), PNAS 1 : 4154-4158; Baldwin et al. (1984), Biochemistry 23: 3663-3667); alkaline phosphatase (Toh et al. (1989) Eur. J. Biochem. 182: 231 -238, Hall et al. (1983) J. Mol. Appl. Gen. 2: 101), human placental secreted alkaline phosphatase (Cullen and Malim (1992) Methods in Enzymol. 216:362-368) and green fluorescent protein (U.S. Pat. No. 5,491 ,084; WO 96/23898).

As used herein, the term "XBP1 -responsive element" refers to a DNA sequence that is directly or indirectly regulated by the activity of the XBP1 (whereby activity of XBP1 can be monitored, for example, via transcription of a reporter gene).

As used herein, the term "cells deficient in XBP1 " includes cells of a subject that are naturally deficient in XBP1, as wells as cells of a non-human XBP1 deficient animal, e.g., a mouse, that have been altered such that they are deficient in XBPl . The term "cells deficient in XBPl " is also intended to include cells isolated from a non-human XBPl deficient animal or a subject that are cultured in vitro.

As used herein, the term "non-human XBPl deficient animal" refers to a non-human animal, preferably a mammal, more preferably a mouse, in which an endogenous gene has been altered by homologous recombination between the endogenous gene and an exogenous DNA molecule introduced into a cell of the animal, e.g., an embryonic cell of the animal, prior to development of the animal, such that the endogenous XBPl gene is altered, thereby leading to either no production of XBPl or production of a mutant form of XBP\1 having deficient XBPl activity. Preferably, the activity of XBPl is entirely blocked, although partial inhibition of XBPl activity in the animal is also encompassed. The term "non-human XBPl deficient animal" is also intended to encompass chimeric animals (e.g., mice) produced using a blastocyst

complementation system, such as the RAG-2 blastocyst complementation system, in which a particular organ or organs (e.g., the lymphoid organs) arise from embryonic stem (ES) cells with homozygous mutations of the XBPl gene.

As used herein, the term "cells deficient in ERE1 " includes cells of a subject that are naturally deficient in IRE1, as wells as cells of a non-human IRE1 deficient animal, e.g., a mouse, that have been altered such that they are deficient in IRE1. The term "cells deficient in IRE1 " is also intended to include cells isolated from a non-human IRE1 deficient animal or a subject that are cultured in vitro.

As used herein, the term "non-human IRE1 deficient animal" refers to a non-human animal, preferably a mammal, more preferably a mouse, in which an endogenous gene has been altered by homologous recombination between the endogenous gene and an exogenous DNA molecule introduced into a cell of the animal, e.g., an embryonic cell of the animal, prior to development of the animal, such that the endogenous IRE1 gene is altered, thereby leading to either no production of IRE1 or production of a mutant form of IRE1 having deficient IRE1 activity. Preferably, the activity of IRE1 is entirely blocked, although partial inhibition of IRE1 activity in the animal is also encompassed. The term "non-human IRE1 deficient animal" is also intended to encompass chimeric animals (e.g., mice) produced using a blastocyst

complementation system, such as the RAG-2 blastocyst complementation system, in which a particular organ or organs (e.g., the lymphoid organs) arise from embryonic stem (ES) cells with homozygous mutations of the IREl gene.

II. Role of XBP1 and IREla in Regulation of Lipid Homeostasis

The unfolded protein response (UPR) was originally identified as a signaling system that promotes the transcription of endoplasmic reticulum (ER) chaperone^' genes in response to stresses that burden the ER with increased client proteins for folding (Ron, D. and Walter, P. (2007) Nat Rev Mol Cell Biol 8:519-529; Schroder, M. and Kaufman, R.J. (2005) Annu Rev Biochem 74:739-789). In mammals, UPR is initiated by three families of unique ER

transmembrane proteins, PERK, IREl (IREla and ΠΙΕΙβ), and ATF6 (ATF6a and ATF6 ). IREl is evolutionarily well conserved in all eukaryotes from unicellular organisms to mammals, while the other UPR branches are present only in higher eukaryotes (Mori, K. et al. (1993) Cell 74:743-756; Ron, D. and Walter, P. (2007) Nat Rev Mol Cell Biol 8:519-529; Wang, X.Z. et al. (1998) EMBO J 17:5708-5717; Cox, J.S. et al. (1993) Cell 73: 1197-1206).

XBP1 is the only known transcription factor downstream of IREla that is activated through an unconventional mRNA splicing reaction. XBP1 activates the transcription of a variety of genes involved in protein secretory pathways (Acosta-Alvear, et al. (2007) Mol Cell 27:53-66; Shaffer, A.L. et al. (2004) Immunity 21:81-93; Lee, A.H. et al. (2003) Mol Cell Biol 23:7448-7459). In line with this, IREla and XBP1 are required for the development, survival and the protein secretory function of some professional secretory cells (Lee, A.H. et al. (2005) Embo J 24:4368-4380; Reimold, A.M. et al. (2001) Nature 412:300-307; Kaser, A. et al. (2008) Cell 134:743-756; Huh, W.J. et al. (2010) Gastroenterology 139:2038-2049; Iwawaki, T. et al. (2010) PLoS One 5, el3052; Zhang, K. et al. (2005) J Clin Invest 115:268-281. In addition to activating XBP1, IREla can also activate Jun N-terminal kinase (JNK) (Urano, F. et al. (2000) Science 287:664-666) and induce the degradation of certain mRNAs, a process known as regulated IREl -dependent decay (RIDD) (Han, D. et al. (2009) Cell 138:562-575; Hollien, J. et al. (2009) J Cell Biol 186:323-331 ; Lipson, K.L. et al. (2008) PLoS One 3, el 648; Oikawa, D. et al. (2010) Nucleic Acids Res 38:6265-6273; Lee, A.H. et al. (2011) Proc Natl Acad Sci U S A 108:8885-8890). The physiological significance of REDD was first explored in insect cells, where it was postulated to be a mechanism to reduce ER stress by limiting the entry of cargo proteins to the ER, given the preferential degradation of mRNAs encoding secretory proteins by RIDD (Hollien, J. and Weissman, J.S. (2006) Science 313:104-107). Interestingly, in mammalian cells, IRE la appears to cleave mRNAs encoding not only secretory cargo proteins, but also ER resident proteins that serve in protein folding and secretory pathways. This has led to the hypothesis that ERE la might promote apoptosis under severe ER stress conditions by diminishing ER capacity to handle stress (Han, D. et al. (2009) Cell 138:562-575). The in vivo functions of REDD are only beginning to be explored. We and others have demonstrated that IRE la degrades insulin mRNA as well as proinsulin-processing enzyme mRNAs, uncovering an important function of REDD in insulin secretion from β cells (Han, D. et al. (2009) Cell 138:562- 575; Lee, A.H. et al. (2011) Proc Natl Acad Sci U S A 108:8885-8890 Lipson, K.L. et al. (2008) PLoS One 3, el648).

We previously reported that XBP1 ablation in the liver profoundly decreased plasma triglyceride (TG) and cholesterol levels in mice, revealing an important role for this factor in hepatic lipid metabolism (Lee, A.H. et al. (2008) Science 320:1492-1496). Contrary to our speculation that XBP1 deficiency might induce ER stress in hepatocytes, leading to decreased very-low-density lipoprotein (VLDL) secretion, XBP1 deficient hepatocytes did not exhibit morphological signs of ER dysfunction, defects in apoB 100 secretion, TG accumulation, increased apoptosis, or activation of XBP1 independent UPR markers, arguing against the contribution of ER stress to the hypolipidemic phenotype of the mutant mice. Instead, we found that the expression of key lipogenic enzyme genes was reduced in XBP1 deficient liver. Some but not all of these genes were directly induced by XBPls overexpression, indicating that XBP1 acts as a pro-lipogenic transcription factor.

While XBP1 plays an important role in hepatic lipid metabolism, several studies have also reported UPR activation in alcoholic and non-alcoholic fatty liver diseases (Ji, C. and Kaplowitz, N. (2003) Gastroenterology 124: 1488-1499; Ozcan, U. et al. (2004) Science

306:457-461 ; Puri, P. et al. (2008) Gastroenterology 134:568-576), suggesting the presence of ER stress in these metabolic abnormalities (Hotamisligil, G.S. (2008) Int J Obes (Lond) 32 Suppl 7:S52-54). Although it remains unclear how lipids activate the UPR (or cause ER stress), the idea that ER stress contributes to the pathogenesis of dyslipidemic metabolic diseases remains an interesting one. Along this line, it has been shown that XBP1 increases insulin sensitivity in type 2 diabetes animal models by resolving ER stress (Park, S.W. et al. (2010) Nat Med 16:429-437; Winnay, J.N. et al. (2010) Nat Med 16:438-445; Zhou, Y. et al. (2011) Nat Med 17:356-365).

Although XBP1 ablation did not activate PERK or ATF6, it strongly activated its upstream enzyme, EREloc, indicating feedback regulation of IREla activity by the abundance of its substrate XBPls (Lee, A.H. et al. (2008) Science 320: 1492-1496). Hyperactivated IREla possesses ribonuclease activity to induce the degradation of certain mRNAs by RIDD, such as those encoding cytochrome P450 enzymes that carry out detoxification of xenobiotics (Hur, K.Y. et al. (2012) J Exp Med.). Here, we demonstrate that RIDD also plays an important role in hepatic lipid metabolism. Suppression of RIDD by IREla siRNA markedly restored plasma TG and cholesterol levels in XBP1 knock out mice, along with the induction of lipid metabolism genes. Gene expression profiling revealed a group of lipid metabolism genes regulated by RIDD, which included Angptl3 and the carboxylesterase 1 (Cesl) gene family, whose suppression decreases plasma lipids (Koishi, R. et al. (2002) Nat Genet 30: 151-157; Romeo, S. et al. (2009) J Clin Invest 119:70-79; Musunuru, K. et al. (2010) N Engl J Med 363:2220-2227; Quiroga, A.D. and Lehner, R. (2011) Trends Endocrinol Metab 22:218-225). Contrary to the notion that XBP1 mitigates ER stress induced by fat accumulation, XBP1 ablation ameliorated hepatic steatosis in ob/ob mice, and liver damage induced by long-term high fat diet feeding, consistent with the distinct role of IREla/XBPl in lipid metabolism. We also demonstrate that siRNA-mediated silencing of XBP1 mRNA in the liver effectively lowers plasma lipids in mice, providing "proof of principle" that targeting XBP1 may be a viable approach to the treatment of dyslipidemias.

We have previously demonstrated that ablation of XBP1 in the liver resulted in a profound reduction of plasma TG and cholesterol unaccompanied by hepatic steatosis (Lee, A.H. et al. (2008) Science 320: 1492-1496). Here, we demonstrate that IREla hyperactivation and the consequent RIDD substantially contributes to the hypolipidemia phenotype of XBP1 deficient mice. XBP1 deficiency results in a feedback activation of IREla, inducing the degradation of mRNAs of a cohort of lipid metabolism genes, such as Dgat2, Acacb, Pcsk9, Angpt and Cesl, which regulate TG and cholesterol metabolism at multiple levels. Dgat2 and Acacb encode key lipogenic enzymes; Pcsk9 is involved in LDL clearance; AngptB suppresses LPL mediated TG clearance; Cesl possesses TG hydrolyzing activity and is implicated in fatty acids mobilization from lipid droplets to nascent VLDL. Hence, suppression of these genes by IREla is likely to contribute to the striking hypolipidemic phenotype mediated by liver-specific loss of XBP1. IRE la regulates hepatic lipid metabolism via two distinct mechanisms. First, IRE la promotes the degradation of mRNAs encoding lipid metabolism genes such as Dgat2, Acacb, Cesl, AngptU, and Pcsk9, which play important roles in de novo lipogenesis, hydrolysis of cholesterol ester and TG, and lipoprotein catabolism. Second, IREl a activates its downstream transcription factor XBPl, which can directly activate certain lipid metabolism genes. Hence, the combined effects of XBPl deficiency coupled with IRE la hyperactivation causes a profound reduction of plasma lipids in XBPl deficient mice. IREla deficient mice are defective in both XBPls-mediated activation of lipogenic genes and IRE la-mediated mRNA degradation. Under regular chow-fed conditions, IREla activity in liver is low, exerting minimal effect on the levels of RIDD substrates, but producing measurable amounts of XBPls protein, suggesting that the modest hypolipidemia of IREla deficient mice is caused mainly by the lack of XBPls.

In that case, how does XBPls deficiency cause hypolipidemia independently of RIDD in IREla deficient mice? We previously demonstrated that XBPl directly activated Dgat2 and Acacb lipogenic genes, which might contribute to the hypolipidemia observed in the absence of XBPl. However, these genes were not suppressed in IREla deficient liver, but were induced by IREla siRNA in XBPl deficient liver, indicating that these genes are primarily regulated by RIDD. In contrast, microarray analysis identified various lipid metabolism genes such as Fdps, Sqle, Cyp51, Pmvk, Fdftl, Hsdl7b7, Sc4mol, Mvk, and Idil that were suppressed in both XBPl and IREla deficient liver, hence representing direct XBPl targets. We speculate that XBPl directly regulates the expression of a subset of lipid metabolism genes including those listed above, contributing to the hypolipidemia phenotype observed in IREla deficient mice. It is also possible that the alteration of ER protein folding homeostasis caused by XBPl deficiency indirectly affected lipid metabolism. For example, VLDL assembly occurs in the ER lumen and involves the folding and lipidation of apoB-100. Although apoB turnover was not substantially altered in XBPl or IREla deficient primary hepatocytes (Lee, A.H. et al. (2008) Science 320: 1492-1496; Zhang, K. et al. (2011) EMBO J 30: 1357-1375), it is possible that XBPl deficiency impairs VLDL assembly and/or secretion in vivo. Supporting this scenario, IREla deficient mice were reported to be more susceptible to tunicamycin-induced hepatic steatosis (Zhang, K. et al. (2011) EMBO J 30: 1357-1375), which is likely to be caused by the inhibition of hepatic VLDL secretion (Liao, W. and Chan, L. (2001) Biochem J 353:493-501 ; Lee, A.H. and Glimcher, L.H. (2009) Cell Mol Life Sci 66:2835-2850).

It is notable that XBP1 ablation did not cause any deleterious effect on the viability of hepatocytes. On the contrary, XBP1 ablation in the liver protected mice from hepatic steatosis and liver damage caused by prolonged HFD feeding or by leptin deficiency (ob/ob mice). XBP1 ablation in the liver of ob/ob mice markedly ameliorated hepatic steatosis, paralleled by the suppression of the expression of lipogenic genes. Feeding a high-fat diet caused liver damage in WT mice, which was ameliorated in XBP1 deficient mice. Oxidative stress has been implicated in HFD-induced liver cell death (Anstee, Q.M. and Goldin, R.D. (2006) Int J Exp Pathol 87:1- 16). It remains to be determined if XBP1 deficiency altered the generation of ROS, or the sensitivity of hepatocytes to oxidative damage.

The ER is a crucial subcellular compartment for many physiological functions of the liver, including lipogenesis, lipoprotein production, protein secretion, and detoxification of xenobiotics. One might speculate that an increased burden of lipids, proteins, and xenobiotics on the ER would impair ER function, evoking a stress response. Supporting this hypothesis, several papers have reported UPR activation in various liver diseases as a marker of ER stress

(Hotamisligil, G.S. (2008) Int J Obes (Lond) 32 Suppl 7:S52-54; Ozcan, U. et al. (2004) Science 306:457-461 ; Puri, P. et al. (2008) Gastroenterology 134:568-576). Notably, it has been shown that TG accumulation in the liver in obesity causes ER stress, and the disruption of the UPR exacerbates the consequences of metabolic abnormalities such as nsulin resistance (Hotamisligil, G.S. (2008) Int J Obes (Lond) 32 Suppl 7:S52-54; Ozcan, U. et al. (2004) Science 306:457-461 ; Park, S.W. et al. (2010) Nat Med 16:429-437; Winnay, J.N. et al. (2010) Nat Med 16:438-445; Zhou, Y. et al. (2011) Nat Med 17:356-365). In stark contrast, we failed to detect any measurable UPR activation (IREla and PERK phosphorylation; ATF6oc processing; CHOP and BiP mRNA induction) in the liver of ob/ob mice, arguing against the involvement of the UPR in the metabolic abnormalities in this animal model. Our data are consistent with a recent report that demonstrated increased hepatic insulin sensitivity in XBP1 deficient mice, which was associated with decreased TG accumulation in the liver and plasma (Jurczak, M.J. et al. (201 1) J Biol Chem). It is important to note that our data does not disprove the involvement of ER dysfunction in metabolic stress, given that we actually did not measure "ER stress" in ob/ob mice, which is currently not feasible. Although UPR activation has been used as a marker of ER stress, the immediate upstream stimuli activating UPR in normal and diseased liver is not defined. It is possible that metabolic stresses compromise the ER function, without UPR activation.

XBP1 deficient liver displays a qualitatively and quantitatively normal lipid profile with no hepatic steatosis. This is in contrast to ApoB siRNA treated or Mttp mutant mice where lipid accumulates in the liver due to impaired VLDL assembly/secretion (Tadin-Strapps, M. et al. (2011) lipid research 52: 1084-1097; Raabe, M. et al. (1999) J Clin Invest 103: 1287-1298).

Dramatic reduction of plasma lipids coupled with preservation of the normal hepatic lipid profile in XBP1 deficient mice suggests that XBP1 is a promising target for drug development to treat dyslipidemias. We demonstrated that XBP1 siRNA lowered plasma lipid levels both in WT C57BL/6 and apoE deficient mice, providing proof of principle for targetiqg XBP1 to treat dyslipidemias. Alternatively, small molecule compounds modulating IREl a could be useful to . lower plasma lipid levels. Since genetic ablation of IREla in the liver decreased plasma lipid levels, compounds that inhibit IREla activity are expected to have similar lipid-lowering effects. IREla activators could also decrease plasma lipid levels by promoting the degradation of mRNAs of lipid metabolism genes. IREla activation would also induce XBPls, which can promote lipogenesis. The effects of IREla activation coupled with XBPls induction on lipid homeostasis remain to be further investigated.

ΙΠ. Screening Assays

Based on the biological roles of IRE1 and XBP1 in regulating lipid homeostasis through regulation of the expression and/or activity of various lipid metabolism and lipogenic genes, as demonstrated in the Examples, screening assays are provided that are useful in identifying compounds that can modulate lipid homeostasis, e.g., lipid metabolism.

Accordingly, in one aspect, the invention pertains to a method of identifying candidate modulators of lipid homeostasis comprising, a) determining the ability of a test compound to modulate regulated IRE 1 -dependent decay (RIDD) of a lipid metabolism gene, wherein mRNA of the lipid metabolism gene is degraded by IREl ; and b) selecting said test compound as a compound of interest when said compound modulates RIDD of the lipid metabolism gene, to thereby identify candidate modulators of lipid homeostasis. In another aspect, the invention pertains to a method of identifying candidate modulators of lipid homeostasis comprising, a) determining the ability of a test compound to modulate XBP1 -mediated transcription of a lipogenic target gene; and selecting said test compound as a compound of interest when said compound modulates XBP1 -mediated transcription of the lipogenic target gene, to thereby identify candidate modulators of lipid homeostasis.

In one embodiment, modulation of lipid homeostasis comprises modulation of lipid metabolism. In one embodiment, modulation of lipid metabolism comprises induction of lipid clearance from blood (e.g., clearance of cholesterol, triglyceride and/or other lipids from the blood). In one embodiment, modulation of lipid homeostasis comprises modulation of lipogenesis. In one embodiment, modulation of lipogenesis comprises induction of lipogenesis. In another embodiment, modulation of lipogenesis comprises inhibition of lipogenesis.

In one embodiment, the compound identified in a method presented herein inhibits RIDD of mRNAs of lipid metabolism genes. In another embodiment, the compound identified in a method presented herein stimulates RIDD of mRNAs of lipid metabolism genes. In one embodiment, the compound identified in a method presented herein stimulates XBP1 -mediated transcription of the lipogenic target gene. In another embodiment, the compound identified in a method presented herein inhibits XBP1 -mediated transcription of the lipogenic target gene.

In one embodiment, the lipid metabolism gene or lipogenic target gene is a hepatic gene.

As described in Example 4, microarray analysis has led to the identification of a panel of hepatic genes, e.g., lipogenic genes and/or other genes directly or indirectly involved in lipid metabolism, whose transcription is directly regulated by XBP1, as set forth in Figure 3B.

Accordingly, in one embodiment, the lipogenic target gene regulated by XBP1 and used in a method presented herein is selected from the group consisting of SdfZll, Cflar, Dnajc3, Dnajb9, Gale, Nans, Edeml, Dnaljbl 1, Tmem39a, Erolb, Gmppa, Hyoul, Usol, Sec24d, Slc35bl, Uggtl, Ssr3, Ubxn4, Txndc5, Stt3a, Ssrl, Serpl, Ddost, Sec61al, S rf4, 0610007 LOlRik, Rrbpl, Ostc, Odd, Cnpy3, Cdk5rap3, Fdps, Spcs3, P4hb, Spcs2, 2810482107Rik, Sec61b, Lman2, Krtcap2, Secllc, Sec61g, Ssr4, Sec63, H13, D17Wsul04e, Chidl, Sndl, Ikbke, Traml, Cbaral, Slc33al, Pexllc, Mlec, Sqle, Cyp51, Tnfaip8ll, Pmvk, Fdtfl, Hsdl7b7, Anubll, Lifr, Sc4mol, Mvk, and Idil. This subset of genes represents the 64 genes set forth in Figure 3B.

In another embodiment, the lipogenic target gene regulated by XBP1 and used in a method presented herein is selected from the group consisting of Dnajc3, Edeml, Dnaljbl 1, Erolb, Hyoul, Sec24d, Uggtl, Ssr3, Txndc5, Stt3a, Ssrl, Ddost, Sec61al, Rrbpl, Spcs3, P4hb, Spcs2, Sec61b, Lman.2, Secllc, Sec61g, Ssr4, Sec63 and Traml. This subset of genes represents the 24 genes involved in protein processing in the ER highlighted in Figure 3B.

In yet another embodiment, the lipogenic target gene regulated by XBP1 and used in a method presented herein is selected from the group consisting of Fdps, Sqle, CypSl, Pmvk, Fdtfl, Hsdl7b7, Sc4mol, Mvk and Idil. This subset of genes represents the 9 genes involved in lipid metabolism highlighted in Figure 3B.

Also as described in Example 4, microarray analysis has led to the identification of a panel of hepatic genes, e.g., genes directly or indirectly involved in lipid metabolism, whose expression and/or activity is directly regulated by RIDD mediated by IREl, as set forth in Figure 3C. Accordingly, in one embodiment, the lipid metabolism gene regulated by IREl and used in a method presented herein is selected from the group consisting of Paqr7, Ptprf, Ces3,

AU018778, Cesl, Ceslc, Cesld, Cesle, Ceslf, Ceslg, Lrpl, Bloclsl, Dhcr7, Limd2, Idil, 221041 lKHRik, Es22, Apon, Glrx, Fads2, He, Oaf, Mvk, Gjbl, Sc4mol, Plala, F7, Cpn2, Pxmp2, Abca3, Gm2a, Kcnn2, Nsdhl, Stim2, Dnase2b, Vnn3, Hsdl7b7, Hgsnat, Coll8al, Aldoc, Fadsl, Fnl, Smarcc2, Dera, Spp2, Angptl3, Atrnll, Tppl, Faml08a, Tmprss6, Sidt2, Esl, Dbt, Ropnll, Tex264, Tapbp, Selk, Btbdl, Cyp2c68, Ndufa9, Pmvk, Blvra, Igftr, Proc, Abca.2, Taldol, Bola3, Cypla2, Retsat, Lamp], Rnfl30, Tex2, Afm, Jkamp, 0610007 C21Rik, Plbd2, Cyp51, Tm7sf2, Clu, Hgfac, 3110073H01Rik, Furin, C920025E04Rik, Serpingl, Lipc, Ttcl3, Coll8al, Slc27a5, Pcsk6, Tmem208, Pm20dl, Sqle, Man2a2, Pplrl5b, Azgpl, Ces6, Pla2gl2b, llvbl, Adam9, Mlec, Pdapl, F2, Commd9, Itih4, Dgat2, Cyb561d2, Slc35f5, Algl, Gnptg, Rarresl, Yifla, Slc25al7, Mxd4, Fdps, Papola, Erp44 and Rpnl.

In another embodiment, the lipid metabolism gene regulated by XBP1 and used in a method presented herein is selected from the group consisting of Ces3, AU018778, Cesl, Lrpl, Dhcr7, Idil, Es22, Apon, Fads2, Mvk, Gjbl, Sc4mol, Plala, Abca3, Gm2a, Nsdhl, Hsdl7b7, Aldoc, Fadsl, Dera, AngptU, Esl, Pmvk, Abca.2, Taldol, Retsat, Plbd2, CypSl, Tm7sf2, Lipc, Slc27a5, Sqle, Pla2gl2b, Dgat2 and Fdps. This subset of genes represents the 35 genes involved in lipid metabolism highlighted in Figure 3C.

In yet another embodiment, the lipid metabolism gene regulated by XBP1 and used in a method presented herein is selected from the group consisting of Ceslc, Cesld, Cesle, Ceslf, Ceslg and Angplt3. This subset of genes represents a particular group of REDD substrates described further in Examples 5 and 6.

In one embodiment, the screening methods of the invention can be carried out by testing a single test compound in the method. Alternatively, the test compound is comprised in a combination of test compounds, such as a library of test compounds. In one embodiment, the test compound is a small molecule. In another embodiment, the test compounds is comprised in a library of small molecules. Suitable test compounds for use in the methods presented herein are described further in subsection IV below.

In one embodiment of the invention, the cell based and/or cell free assays are performed in a high-throughput manner. In one embodiment, the assays are performed using a 96-well format. In another embodiment, the assays of the invention are performed using a 192-well format. In another embodiment, the assays of the invention are performed using a 384-well format. In one embodiment, the assays of the invention are semi-automated, e.g., a portion of the assay is performed in an automated manner, e.g., the addition of various reagents. In another embodiment, the assays of the invention are fully automated, e.g., the addition of all reagents to the assay and the capture of assay results are automated.

The assays of the invention generally involve contacting an assay composition with a test compound or a compound of interest or a library of compounds for a predetermined amount of time or at a predetermined time of growth (either in vitro or in vivo) and assaying for the effect of the compound on a particular read-out. In one embodiment, an assay composition is contacted with a compound of interest or a library of compounds for the duration of the assay. In another embodiment, an assay composition is contacted with a compound of interest or a library of compounds for a period of time less than the entire assay time period. For example, cells may be cultured for a period of days or weeks and may be contacted with a compound following, for example, 14 days in culture. In one embodiment, cells are contacted with a compound of interest for 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, or 21 days. In one embodiment, assay compositions of the invention are contacted with a compound for a predetermined time period, the compound is removed, and the assay composition is maintained in the absence of the compound for a predetermined period prior to assaying for a particular read-out. The compounds of the invention may be assayed at concentrations suitable to the assay and readily determined by one of skill in the art. For example in one embodiment, assay compositions are contacted with millimolar concentrations of compounds. In another embodiment, assay compositions are contacted with micromolar concentrations of compounds. In another embodiment, assay compositions are contacted with nanomolar concentrations of compounds.

To determine the ability of a test compound to modulate RIDD of a lipid metabolism gene, typically the test compound is contacted with an indicator composition that comprises IREl, or a biologically active portion thereof, and at least one lipid metabolism gene, or mRNA thereof. The indicator composition can be, for example, a cell that comprises IREl and the lipid metabolism gene(s)/mRNA(s), a cell extract that comprises IREl and the lipid metabolism gene(s)/mRNA(s), or a cell-free composition that comprises IREl and the lipid metabolism gene(s)/mRNA(s). Preferably, the IREl is IREloc, such as recombinant IREl a protein. The ability of a test compound to modulate RIDD of a lipid metabolism gene can be measured by determining the effect of the test compound on the degree of RIDD (i.e., mRNA degradation), mediated by IREl, of mRNA of the lipid metabolism gene. This can be determined by comparing the amount of RIDD of mRNA of the lipid metabolism gene in the presence of the test compound and in the absence of the test compound. The level of mRNA of the lipid metabolism gene can be directly measured or, alternatively, the amount of protein product of the lipid metabolism gene can be determined as a measure of the level of mRNA of the lipid metabolism gene.

Thus, in one embodiment, determining the ability of a test compound to modulate RIDD of the lipid metabolism gene comprises contacting an indicator composition with the test compound and determining a level of lipid metabolism gene mRNA in the presence of the test compound. In another embodiment, determining the ability of a test compound to modulate RIDD of the lipid metabolism gene comprises contacting an indicator composition with the test compound and determining a level of lipid metabolism gene-encoded protein in the presence of the test compound. In another embodiment, an in vitro mRNA cleavage assay, such as that described in Example 6, is used to determine the ability of the test compound to modulate RIDD of the lipid metabolism gene mediated by IREl . To determine the ability of a test compound to modulate XBP1 -mediated transcription of the lipogenic target gene, typically the test compound is contacted with an indicator composition that comprises XBP1, or a biologically active portion thereof, and at least one lipogenic target gene, or a reporter gene thereof. The indicator composition can be, for example, a cell that comprises XBP1 and the lipogenic target gene(s)/reporter(s), a cell extract that comprises XBP1 and the lipogenic target gene(s)/reporter(s), or a cell-free composition that comprises XBP1 and the lipogenic target gene(s)/reporter(s). Preferably, the XBP1 is spliced XBP1, such as recombinant spliced XBP1 protein. The ability of a test compound to modulate XBPl-mediated transcription of the lipogenic target gene can be measured by determining the effect of the test compound on the degree of lipogenic target gene transcription (i.e., as measured by mRNA or gene product levels), mediated by XBP1. This can be determined by comparing the amount of XBPl-mediated transcription of the lipogenic target gene in the presence of the test compound and in the absence of the test compound. The level of mRNA of the lipogenic target gene can be directly measured or, alternatively, the amount of protein product of the lipogenic target gene can be determined as a measure of the level of mRNA of the lipogenic target gene. Additionally or alternatively, a reporter gene construct can be used in which the transcriptional regulatory region of the lipogenic target gene is operatively linked to a reporter gene whose gene product is readily measurable (such as, for example, CAT or luciferase). Cell-based and cell-free assay systems for measuring the transcriptional activity of XBP1 for a target gene, including systems using a reporter gene, are described further in US Patent Publication No. 20040170622, the entire contents of which is expressly incorporated herein by reference. The ordinarily skilled artisan can adapt such cell-based and cell-free XBP1 assay systems for use in the methods described herein using lipogenic target genes as described herein.

Thus, in one embodiment, determining the ability of a test compound to modulate XBPl- mediated transcription of the lipogenic target gene comprises contacting an indicator

composition with the test compound and determining a level of lipogenic target gene mRNA in the presence of the test compound. In another embodiment, determining the ability of a test compound to modulate XBPl-mediated transcription of the lipogenic target gene comprises contacting an indicator composition with the test compound and determining a level of a reporter of lipogenic target gene transcription in the presence of the test compound. In yet another embodiment, the test compound is further determined to modulate XBPl -mediated transcription of the lipogenic target gene independent of unfolded protein response (UPR) activation.

In another aspect, the invention pertains to methods of validating a compound as a compound useful in modulating lipid homeostasis in vivo using one or more animal model systems. For example, in one embodiment, the invention provides a method for validating a compound as a compound useful in modulating lipid homeostasis in vivo comprising, selecting a candidate compound identified according to a screening method as disclosed herein, and testing said candidate compound for modulation of lipid metabolism in an animal model of

dyslipidemia. Such animal models of dyslipidemia are well established in the art, including but not limited to, leptin deficient ob/ob mice, mice fed a high-fat diet and ApoE deficient mice. In the validation method, the animal can be tested, for example, for modulation of one or more of plasma cholesterol levels, plasma and/or hepatic triglyceride levels, plasma and/or hepatic lipid levels, steatosis, and lipoprotein metabolism. Methods for testing these parameters in animal models are well established in the art and are also described further in the Examples.

In another embodiment, the invention provides a method for validating a compound as a compound useful in modulating lipid homeostasis in vivo comprising, selecting a candidate compound identified according to a screening method as presented herein, and testing said candidate compound for modulation of lipid metabolism gene activity in an animal model deficient in XBPl and/or IRE la. In the validation method, the animal can be tested, for example, for modulation of one or more of plasma cholesterol levels, plasma and/or hepatic triglyceride levels, plasma and/or hepatic lipid levels, steatosis, and lipoprotein metabolism. Methods for testing these parameters in animal models are well established in the art and are also described further in the Examples.

IV. Test Compounds

A variety of test compounds can be evaluated using the screening assays described herein. The term "test compound" includes any reagent or test agent which is employed in the assays of the invention and assayed for its ability to modulate RIDD of a lipid metabolism gene mediated by IRE1 and/or its ability to modulate transcriptional activation of a lipogenic target gene mediated by XBPl . More than one compound, e.g., a plurality of compounds, can be tested at the same time for their ability to modulate RIDD of a lipid metabolism gene mediated by IREl and/or their ability to modulate transcriptional activation of a lipogenic target gene. The term "screening assay" preferably refers to assays which test the ability of a plurality of compounds to influence the readout of choice rather than to tests which test the ability of one compound to influence a readout. Preferably, the subject assays identify compounds not previously known to have the effect that is being screened for. In one embodiment, high throughput screening can be used to assay for the activity of a compound.

In certain embodiments, the compounds to be tested can be derived from libraries (i.e., are members of a library of compounds). While the use of libraries of peptides is well established in the art, new techniques have been developed which have allowed the production of mixtures of other compounds, such as benzodiazepines (Bunin et al. (1992). J. Am. Chem. Soc.

1 14: 10987; DeWitt et al. (1993). Proc. Natl. Acad. Sci. USA 90:6909) peptoids (Zuckermann. (1994). J. Med. Chem. 37:2678) oligocarbamates (Cho et al. (1993). Science. 261 : 1303-), and hydantoins (DeWitt et al. supra). An approach for the synthesis of molecular libraries of small organic molecules with a diversity of 104-105 has been described (Carell et al. (1994). Angew. Chem. Int. Ed. Engl. 33:2059-; Carell et al. (1994) Angew. Chem. Int. Ed. Engl. 33:2061-).

The compounds of the present invention can be obtained using any of the numerous approaches in combinatorial library methods known in the art, including: biological libraries; spatially addressable parallel solid phase or solution phase libraries, synthetic library methods requiring deconvolution, the ^v one-bead one-compound library method, and synthetic library methods using affinity chromatography selection. The biological library approach is limited to peptide libraries, while the other four approaches are applicable to peptide, non-peptide oligomer or small molecule libraries of compounds (Lam, K. S. (1997) Anticancer Drug Des. 12: 145). Other exemplary methods for the synthesis of molecular libraries can be found in the art, for example in: Erb et al. (1994). Proc. Natl. Acad. Sci. USA 91 : 11422-; Horwell et al. (1996) Immunopharmacology 33:68-; and in Gallop et al. (1994); J. Med. Chem. 37: 1233.

Libraries of compounds can be presented in solution (e.g., Houghten (1992)

Biotechniques 13:412-421), or on beads (Lam (1991) Nature 354:82-84), chips (Fodor (1993) Nature 364:555-556), bacteria (Ladner U.S. Pat. No. 5,223,409), spores (Ladner U.S. Pat. No. '409), plasmids (Cull et al. (1992) Proc Natl Acad Sci USA 89: 1865-1869) or on phage (Scott and Smith (1990) Science 249:386-390); (Devlin (1990) Science 249:404-406); (Cwirla et al. (1990) Proc. Natl. Acad. Sci. 87:6378-6382); (Felici (1991) J. Mol. Biol. 222:301-310); In still another embodiment, the combinatorial polypeptides are produced from a cDNA library.

Exemplary compounds which can be screened for activity include, but are not limited to, peptides, nucleic acids, carbohydrates, small organic molecules, and natural product extract libraries.

Candidate/test compounds include, for example, 1) peptides such as soluble peptides, including Ig-tailed fusion peptides and members of random peptide libraries (see, e.g., Lam, K. S. et al. (1991) Nature 354:82-84; Houghten, R. et al. (1991) Nature 354:84-86) and

combinatorial chemistry-derived molecular libraries made of D- and/or L-configuration amino acids; 2) phosphopeptides (e.g., members of random and partially degenerate, directed phosphopeptide libraries, see, e.g., Songyang, Z. et al. (1993) Cell 72:767-778); 3) antibodies (e.g., polyclonal, monoclonal, humanized, anti-idiotypic, chimeric, and single chain antibodies as well as Fab, F(ab').sub.2, Fab expression library fragments, and epitope-binding fragments of antibodies); 4) small organic and inorganic molecules (e.g., molecules obtained from

combinatorial and natural product libraries); 5) enzymes (e.g., endoribonucleases, hydrolases, nucleases, proteases, synthatases, isomerases, polymerases, kinases, phosphatases, oxido- reductases and ATPases), and 6) mutant forms of KRC (e.g., dominant negative mutant forms of the molecule).

The test compounds of the present invention can be obtained using any of the numerous approaches in combinatorial library methods known in the art, including: biological libraries; spatially addressable parallel solid phase or solution phase libraries; synthetic library methods requiring deconvolution; the 'one-bead one-compound' library method; and synthetic library methods using affinity chromatography selection. The biological library approach is limited to peptide libraries, while the other four approaches are applicable to peptide, non-peptide oligomer or small molecule libraries of compounds (Lam, K. S. ( 1997) Anticancer Drug Des. 12: 145).

Examples of methods for the synthesis of molecular libraries can be found in the art, for example in: DeWitt et al. (1993) Proc. Natl. Acad. Sci. U.S.A. 90:6909; Erb et al. (1994) Proc. Natl. Acad. Sci. USA 91 : 1 1422; Zuckermann et al. (1994) J. Med. Chem. 37:2678; Cho et al. (1993) Science 261 : 1303; Carrell et al. (1994) Angew. Chem. Int. Ed. Engl. 33:2059; Carell et al. (1994) Angew. Chem. Int. Ed. Engl. 33:2061 ; and Gallop et al. (1994) J. Med. Chem.

37: 1233. Libraries of compounds can be presented in solution (e.g., Houghten (1992) Biotechniques 13:412-421), or on beads (Lam (1991) Nature 354:82-84), chips (Fodor (1993) Nature 364:555-556), bacteria (Ladner U.S. Pat. No. 5,223,409), spores (Ladner U.S. Pat. No. '409), plasmids (Cull et al. (1992) Proc Natl Acad Sci USA 89: 1865-1869) or phage (Scott and Smith (1990) Science 249:386-390; Devlin (1990) Science 249:404-406; Cwirla et al. (1990) Proc. Natl. Acad. Sci. 87:6378-6382; Felici (1991) J. Mol. Biol. 222:301-310; Ladner supra.).

Compounds identified in the subject screening assays can be used in methods of modulating lipid homeostasis, for example in therapy for disorders associated with dyslipidemia, such as atherosclerosis, obesity, metabolic syndrome and type 2 diabetes. It will be understood that it may be desirable to formulate such compound(s) as pharmaceutical compositions prior to contacting them with cells.

Once a test compound is identified that directly or indirectly modulates lipid homeostasis by one of the variety of methods described hereinbefore, the selected test compound (or

"compound of interest") can then be further evaluated for its effect on cells, for example by contacting the compound of interest with cells either in vivo (e.g., by administering the compound of interest to a subject) or ex vivo (e.g., by isolating cells from the subject and contacting the isolated cells with the compound of interest or, alternatively, by contacting the compound of interest with a cell line) and determining the effect of the compound of interest on the cells, as compared to an appropriate control (such as untreated cells or cells treated with a control compound, or carrier, that does not modulate the biological response).

V. Kits of the Invention

Another aspect of the invention pertains to kits for carrying out the screening assays of the invention. For example, in one embodiment, a kit for carrying out a screening assay of the invention can include an indicator composition comprising IREl (e.g., recombinant IREl a or a cell extract comprising IREl), means for measuring a readout (e.g., mRNA levels of one or more lipid metabolism genes regulated by RIDD) and instructions for using the kit to identify modulators of lipid homeostasis. Additionally, the kit can include mRNA or means for preparing mRNA of one or more lipid metabolism genes regulated by RIDD. In another embodiment, a kit for carrying out a screening assay of the invention can include an indicator composition comprising XBPl (e.g., recombinant XBPl or a cell extract comprising XBPl), means for measuring a readout (e.g., mRNA levels of one or more lipogenic target genes regulated byXBPl or a reporter gene readout of one or more lipogenic target genes regulated by XBPl) and instructions for using the kit to identify modulators of lipid homeostasis. Additionally, the kit can include one or more lipogenic target genes, or one or more reporter genes of a lipogenic target gene(s) regulated by XBPl .

The practice of the present invention will employ, unless otherwise indicated, conventional techniques of cell biology, cell culture, molecular biology, transgenic biology, microbiology, recombinant DNA, and immunology, which are within the skill of the art. Such techniques are explained fully in the literature. See, for example, Molecular Cloning A

Laboratory Manual, 2nd Ed., ed. by Sambrook, Fritsch and Maniatis (Cold Spring Harbor Laboratory Press: 1989); DNA Cloning, Volumes I and II (D. N. Glover ed., 1985);

Oligonucleotide Synthesis (M. J. Gait ed., 1984); Mullis et al. U.S. Pat. No. 4,683,195; Nucleic Acid Hybridization (B. D. Hames & S. J. Higgins eds. 1984); Transcription And Translation (B. D. Hames & S. J. Higgins eds. 1984); Culture Of Animal Cells (R. I. Freshney, Alan R. Liss, Inc., 1987); Immobilized Cells And Enzymes (IRL Press, 1986); B. Perbal, A Practical Guide To Molecular Cloning (1984); the treatise, Methods In Enzymology (Academic Press, Inc., N.Y.); Gene Transfer Vectors For Mammalian Cells (J. H. Miller and M. P. Calos eds., 1987, Cold Spring Harbor Laboratory); Methods In Enzymology, Vols. 154 and 155 (Wu et al. eds.), Immunochemical Methods In Cell And Molecular Biology (Mayer and Walker, eds., Academic Press, London, 1987); Handbook Of Experimental Immunology, Volumes I- IV (D. M. Weir and C. C. Blackwell, eds., 1986); Manipulating the Mouse Embryo, (Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y., 1986).

This invention is further illustrated by the following examples which should not be construed as limiting. The contents of all references, patents, and published patent applications cited throughout this application, as well as the figures and the sequence listing, are hereby incorporated by reference. EXAMPLES

Example 1: Materials and Methodologies Used in the Examples

The following materials and methodologies were used in Examples 2-9.

Mice. Xbplⁿ° mice were crossed with interferon inducible B6.Cg-Tg(Mxl-cre)lCgn/J or C57BL/6-Tg(Alb-cre)21Mgn/J strains of mice (Jackson Laboratory) that produce ere recombinase under the control of the mouse albumin promoter and efficiently delete the floxed gene in the liver as previously described (Lee, A.H. et al. (2008) Science 320: 1492-1496) Ernlⁿ° mice have been previously described (Iwawaki, T. et al. (2009) Proc Natl Acad Sci U S A 106: 16657-16662), and crossed with Mxl-cre mice to inducibly delete IRE la in the liver. Mice were housed in a specific pathogen free facility at the Harvard School of Public Health and had free access to water and standard chow diet (PicoLab Rodent diet 20, #5058, Lab diet), which consisted of 12% fat, 23.5% protein, and 64.5% carbohydrate, or a high-fat diet (45% fat, 18.6% protein, and 36.4% carbohydrate; TestDiet, #58G8). Mice heterozygous for a leptin mutation (B6.V-Lep^ob/J) were obtained from The Jackson Laboratory, and intercrossed to produce homozygous ob/ob mice, or bred onto Xbpl mice to generate Xbpl ;ob/ob mice. Xbpl^{f f};Mxl -ere mice were crossed to ApoE^{_ "} mice (Jackson laboratory, B6.129P2- Apoe^tmlUnc/J), and injected with poly (I:C) to generate XBP1 and ApoE double deficient mice. Animal studies and experiments were approved and carried out according to the guidelines of the Animal Care and Use Committee of Harvard University.

In vivo siRNA delivery. XBP1 siRNA (sense, 5'-CACCCUGAAUUCAUUGUCU-3' (SEQ ID NO: 1); antisense, 5 ' - AG AC A AUG A AUUC AGGGUG-3 ' ) (SEQ ID NO: 2) was formulated with lipidoid 98Ni₂-5 as described previously (Frank-Kamenetsky, M. et al. (2008) Proc Natl Acad Sci U S A 105:11915-1 1920). XBP1 and luciferase siRNAs (Frank-Kamenetsky, M. et al. (2008) Proc Natl Acad Sci U S A 105:11915-11920) were diluted in PBS and injected into mice via the tail vein at 5 μΐ/g body weight. IREla and control luciferase siRNAs formulated with lipidoid C12-200 were injected via tail vein as described previously (Love, K.T. et al. (2010) Proc Natl Acad Sci U S A 107: 1864-1869; Hur, K.Y. et al. (2012) J Exp Med.). Blood chemistry and measurement of liver TG contents. Plasma TG and cholesterol levels were measured by using commercial Triglyceride Reagent (Sigma) and the Amplex® Red Cholesterol Assay Kit (Invitrogen), respectively, following manufacturer's instructions. Plasma Angplt3 levels were measured using a commercial ELISA kit (R&D Systems). Fast performance liquid chromatography (FPLC) analysis of plasma samples was performed as described previously (Lee, A.H. et al. (2011) Proc Natl Acad Sci U S A 108:8885-8890). Serum alanine

aminotransferase (ALT) activity was measured using a commercial reagent (Bioquant). Lipids were extracted from liver tissue by Folch's method, and subjected to TG assay.

RNA isolation, qRT-PCR and microarray analysis. Total RNA was extracted from liver using Qiazol lysis reagent (Qiagen), and reverse transcribed into cDNA using High Capacity cDNA Reverse Transcription Kit (Applied Biosystems). qRT-PCR was performed as described previously (Lee, A.H. et al. (2008) Science 320: 1492-1496). RNA samples from individual mice were further purified using RNeasy MinElute Cleanup Kit (Qiagen), and then used for the production of biotin-labeled cRNA followed by hybridization with HT MG-430 PM Array Plate (Affymetrix). 237 XBP1 -dependent genes were identified by comparing WT and Xbpl A liver samples with minimum fold-change criterion of 1.6 and P-value < 0.005. Relaxed criteria (P- value < 0.05) were used for the comparison between WT and IrelA; untreated and tunicamycin treated; luciferase and IRE la siRNA.

Cell culture and Transfection. HEK293T cells were cultured in DMEM supplemented with 10% fetal bovine serum. Transient transfection was performed using Lipofectamine 2000 (Invitrogen), as described previously (Lee, A.H. et al. (2011) Proc Natl Acad Sci U S A 108:8885-8890).

Western blot. Liver tissues were homogenized in RIPA buffer (50 mM Tris-Cl pH 8.0, 150 mM NaCl, 1 % NP40, 0.5% deoxycholate, 0.1 % SDS, 50 mM NaF) supplemented with protease inhibitor tablet (Roche). Homogenates were centrifuged at 12,000g for 10 min at 4°C, and the supernatants were collected. Liver nuclear extracts were prepared as described previously (Lee, A.H. et al. (2008) Science 320: 1492-1496). Liver lysates were used for western blotting as described previously (Hur, K.Y. et al. (2012) J Exp Med.). In vitro IRE Ice-mediated mRNA cleavage assays Cleavage of in vitro transcribed Angptl3 mRNA by recombinant IREla was tested as described previously (Lee, A.H. et al. (2011) Proc Natl Acad Sci U S A 108:8885-8890). Briefly, pCMV-SPORT6-Angptl3 (BCO 19491, Open Biosystems) plasmid was linearized by Bglll digestion and incubated with SP6 polymerase to produce AngptO RNA. Angptl3mut construct was generated using QuikChange Site-Directed Mutagenesis Kit (Stratagene), based on the predicted mRNA secondary structure

(http://ma.tbi. univie.ac.at/cgi-bin/RNAfold.cgi). In vitro transcribed RNAs were incubated with the recombinant protein including the cytosolic domain of human IREla, resolved on a 1.2% denaturing agarose gel and visualized by ethidium bromide staining.

Quantification of atherosclerosis lesions. The aortas were harvested and fixed in 10% buffered formalin solution overnight. Atherosclerotic lesions were analyzed in the aortic arch, and descending aorta by oil red O staining as previously described (Gotsman, I. et al. (2006) Circulation 114:2047-2055; Gotsman, I. et al. (2007) J Clin Invest 117:2974-2982).

Glucose tolerance test. Mice were fasted for 16 h with free access to water, and then i.p. injected with glucose at 1.5 g Kg of body weight. Blood glucose levels were measured along the time course using an Ascensia Breeze glucometer (Bayer).

Insulin tolerance test. Mice were fasted for 6 h with free access to water, and then i.p. injected with insulin (Eli Lilly) at 0.75 units/kg body weight. Blood glucose levels were measured as above.

Histological analysis . Liver tissues were fixed in 10% neutral buffered formalin solution, processed into paraffin blocks, sectioned at 5 μπι, and stained with hematoxylin and eosin (H&E). Example 2: Partial restoration of plasma lipid levels by suppression of IREla in XbplA mice

We previously demonstrated that an inducible deletion of Xbpl in the liver decreased plasma cholesterol and TG levels in mice, while preserving normal hepatic lipid contents and composition (Lee, A.H. et al. (2008) Science 320: 1492-1496). As XBPl deficiency also resulted in constitutive activation of IREla, which could potentially mediate XBPl -independent functions, we sought to determine the contribution of hyperactivated IREla to the hypolipidemic phenotype of XBPl deficient mice. Xbpl^f/f;Alb-cre (Xbpl^LKO) mice which expressed Cre recombinase in postnatal liver under the control of the mouse albumin enhancer/promoter displayed markedly lower plasma TG and cholesterol levels compared to their Xbpl^{f f} littermate controls (Figure 1A and IB). Plasma lipid levels were not altered in heterozygous Xbpl^f/+;Alb- cre mice. Lipogenic enzyme genes such as Dgat2, Acacb, and Scdl, were suppressed in the liver of Xbpl^LKO mice (Figure 8), similar to what we observed in the inducible XBPl deficient mouse strain (Lee, A.H. et al. (2008) Science 320: 1492-1496).

To investigate the function of hyperactivated IREla in hepatic lipid metabolism, we silenced IREla expression in the liver of Xbpl^LKO mice using siRNA, as described elsewhere (Hur, K.Y. et al. (2012) J Exp Med.). IREla siRNA markedly increased plasma TG and cholesterol levels in Xbpl^LKO mice, indicating that hyperactivated IREla contributed to the reduction of plasma lipids in the setting of XBPl deficiency (Figure 1C and ID). Injection of IREla siRNA into Xbpl^LKO mice significantly increased Dgat2 and Acacb mRNAs, which play crucial roles in fatty acid metabolism and were decreased in Xbpl deficient liver (Figure IE and IF). Pcsk9, which promotes the degradation of LDL receptor (Horton, J.D. et al. (2007) Trends Biochem Sci 32:71-77), was also induced by IREla siRNA, suggesting that these mRNAs are degraded after cleavage by hyperactivated IREla in XBPl deficient liver.

Example 3: Modest hypolipidemia in IREla deficient mice

Since IREla silencing restored the plasma lipid concentrations to near normal levels in Xbpl knock-out mice, we asked whether XBPl regulates plasma lipid levels solely by modulating IREla activity or has a separate function in hepatic lipid metabolism. We sought to determine the relative contributions of XBPl deficiency and IREla hyperactivati on to the dyslipidemia phenotype of Xbpl^LK0 mice. It is notable that IREla siRNA only partially restored plasma lipid levels in Xbpl^LK0 mice (Figure 1), suggesting that XBPl itself may also play a role in lipid metabolism independent of RIDD. XBPl deficient mice with normal IREla activity would reveal the direct role of XBPl in hepatic lipid metabolism, but this was not a feasible approach. As an alternative, we examined IREla deficient mice that do not produce the active XBPls protein. IREla deficient mice were generated by crossing Ernlⁿ°^x mice with Mxl-cre mice, which allowed inducible deletion of IREla in the liver upon poly (I:C) administration. Without poly (I:C) administration, plasma TG and cholesterol levels were comparable between Ernl^{f f} and Ernl^f/f;Mxl-cre mice (Figure 2A and 2B). Notably, poly (I:C) administration significantly decreased both plasma TG and cholesterol levels in Ernl^f/f;Mxl-cre mice (ErnlA), but not in the control Ernl^{flo /flox} mice, indicating that XBPls is required to maintain optimal plasma lipid levels. Moreover, IREla and XBPl double-deficient mice exhibited modestly decreased plasma TG and cholesterol levels compared with WT littermates (Figure 2C and 2D), consistent with the results from IREla siRNA-injected Xbpl^LKO mice (Figure 1C and ID). These data suggest that the lack of XBPls and the hyperactivation of IREla both contribute to the dramatic decrease of plasma lipids observed in the setting of XBPl deficiency and do so via separate mechanisms.

Example 4; Identification of direct XBPl targets and RIDD substrates in liver

We next sought to identify lipid metabolism genes that are directly regulated by XBPl and those regulated by RIDD. We reasoned that genes down-regulated in XBPl deficient liver would fall into two groups: RIDD substrates and direct XBPl targets. To identify RIDD substrate mRNAs and direct XBPl targets in the liver, we performed a comprehensive microarray analysis of three groups of RNA samples: WT and XBPl deficient mice, WT and IREla deficient mice untreated or injected with tunicamycin, and XBPl deficient mice injected with luciferase or IREla siRNA. We identified 237 genes whose mRNA levels were significantly lower in XBPl knockout than in the WT littermate controls. Among these XBP1- dependent genes, 64 genes were also down regulated in IREla deficient liver, hence representing likely direct XBP1 target genes (Figure 3B). A majority of these genes were induced by tunicamycin treatment in WT, but not in ErnlA mice, indicating that XBPls directly activates their transcription (Figure 3A and 3B). Gene ontology analysis revealed that genes involved in "Protein processing in endoplasmic reticulum (KEGG mmu04141)" were highly enriched in this group, consistent with the critical function of XBP1 in protein secretory pathways (Shaffer, A.L. et al. (2004) Immunity 21:81-93; Lee, A.H. et al. (2003) Mol Cell Biol 23:7448-7459). Notably, the expression of a number of sterol biosynthesis-related genes such as Fdps, Sqle, CypSl, Pmvk, Fdfil, Hsdl7b7, Sc4mol, Mvk, and Idil was also reduced in both XBP1 and IREla deficient mice, and hence might contribute to the decreased lipid levels in these mutant mice.

Among the 237 genes that were suppressed in XBP1 deficient liver, 112 genes were induced by IREla siRNA treatment in XBP1 deficient mice- these genes represent REDD substrates (Figure 3C). Notably, several genes involved in lipid metabolism in addition to Dgat2, Acacb and Pcsk9 were identified as RIDD substrates, likely contributing to the suppression of plasma lipid levels by activated ERE la. Acacb and Pcsk9 were not included in the microarray analysis because of weak signals. Tunicamycin activates ERE la and hence would decrease the expression of REDD substrates. Indeed, most of the genes that were suppressed by tunicamycin were induced by EREla siRNA in XBP1 deficient liver. (Figure 3 A). Among REDD substrates, Hsdl7b7, Idil, Mvk, Sc4mol and Pmvk, which participate in sterol biosynthesis were suppressed in EREla deficient mice, suggesting that XBP1 regulates these genes in both REDD-dependent and -independent manners.

Example 5: Identification of Cesl and Angptl3 mRNAs as RIDD substrates

A notable group of REDD substrates was the carboxylesterase 1 gene family. The mouse genome contains eight highly homologous CESl-like genes in tandem, generated by gene duplication events during evolution (Figures 9A and 9B) (Holmes, R.S. et al. (2010) Mamm Genome 21:427-441). Cesl genes encode enzymes that possess TG and cholesterol-ester hydrolase activities, and have been implicated in the hydrolysis of neutral lipids stored in lipid droplet and VLDL formation (Parathath, S. et al. (2011) J Biol Chem 286:39683-39692; Quiroga, A.D. and Lehner, R. (2011) Trends Endocrinol Metab 22:218-225; Wei, E. et al. (2010) Cell Metab 11:183-193). Notably, ablation of Ces3/TGH encoded by the Cesld gene decreased plasma TG and cholesterol levels without causing steatosis (Wei, E. et al. (2010) Cell Metab 11 : 183-193). Ceslc, Cesld, Cesle, Ceslf and Ceslg mRNA levels were strikingly lower in XBP1 deficient, but not in IRE la deficient liver compared with WT controls (Figure 4A).

Further, these mRNAs were induced by IRE la siRNA in Xbpl^LKO liver, and suppressed by tunicamycin treatment in WT mice. Western blot analysis of liver lysates revealed marked reduction of TGH (Cesld) and Es-X (Ceslg) proteins in XBP1 deficient liver (Figure 4B).

IREla-induced degradation of Cesl mRNAs may contribute to the hypolipidemic phenotype of XBP1 deficient mice.

Interestingly, angiopoietin-like protein 3 (AngptU) mRNA was also identified as a REDD substrate in the microarray analysis. ANGPTL3 protein is produced and secreted mainly by liver both in humans and mice, and possesses inhibitory activity toward lipoprotein lipase and endothelial lipase (Oike, Y. et al. (2005) Trends Mol Med 11 :473-479). Accordingly, genetic loss of ANGPTL3 resulted in decreased plasma TG and cholesterol levels in both species (Koishi, R. et al. (2002) Nat Genet 30: 151-157; Fujimoto, K. et al. (2006) Exp Anim 55:27-34; Shimamura, M. et al. (2007) Arterioscler Thromb Vase Biol 27:366-372; Musunuru, K. et al. (2010) Engl J Med 363:2220-2227. qRT-PCR confirmed that AngptB mRNA levels were markedly lower in XBP1 deficient liver compared with WT (Figure 4C). Angptl4 and Angptl6 mRNAs that are also abundantly produced in liver were not altered by XBP1 deficiency (Figure 4C). Consistent with the reduction of AngptB mRNA by XBP1 deficiency, plasma Angptl3 protein concentration was also markedly decreased in XbplA mice (Figure 4D). Administration of IREla siRNA into Xbpl^LKO mice significantly induced hepatic Angptl3 mRNA and plasma AngptB protein levels, indicating that hyperactivated IREla reduces AngptB expression in the liver (Figure 4E and 4F).

Example 6: Cleavage of Angptl3mRNA by IREla

To directly demonstrate that AngptB mRNA is cleaved by IREla, we performed in vitro cleavage assays using recombinant IREla. AngptB RNA synthesized in vitro was efficiently cleaved by IREla into two smaller fragments (Figure 4G). Analysis of the mRNA secondary structure predicted the presence of a hairpin structure in AngptB mRNA, which was similar to the IRE la cleavage site in XBPl mRNA (Figure 4H). Mutation of the potential cleavage site abolished the cleavage of Angptl3 mRNA by E Elcc, demonstrating that IRE1 a recognizes the hairpin structure and cleaves AngptB mRNA (Figure 4G). To further validate the IREla cleavage site in Angptl3 mRNA, we transfected 293T cells with WT or cleavage site mutant AngptB expression plasmids together with EREla, and measured the mRNA levels. IREla is activated by overexpression in 293T cells and reduced the expression of WT, but not the mutant AngptB mRNA (Figure 41). Given the critical role of Angplt3 in the catabolism of lipoprotein particles (Lichtenstein, L. and Kersten, S. (2010) Biochim Biophys Acta 1801:415-420), IRE la- induced degradation of Angplt3 mRNA may contribute to the hypolipidemia phenotype of XBPl deficient mice.

Example 7; Ablation of XBPl ameliorates hepatic steatosis and liver damage

in dietary and genetic animal disease models

IREla and XBPl activate hepatic lipid metabolism via transcriptional and post- transcriptional regulation of genes in lipid metabolic pathways. However, the IREla/XBPl signaling pathway is also considered to have a cytoprotective role against ER stress, which has been implicated in metabolic abnormalities in various organs (Cnop, M. et al. (2011) Trends Mol Med.; Ron, D. and Walter, P. (2007) Nat Rev Mol Cell Biol 8:519-529). Several reports have demonstrated that fat accumulation causes ER stress in liver (Sha, H. et al. (2011) Trends Endocrinol Metab 22:374-381 ; Ozcan, U. et al. (2004) Science 306:457-461 ; Puri, P. et al.

(2008) Gastroenterology 134:568-576; Nakatani, Y. et al. (2005) J Biol Chem 280:847-851). We asked whether XBPl deficiency would ameliorate fat accumulation in liver and blood under metabolic stress conditions, given its lipid-lowering effect, or would instead heighten ER stress and lipotoxicity, leading to increased cell death. We investigated the effects of XBPl deficiency in two dyslipidemic animal models: leptin deficient ob/ob mice, and mice fed a high-fat diet.

Ablation of Xbpl in the liver of ob/ob mice was achieved by generating compound mutant mice (Xbpl^LKO;ob/ob) harboring a homozygous Xbpl flox allele, Alb-cre transgene, and ob/ob mutation. Compared with the Xbpl^{f f};ob/ob mice littermate control, Xbpl ^LKO;ob/ob mice displayed markedly lower hepatic TG and plasma cholesterol levels (Figure 5 A and Figure 10), associated with decreased lipogenic gene mRNA levels (Figure 5B), similar to Xbpl mice. Plasma ALT levels were comparable between Xbpl^{f f};ob/ob and Xbpl^LKO;ob/ob mice, indicating that the ablation of Xbpl did not augment liver damage in ob/ob mice (Figure 5 A). CHOP, GADD34, Herpudl and BiP mRNAs which are strongly induced under ER stress conditions by the PERK and ATF6 pathways were not induced in Xbpl^LKO;ob/ob mice (Figure 5B), suggesting that XBP1 ablation did not induce measurable ER stress in ob/ob mouse liver. Furthermore, contrary to the notion that fat accumulation induces ER stress in liver, we failed to detect any measurable induction of UPR markers such as phospho-IREla, phospho-PERK, phospho-eIF2a, processed nuclear ATF6a and XBPls proteins, and BiP, CHOP and ERdj4 mRNAs in ob/ob mouse liver (Figure 10), suggesting that the causal relationship between fat accumulation and ER stress induction in liver should be reassessed.

Feeding a high-fat diet increased plasma TG in XbplA mice to a level similar to WT mice (Figure 5C). However, plasma cholesterol levels were lower in HFD fed XbplA mice than in WT mice. Interestingly, the plasma ALT level was significantly lower in XbplA compared to WT mice (Figure 5C), indicating decreased hepatotoxicity in the absence of XBP1. We did not observe any differences in glucose tolerance and insulin tolerance tests between WT and XbplA mice under HFD conditions (Figure 10).

Example 8: Ablation of XBP1 in ApoE deficient mice can significantly

decrease plasma cholesterol

ApoE deficient mice display high levels of plasma cholesterol and are widely used as a model for atherosclerosis (Zadelaar, S. et al. (2007) Arterioscler Thromb Vase Biol 27: 1706- 1721). To test whether XBP1 ablation could lower cholesterol levels and reduce atherosclerotic lesion development under such extreme conditions, we generated ApoE^{" "};Xbpl ;Mxcre mice that harbored a poly (I:C)-inducible ere recombinase transgene. Poly (I:C) administration activated the Cre and ablated XBP1 to generate ApoE and XBP1 double knock out (ApoE^"'^" ;XbplA) mice. Inducible XBP1 deletion in ApoE^"'^" mice significantly decreased plasma TG and cholesterol levels, although the latter still remained substantially elevated at >200 mg/dL (Figure 6A and 6B). FPLC analysis of the lipoprotein profile demonstrated that the cholesterol distribution was not altered by XBP1 ablation in ApoE^"'^" mice (Figure 6C). Not surprisingly, despite reductions in plasma cholesterol and TG levels in ApoE^{" "};XbplA compared with ApoE^{" "} mice, atherosclerotic lesion formation was not significantly different between these two groups, as measured by the oil red O stained areas in the descending aorta (Figure 6D) and aortic arch (Figure 6E). Thus the 30-40% reduction in plasma cholesterol levels achieved by XBPl ablation, while very significant, was not sufficient to prevent lesion formation in the setting of ApoE deficiency.

Example 9; Effect of XBPl siRNA on plasma lipids

In recent years, RNA interference has received considerable attention as a novel therapeutic strategy, as sequence specific siRNA molecules can target virtually any disease- associated gene with high specificity (Vaishnaw, A.K. et al. (2010) Silence 1, 14). XBPl is an attractive target for RNA therapeutics, given the profound reduction of plasma TG and cholesterol by XBPl ablation. We investigated if the targeted delivery of XBPl siRNA to the liver would decrease plasma lipid levels. Intravenous injection of lipidoid formulated XBPl siRNA decreased XBPl mRNA and protein levels in the liver in a dose-dependent manner, resulting in near complete suppression of XBPls protein at doses > 5 mg/kg (Figure 7A and 7B). It is notable that a intermediate dose of 2.5 mg/kg siRNA had no effect on XBPls protein level, despite ~50% reduction of XBPl mRNA, indicating precise regulation of IRE la activation to maintain constant XBPls protein levels in the liver. Silencing of XBPls expression was stably maintained for more than a week after a single injection of XBPl siRNA at 7.5 mg/kg (Figure 7C). Similar to the effect of genetic ablation of XBPl, XBPl siRNA induced IREloc

hyperactivation and decreased the expression of genes involved in lipid metabolism, similar to what we observed in XBPl knock-out mice (Figure 7D and 7E). XBPl siRNA transiently decreased plasma TG and cholesterol levels not only in C57BL/6 mice (Figure 7F and 7G), but also in apoE deficient mice (Figure 7H and 71), indicating that XBPl siRNA can efficiently reduce even very elevated plasma lipid levels. EQUIVALENTS

Those skilled in the art will recognize, or be able to ascertain using no more than routine experimentation, many equivalents to the specific embodiments of the invention described herein. Such equivalents are intended to be encompassed by the following claims.

Best Available Copy

Table SI. Genes that are suppressed in XBPl deficient liver.

Best Available Copy

Best Available Copy

Best Available Copy

Claims

We claim:

1. A method of identifying candidate modulators of lipid homeostasis comprising, a) determining the ability of a test compound to modulate regulated IRE 1 -dependent decay (RIDD) of a lipid metabolism gene, wherein mRNA of the lipid metabolism gene is degraded by IRE1; and b) selecting said test compound as a compound of interest when said compound modulates REDD of the lipid metabolism gene, to thereby identify candidate modulators of lipid homeostasis.

2. A method of identifying candidate modulators of lipid homeostasis comprising, a) determining the ability of a test compound to modulate XBP1 -mediated transcription of a lipogenic target gene; and selecting said test compound as a compound of interest when said compound modulates XBP1 -mediated transcription of the lipogenic target gene, to thereby identify candidate modulators of lipid homeostasis.

3. The method of claim 1 or 2, wherein said modulation of lipid homeostasis comprises modulation of lipid metabolism.

4. The method of claim 3, wherein said modulation of lipid metabolism comprises induction of lipid clearance from blood.

5. The method of claim 1 or 2, wherein said modulation of lipid homeostasis comprises modulation of lipogenesis.

6. The method of claim 5, wherein said modulation of lipogenesis comprises induction of lipogenesis.

7. The method of any one of claims 1, 5 and 6, wherein said compound inhibits RIDD of mRNAs of lipid metabolism genes.

8. The method of any one of claims 1, 5 and 6, wherein said compound stimulates XBP1- mediated transcription of the lipogenic target gene.

9. The method of any one of claims 1 to 8, wherein said lipid metabolism gene or lipogenic target gene is a hepatic gene.

10 The method of any one of claims 2 to 5, wherein the lipogenic target gene is selected from the group consisting of Sdflll, Cflar, Dnajc3, Dnajb9, Gale, Nans, Edeml, Dnaljbll, Tmem39a, Erolb, Gmppa, Hyoul, Usol, Sec24d, Slc35bl, Uggtl, Ssr3, Ubxn4, Txndc5, Stt3a, Ssrl, Serpl, Ddost, Sec61al, Surf4, 0610007L01Rik, Rrbpl, Ostc, Odd, Cnpy3, Cdk5rap3, Fdps, Spcs3, P4hb, Spcsl, 2810482107Rik, Sec61b, Lman2, Krtcap2, Secllc, Sec61g, Ssr4, Sec63, H13, D17Wsul04e, Chidl, Sndl, Ikbke, Traml, Cbaral, Slc33al, P exile, Mlec, Sqle, Cyp51, Tnfaip8ll, Pmvk, Fdtfl, Hsdl7b7, Anubll, Lifr, Sc4mol, Mvk, and Idil

11. The method of any one of claims 1 and 3 to 5, wherein the lipid metabolism gene is selected from the group consisting of Paqr7, Ptprf, Ces3, AU018778, Cesl, Ceslc, Cesld, Cesle, Ceslf, Ceslg, Lrpl, Bloclsl, Dhcr7, Limd2, Idil, 221041 lKHRik, Es22, Apon, Glrx, Fads2, He, Oaf, Mvk, Gjbl, Sc4mol, Plala, F7, Cpn2, Pxmp2, Abca3, Gm2a, Kcnn2, Nsdhl, Stim2, Dnase2b, Vnn3, Hsdl7b7, Hgsnat, Coll8al, Aldoc, Fadsl, Fnl, Smarcc2, Dera, Spp2, Angptl3, Atrnll, Tppl, Faml08a, Tmprss6, Sidt2, Esl, Dbt, Ropnll, Tex264, Tapbp, Selk, Btbdl, Cyp2c68, Ndufa9, Pmvk, Blvra, Igf2r, Proc, Abcal, Taldol, Bola3, Cyplal, Retsat, Lampl, Rnfl30, Tex2, Afin, Jkamp, 0610007 C21Rik, Plbd2, Cyp51, Tm7sf2, Clu, Hgfac, 3110073H01Rik, Furin, C920025E04Rik, Serpingl, Lipc, Ttcl3, Coll8al, Slc27a5, Pcsk6, Tmem208, Pm20dl, Sqle, Man2a2, Pplrl5b, Azgpl, Ces6, Pla2gl2b, llvbl, Adam9, Mlec, Pdapl, F2, Commd9, Itih4, Dgat2, Cyb561d2, Slc35f5, Algl, Gnptg, Rarresl, Yifla, Slc25al7, Mxd4, Fdps, Papola, Erp44 and Rpnl .

12. The method of claim 11, wherein the lipid metabolism gene is selected from the group consisting of Ceslc, Cesld, Cesle, Ceslf, Ceslg and Angplt3.

13. The method of any one of the preceding claims, wherein said test compound is comprised in a combination of test compounds.

14. The method of claim 1, wherein determining the ability of a test compound to modulate REDD of the lipid metabolism gene comprises contacting an indicator composition with the test compound and determining a level of lipid metabolism gene mRNA in the presence of the test compound.

15. The method of claim 1, wherein determining the ability of a test compound to modulate REDD of the lipid metabolism gene comprises contacting an indicator composition with the test compound and determining a level of lipid metabolism gene-encoded protein in the presence of the test compound.

16. The method of claim 2, wherein determining the ability of a test compound to modulate XBPl -mediated transcription of the lipogenic target gene comprises contacting an indicator composition with the test compound and determining a level of lipogenic target gene mRNA in the presence of the test compound.

17. The method of claim 2, wherein determining the ability of a test compound to modulate XBPl -mediated transcription of the lipogenic target gene comprises contacting an indicator composition with the test compound and determining a level of a reporter of lipogenic target gene transcription in the presence of the test compound.

18. The method of any one of claims 2, 16 and 17, wherein the test compound is further determined to modulate XBPl -mediated transcription of the lipogenic target gene independent of unfolded protein response (UPR) activation.

19. A method for validating a compound as a compound useful in modulating lipid homeostasis in vivo comprising, selecting a candidate compound identified according to the method of any one of the proceeding claims, and testing said candidate compound for modulation of lipid metabolism in an animal model of dyslipidemia.

20. The method of claim 19, wherein said animal is tested for modulation of one or more of plasma cholesterol levels, plasma and/or hepatic triglyceride levels, plasma and/or hepatic lipid levels, steatosis, and lipoprotein metabolism.

21. A method for validating a compound as a compound useful in modulating lipid homeostasis in vivo comprising, selecting a candidate compound identified according to the method of any one of the claims 1 to 18, and testing said candidate compound for modulation of lipid metabolism gene activity in an animal model deficient in XBP1 and/or IRE la.