Pancreatic B-cells expressing ABCAl and uses thereof

FIELD OF INVENTION

[0001] The present invention relates to the field of therapeutic interventions for type 2 diabetes mellitus. More specifically, the therapeutic interventions include modulation of ABCAl expression or biological activity in pancreatic beta-cells.

BACKGROUND OF THE INVENTION

[0002] The pancreas is comprised of exocrine and endocrine tissue. The exocrine pancreatic tissue secretes a battery of proteases, lipases, amylases and other enzymes that are key for normal digestion. The endocrine pancreatic tissue is comprised of three cell types, distinguished by their selective expression of insulin, glucagon or somatostatin. In the adult mammal, insulin is expressed only in the beta-cell of the pancreatic islet, and is regulated in a cell-specific manner. Diabetes mellitus (DM) is a collective term representing several metabolic disorders with a common hallmark of hyperglycemia. Hyperglycemia may result from inadequate or in some cases absent insulin production, inadequate secretion of insulin from the beta cells of the pancreatic islets, or resistance of peripheral tissues to insulin, resulting in reduced uptake of glucose from the circulation. Type 1 DM is characterized by insulin deficiency resulting from autoimmune destruction of insulin-producing cells, while type 2 DM may be more broadly defined and encompasses a variety of disorders characterized by variable degrees of insulin resistance, impaired insulin secretion and increased glucose production. A variety of metabolic, genetic and environmental factors may play in development of type 2 DM, however, the common therapeutic obstacles to be overcome in almost all cases are dysfunctional beta cells and insulin resistance in the peripheral tissues. Obesity is a major risk factor for type 2 DM development, and confers this increased risk through obesity-associated insulin resistance.

[0003] 'Insulin sensitizer' is a broad term to describe the two main classes of drugs used to stimulate insulin production and secretion, and glucose uptake (stimulated by insulin) in the peripheral tissues. Resistance to the action of insulin impairs glucose utilization and increases hepatic glucose output, with both of these effects contributing to the hyperglycemia of diabetes. The precise mechanism of this insulin resistance is not known,

however the pancreatic beta cells respond by increasing insulin production. If this increased production is coupled with decreased insulin secretory capacity, the beta cells may effectively 'burn out', leading to beta cell failure. The clinical outcome of this is dependence on exogenous insulin, but not before significant irreversible damage is done to the tissues of the body as a result of prolonged hyperglycemia.

[0004] The two main classes of insulin sensitizing drugs are sulfonylureas and thiazolidinediones. Sulfonylureas have been in use since the 1950's. Sulfonylureas stimulate insulin release from pancreatic β-cells and also reduce hepatic clearance of insulin, enabling peripheral tissue to increase uptake of blood glucose. Examples of sulfonylureas include repaglinide (PRANDIN), nateglinide (STARLIX) and metformin (GLUCOPHAGE). Thiazolidinediones (TDZs) are a second major class and recent addition to the family of insulin sensitizers. Examples of thiazolidinediones include troglitazone, rosiglitazone (AVANDIA) and pioglitazone (ACTOS). Thiazolidinedione compounds act as agonists of peroxisome proliferators-activated receptors, primarily the gamma subtype (PPARgamma), and indirectly as ABCAl agonists. PPARgamma is highly expressed in beta-cells and is required for normal beta cell development. (Dubois, M. et al. 2000. Diabetologia 43, 1165-1169; Rosen, E. D. et al. 2003. MoI. Cell Biol. 23, 7222-7229). PPARγ activates ABCAl through an LXR mediated pathway in macrophages, and upregulation of ABCAl by TDZs has been demonstrated in macrophages (Llaverias, G., et al., 2004. Biochem. Pharmacol. 68, 155-163; Chinetti G et a. 2001. Nature Medicine 7:53-58). The interaction of a TGZ with PPARgamma may resull in activation insulin-responsive genes involved in regulation of carbohydrate and lipid metabolism and may in turn lead to reduction of insulin resistance in peripheral tissues.

[0005] TDZs are used clinically to reduce insulin resistance in muscle, liver and adipose tissue in subjects with glucose intolerance

Type 2 DM is considered to be not only a disease of glucose intolerance, but a lipid disorder. Free fatty acids (FFA) are the primary endogenous energy source for unstimulated islet cells. Islet metabolism shifts to glucose as an oxidative fuel as islets are stimulated by glucose to produce and secrete insulin. Short term elevation of plasma FFA

enhances glucose-stimulated insulin secretion, however long-term exposure to elevated FFA has an opposite effect, resulting in an impaired beta-cell response to glucose in mammals (Kashyap SR, et al .2004. Am J Physiol Endocrinol Metab, 287:E537-E546; Stefan N, et al. 2001. Diabetes, 50:1143-1148). Lipotoxicity describes this deleterious effect of chronic FFA elevation on insulin secretion from the pancreatic beta cell. FFA that are not metabolized accumulate in islets and are esterified. In vitro, human islet cells incubated with FFA undergo apoptosis (Lupi R. et al. 2002. Diabetes, 51:1437-1442).

[0006] Lipotoxicity is also observed in cells affected by excessive intracellular cholesterol levels. Cholesterol homeostasis is maintained by regulation of its synthesis, transport and degradation. Degradation occurs in the liver where it is metabolized to bile acids and excreted. Cholesterol is transported from peripheral tissues where it is synthesized or stored to the liver by high density lipoprotein (HDL) particles. This transport is generally a protective effect, as sterol overload, is detrimental to cells. Cholesterol buildup in arterial wall macrophages is a causative factor in atherosclerosis. In macrophages, intracellular cholesterol accumulation results in dysfunction of the sarcoplasmic- endoplasmic reticulum calcium ATPase (SERCA) and depletion of ER calcium stores, which ultimately triggers apoptosis.

[0007] The ATP-binding cassette transporter Al (ABCAl) acts to transfer phospholipids and cholesterol to ApoAl of the HDL particles. ABCAl is part of the ATP-binding cassette (ATP transporter) superfamily, which is involved in energy dependent transport of a wide variety of substrates across membranes. The human ABCAl gene maps to chromosome 9q22-q31 and extends over 149 kb. A representative Homo sapiens ABCAl sequence is listed in GenBank under accession number NMJ305502.

[0008] . Upregulating ABCAl expression or activity increases cholesterol efflux and elevates circulating HDL levels in mouse models (Singaraja, R.R. et al., 2001. J. Biol. Chem. 276: 33969-33979). hi non-transgenic mice, an atherogenic diet upregulated ABCAl expression in the liver (Wellington, CL. et al. 2002. Lab Invest. 82: 273-283).

[0009] WO 00/55318 describes methods and reagents for modulating cholesterol levels. US2003/0056234 discusses polymorphisms of the Abcal gene, transgenic animals expressing recombinant Abcal genes, methods of detecting polymorphisms, and kits and

reagents for detection. WO 03/004692 discusses methods for identification of agents that modulate phosphorylation of ABCAl. WO 02/084301 discusses screening assays for agents that may modulate lipid metabolism, specifically agents that may modulate ABCAl activity. WO 02/097123 discusses specific non-coding polymorphism of the abcal gene their use in diagnostic and therapeutic applications. WO 02/101392 discusses methods for treating, preventing or modulating neurological disease, and modulation of pain or fertility with compounds that modulate ABCAl expression or activity. WO 03/033023 discusses drugs that affect expression of ABCAl to ameliorate cholesterolemia and arteriosclerosis. WO 03/092467 discusses compositions and methods for identification of ABCAl agonists.

[0010] US2004/0137423 discusses modulation of ABCAl generally, as well as the role of ABCAl in transporting an interleukin and in some inflammatory disorders and other biological processes besides regulation of cholesterol.

[0011] Two studies have suggested an association between ABCAl polymorphisms and aspects of the metabolic syndrome and type 2 diabetes (Daimon et al 2005. Biochem

Biophys Res Comm. 329:205-210; Villarreal-Molina et al 2007. Diabetes O:db 06-0905vl- 0).

SUMMARY OF THE INVENTION

[0012] The present invention relates to agonists of ABCAl .

[0013] This invention is based, in part, on the surprising discovery that ABCAl is expressed specifically in the beta cells of the endocrine pancreatic tissue, but not in the exocrine pancreatic tissue. Specifically, ABCAl is expressed in the beta cells of the pancreatic islets.

[0014] The invention provides methods for treating a mammal incapable of transporting sufficient intracellular cholesterol and phospholipids out of the pancreatic beta-cell to prevent lipotoxicity or glucotoxicity.

[0015] In accordance with one aspect of the invention there is provided a method for identifying an ABC transporter agonist. The method may comprise providing a pancreatic

beta cell capable of expressing an ABC transporter, contacting the pancreatic beta cell with a candidate compound, and measuring ABC transporter activity in the pancreatic beta cell. Increased ABC transporter activity, relative to a cell not contacted with the candidate compound, indicates that the candidate compound increases ABC transporter activity or expression. The ABC transporter may be ABCAl, ABCGl, ABCG5, ABCG8, ABCG2, ABCA7 or ABCA4. Preferably, the ABC transporter may be ABCAl.

[0016] In accordance with another aspect of the invention, there is provided a method for increasing ABC transporter expression or biological activity in a pancreatic beta cell, the method comprising: administering an expression vector comprising a pancreatic beta-cell control sequence and a nucleic acid encoding an ABC transporter. The ABC transporter may be ABCAl, ABCGl, ABCG5, ABCG8, ABCG2, ABCA7 or ABCA4. Preferably, the ABC transporter may be ABCAl.

[0017] In accordance with another aspect of the invention, there is provided a method for increasing ABC transporter expression or biological activity in a pancreatic beta cell, the method comprising administering an agonist of an ABC transporter. The ABC transporter may be ABCAl, ABCGl, ABCG5, ABCG8, ABCG2, ABCA7 or ABCA4. Preferably, the ABC transporter may be ABCAl and the agonist specifically increases expression or biological activity of ABCAl.

[0018] In accordance with another aspect of the invention, there is provided an expression vector comprising a pancreatic beta cell-specific control sequence, and a nucleic acid encoding an ABC transporter. The ABC transporter may be ABCAl, ABCGl, ABCG5, ABCG8, ABCG2, ABCA7 or ABCA4. Preferably, the ABC transporter may be ABCAl. The pancreatic beta-cell specific control sequence may be selected from the group consisting of: rat, human or mouse insulin promoters.

[0019] In accordance with another aspect of the invention, there is provided a rodent comprising a pancreatic beta-cell specific disruption of the Abcal gene. The rodent may be heterozygous for the pancreatic-beta cell specific disruption of the Abcal gene, or may be homozygous for the pancreatic beta-cell disruption of the Abcal gene.

[0020] In another aspect of the invention, there is provided a cell from a rodent comprising a pancreatic beta-cell specific disruption of the Abcal gene.

[0021] In another aspect of the invention, the pancreatic beta-cell specific disruption of the Abcal gene includes a loxP recombination step.

[0022] In another aspect of the invention, the rodent is a mouse.

[0023] hi accordance with another aspect of the invention, there is provided an agonist of an ABC transporter. The ABC transporter may be ABCAl, ABCGl, ABCG5, ABCG8, ABCG2, ABCA7 or ABCA4. Preferably, the ABC transporter may be ABCAl and the agonist specifically increases expression or biological activity of ABCAl .

[0024] In accordance with another aspect of the invention, there is provided a method of identifying an ABCAl agonist, the method comprising: providing a pancreatic beta cell capable of expressing ABCAl, contacting the pancreatic beta cell with a candidate compound, and measuring ABCAl biological activity in the pancreatic beta cell.

[0025] hi accordance with another aspect of the invention, there is provided a method of increasing ABCAl expression in a pancreatic beta cell, the method comprising: transforming the beta cell with a vector comprising a pancreatic beta-cell control sequence and a nucleic acid encoding ABCAl .

[0026] hi accordance with another aspect of the invention, the ABCAl biological activity includes binding or hydrolysis of ATP, transport of cholesterol, phospholipids, lipids or triglycerides across a membrane. Biological activity may be, for example, a result of, or influenced by, an increase or decrease in ABCAl protein expression or an increase or decrease in ABCAl gene transcription. ABCAl biological activity in pancreatic beta cells may enhance the health and/or function of the beta cells. ABCAl biological activity may therefore, be assessed indirectly by measuring the health and/or function of beta cells.

[0027] hi accordance with another aspect of the invention, the ABCAl biological activity is a decrease in apoptosis.

[0028] In accordance with another aspect of the invention, the ABCAl biological activity is cholesterol efflux

[0029] In accordance with another aspect of the invention, the ABCAl biological activity is glucose tolerance.

[0030] In accordance with another aspect of the invention, the ABCAl biological activity is insulin response.

[0031] In accordance with another aspect of the invention, the pancreatic beta cell is from a subject heterozygous for ABCAl expression .

[0032] In accordance with another aspect of the invention, there is provided a kit for testing the ABCAl biological activity of a composition, the kit comprising a control composition having ABCAl biological activity, together with instructions for use.

[0033] Other aspects and features of the present invention will become apparent to those ordinarily skilled in the art upon review of the following description of specific embodiments of the invention in conjunction with the accompanying figures.

[0034] This summary of the invention does not necessarily describe all features of the invention. Other aspects and features of the present invention will become apparent to those of ordinary skill in the art upon review of the following description of specific embodiments of the invention.

BRIEF DESCRIPTION OF THE DRAWINGS [0035] These and other features of the invention will become more apparent from the following description in which reference is made to the appended drawings wherein:

[0036] FIGURE 1 shows western blot results of ABCAl expression in whole pancreas, murine wild-type islet cells (wt islet), Abcal-I- mice islet cells (ko islet), and cultured INS- 1 cells (INS-I cell). Lysates were probed with a polyclonal antibody raised against ABCA 1 , and GAPDH as a loading control.

FIGURE 2 shows western blot results of ABCAl expression in INS-I cells treated for 48 hours with vehicle control (ethanol) or 1 uM rosiglitazone in ethanol. GAPDH served as a loading control.

FIGURE 3 shows experimental results relating to glucose homeostasis in ABCAl global knock-out mice (Abcal-/-) and wildtype control (Abcal ^+/+ ). A) Gluclose tolerance testing of Abcal +/+ (solid square) and Abcal -/- mice (solid triangle); n=8 mice per group. B) Insulin tolerance testing of Abcal +/+ (solid square) and Abcal -/- mice (solid triangle); n=4 mice per group. C) Acute insulin response following intraperitoneal glucose challenge in Abcal +/+ (dark bar) and Abcal -/- mice (light bar) (serum insulin/blood glucose change; * p <0.01).

FIGURE 4 shows experimental results relating to glucose stimulated insulin secretion from islets isolated from mice lacking ABCAl (Abcal-/-) and wildtype littermate controls (Abcat+/+). Isolated islets (3 mice per genotype) were cultured in low (open bar) and high glucose (solid bar) conditions and the media and islet insulin concentrations were determined by ELISA.

FIGURE 5 shows experimental results relating to generation and characterization of ABCAl beta-cell specific knock-out mice (Abcal-P/-P). Mice lacking ABCAl specifically in beta-eels have total plasma cholesterol (A) and HDL cholesterol (B) concentrations that are indistinguishable from littermate controls. Each block represents one mouse.

FIGURE 6 shows experimental results relating to glucose homeostasis in ABCAl beta-cell specific heterozygous (Abcal ^{+/ P} ) knockout mice (solid triangle) and wildtype (Abcal ^+/+ ) mice (solid square). n= 8 females per group; age 14.3+/-2 (wildtype); 15.8+/-2 (heterozygous)(* p<0.01; ** p<0.0005)

FIGURE 7 shows experimental results relating to glucose tolerance in Abcal-Pl/Pl mice on a high-fat diet. Beta-cell Abcal is essential for the response to rosiglitazone treatment. A) Western blot for Abcal in INS-I cells following overnight culture with 1 mM rosiglitazone (Rosi) or control. (B) Glucose tolerance testing in Abcal+/+ mice receiving a chow diet, and after 2 weeks of high-fat or high-fat + rosiglitazone feeding. n= 12 mice per group. Solid circle - control diet; open triangle - high fat (HF) diet; open circle - high

fat diet with 30 mg/kg/day rosiglitazone (HF Rosi). (C) Glucose tolerance testing in Abcal-P/-P receiving a chow diet, and after 2 weeks of high-fat or high-fat + rosiglitazone feeding, n = 12 mice per group. Solid circle - control diet; open triangle - high fat (HF) diet; open circle - high fat diet with 30 mg/kg/day rosiglitazone (HF Rosi). Statistical comparison of IPGTT curves in b and c is by repeated-measures ANOVA with Neuman-Keuls post-hoc test. (D) Area under the intraperitoneal glucose tolerance curve (AUCipgtt) for the mice described in B and C. Wildtype mice (Abcal+/+ or +/+) and beta- cell specific knockout mice (Abcal-P/-P or -P/-P). Open bar - control diet; light bar - high fat (HF) diet; black bar - high fat diet with 30 mg/kg/day rosiglitazone (HF Rosi). (E) Fasting plasma insulin concentrations for the mice described in B and C. n = 6 mice per group. (F) Insulin tolerance testing in Abcal+/+ and Abcal-P/-P mice on high-fat or high- fat + rosiglitazone diets, n = 6 mice per group. (G) Plasma cholesterol levels Abcal+/+ and Abcal-P/-P mice after 2 weeks of control (open bar), high-fat (light bar) or high-fat + rosiglitazone (black bar) feeding. Lipid levels were determined by enzymatic assay. (H-K) Pancreatic islets from Abcal +/+ (H - high fat diet, I - high fat plus rosiglitazone diet) and Abcal -P/-P mice (J - high fat, K - high fat plus rosiglitazone diet).

[0037] FIGURE 8 shows Glucose homeostasis in beta-cell-specific Abcal knockout mice.

(A) Glucose tolerance testing in 2-month-old Abcal -P/-P homozygotes (n = 5 per group).

(B) As in A but for 4-month-old Abcal-P/-P homozygotes (n = 8 per group). (C) As in a but for 2-month-old Abcal +/-P heterozyygotes (n = 4 per group). (D) As in a but for A- month-old Abcal+/-P heterozygotes (n = 8 per group). Solid circle - Abcal +/+; open circle Abcal -P/-P; open diamond Abcal +/-P.

[0038] FIGURE 9 shows In vivo glucose stimulated insulin release (plasma insulin) from Abcal +/+ and Abcal -P/-P at 0 and 15 minutes. N=6 mice per group.

[0039] FIGURE 10 shows acute phase insulin insulin secretion following 3 mg/kg intraperitoneal glucose injection. Solid circles - Abcal +/+, open circles Abcal -P/-P. n=9 per group.

[0040] FIGURE 11 shows Acute-phase insulin secretion after an intraperitoneal injection of L-arginine 0.3 mg/g body weight. Solid circles - Abcal +/+, open circles Abcal -P/-P. n=5 per group.

[0041] FIGURE 12 shows beta cell mass in mice lacking beta cell ABCAL (A) Beta cell mass was determined by histomorphmetry from 10 evenly spaced pancreatic sections. (B) Immunofluorescence for insulin (green) and glucagons (red) from Abcal +/+ and Abcal - P/-P mice. (C :Low power view of pancreata from Abcal +/+ and Abcal -P/-P mice showing immunofluorescence for insulin green).

[0042] FIGURE 13 shows impaired insulin secretion in isolated islets lacking beta-cell Abcal. (a) Insulin secretion from isolated islets. Islets were cultured overnight and then stimulated for 1 h in the conditions indicated. Values represent pooled data from three separate experiments, each consisting of pooled islets from three mice per genotype, and values are expressed as percent of islet content relative to basal secretion (which is arbitrarily set at 1). (b) Islet insulin content, (c) Pancreatic insulin content, (d) mRNA expression of Insl and Ins2 genes in islets isolated from Abcal-P/-P mice and controls, (e) Islet perifusion. (f) Area under the curve (AUC) from islet perfusion experiment., n = 18 per group for a, b and d. n = 3 per group for c, e and f. *P < 0.05, ,**P < 0.01, ***P < 0.001 (two-tailed Student's t-test).

[0043] FIGURE 14 shows Functional characterization of beta cell ABCAL Cholesterol efflux of islets isolated from beta-cell specific Abcal knockout mice (Abcal-P/-P) and littermate controls mice (Abcal fl/fl). Open bar, control, black bar ApoAl

[0044] FIGURE 15 shows the absence of beta-cell Abcal results in altered cholesterol homeostasis in isolated islets, (a) Total (TC), free (FC) and esterified (CE) cholesterol and triglyceride (TG) levels in islets isolated from beta-cell-specific Abcal knockout mice and controls (Abcal+/+ and Abcal-P/-P) and Abcal global knockout mice and controls

(Abcal+/+ and Abcal-/-) determined by gas-liquid chromatography and enzymatic assay, n = 3-6 mice per group, (b) Relative amount of various mRNAs in isolated islets from Abcal+/+ and Abcal-P/-P mice. The Srebfl gene was amplified with primers specific for the Srebp-lc gene product. Each value represents the amount of mRNA relative to that in Abcal+/+ mice (arbitrarily set at 1 for each transcript in these mice). Values represent

pooled data from 2-5 separate experiments in which equal amounts of total RNA were pooled from three mice. *P < 0.05, **P < 0.01.

[0045] FIGURE 16 shows Islet free cholesterol levels and triglyceride levels in islets isolated from Abcal+/+ and Abcal-P/-P mice after 2 weeks of high-fat or high-fat + rosiglitazone feeding. Lipid levels were determined by gas-liquid chromatography and enzymatic assay on pooled islets from n = 3 mice per group. Error bars represent standard deviation of interassay variation. **P < 0.01. Light bars High fat diet; black bars, high fat + 30 mg/kg/day rosiglitazone.

[0046] FIGURE 17 shows blood glucose (A) and insulin (B) levels in subjects heterozygous for disease-causing mutations in the ABCAl gene, compared with unaffected relatives. Total area under the curve for glucose (AUCglucose) (C) or insulin (AUCinsulin) (D) for the groups in A and B, respectively. Control - open circles, open bar. Heterozygotes - solid square, solid bars.

[0047] FIGURE 18 shows an acute insulin response to glucose over 60 minutes. Subjects heterozygous for disease-causing mutations in the ABCAl gene (black bar), unaffected relative controls (open bar).

DETAILED DESCRIPTION

[0048] hi the description that follows, a number of terms are used extensively, the following definitions are provided to facilitate understanding of various aspects of the invention. Use of examples in the specification, including examples of terms, is for illustrative purposes only and is not intended to limit the scope and meaning of the embodiments of the invention herein.

[0049] Use of the term 'a' or 'an' includes both singular and plural references.

[0050] The invention provides, generally, methods for identification of an ABCAl agonist. The method may comprising contacting a pancreatic beta cell capabable of expressing ABCAl with a test compound and measuring the ABCAl biological activity in the pancreatic beta cell. The pancreatic beta cell may be from any of several species of test subject and may be cultured ex vivo, or in vitro, or may be in vivo, for example, in a test

subject. The test subject may be human, or a non human animal such as a rodent. The rodent may be a 'wild-type' or may be a transgenic animal comprising a defined mutation in at least one copy (a heterozygote) or both copies (a homozygote) of an Abcal gene.

[0051] The diabetogenic effect of increased cellular fat content. Increased intracellular fat content may manifest in several tissues including muscle, liver and pancreatic islets, and may generally be referred to as lipotoxicity. The intracellular fats may comprise cholesterol, free fatty acids, phospholipids, triglycerides or other lipids.

[0052] The diabetogenic effect of elevated blood glucose concentration may be generally referred to as glucotoxicity. Increased plasma glucose may lead to altered levels of intracellular glucose.

[0053] The ability of a mammal to metabolize glucose may be generally referred to as glucose tolerance. A glucose tolerance test may involve administration of a measured dose of glucose to a fasting mammal, and subsequent determination of glucose levels in the blood or urine at measured intervals thereafter. A glucose tolerance test may be useful for the detection of hyperglycemia or diabetes mellitus.

[0054] The ability of a mammal to respond to low blood glucose levels may be generally referred to as insulin tolerance. An insulin tolerance test may involve administration of a measured dose of insulin to a fasting mammal, and subsequent determination of glucose levels in the blood at measured intervals thereafter. An insulin tolerance test may be useful for detection of peripheral insulin resistance, pituitary or adrenal gland dysfunction, hypoglycemia, hyperglycemia or diabetes mellitus.

[0055] Type 2 diabetes mellitus (type 2 DM) is broadly defined and may correlate with a variety of genetic, metabolic and environmental factors. Pancreatic beta cells are not destroyed in entirety, and insulin is produced by the beta cells, however there are obstacles relating to sufficient production of insulin, secretion, response of peripheral tissues to secreted insulin and regulation of blood glucose generally. Risk factors for type 2 DM include family history, obesity (body-mass index >27), age (>45), previously identified impaired fasting glucose or impaired glucose tolerance, hypertension, low HDL cholesterol

level (<90 mmol/L), elevated triglyceride level (>2.82 mmol/L) and ethnicity (African, Asian, Hispanic, Pacific Islands or Native American).

[0056] In individuals with low HDL cholesterol, upregulation of reverse cholesterol transport may be beneficial. ABC transport proteins that may participate in reverse cholesterol transport include ABCAl, ABCGl, ABCG5, ABCG8, ABCG2, ABCA7 or ABCA4. Reverse cholesterol transport refers to the transport of cholesterol produced in the tissue to the liver for degradation. Cholesterol is transferred from the cell by an ATP transporter and bound to the lipoprotein ApoE, forming high density lipoprotein particles (HDL particles). HDL particles are carried by the blood to the liver, where it is removed from the HDL particles and degraded.

[0057] Insufficient ABCAl activity leads to a buildup of intracellular cholesterol and phospholipids, which in turn causes lipotoxicity in the affected cells and tissues. In the pancreatic islets, this buildup of lipids adversely affects the energy metabolism of the islet cells and results in apoptotic death of the pancreatic islets. Without wishing to be bound by theory, upregulation of islet ABCAl activity may be protective against lipotoxicity and prevent apoptotic death of islet cells.

[0058] Therapeutic strategies or treatments that increase beta-cell ABCAl biological activity, such as ABCAl direct agonists, replacement-of -function approaches such as gene therapy or transplantation of pancreatic islets or islet precursors expressing ABCAl may be useful for preserving or modulating beta-cell function, thus preventing the onset of symptomatic type 2 DM.

ABCAl expression and glucose tolerance

[0059] The present invention provides a method for identification of an ABCAl agonist in pancreatic islet cells.

[0060] The ABCAl gene is regulated by a complex transcriptional network, and most ABCAl transcriptional agonists act through these transcription factors. ABCAl transcription may be increased by several indirect, transcriptional agonists. Such transcriptional agonists may include lipids such as oxysterols, retinoids, PPAR-alpha agonists, PPAR-gamma agonists or PPAR-delta agonists, hormones such as estrogen,

cytokines such as oncostatin M, pharmaceuticals such as A2A-receptor agonists, niacin, verapamil, bisphosphonate or statins (Schmitz, G. and Langmann, T. 2005. Biochem Biophys Acta 1735:1-19; incorporated herein by reference). Other examples of transcriptional agonists or direct agonists, of ABCAl include thiazolidinediones, such as those described herein. Increasing ABCAl expression in the pancreatic beta-cells may increase the ABCAl biological activity in the pancreatic beta-cell, thus preventing the onset of symptomatic type 2 DM.

[0061] In mice, ABCAl expression is unusually high in wild-type isolated mouse islets in comparison to published results (Figure 1). ABCAl is upregulated in response to a thiazolidinedione, indicating that the PPARgamma-LXR-ABCAl pathway is active and inducible by TDZs. Glucose tolerance testing on the Abcal-/- mice demonstrates significant impairment in glucose tolerance in these animals, while insulin tolerance testing revealed no difference in glucose clearance in the Abcal-/- mice compared with controls, indicating that the impaired glucose tolerance in the Abcal -/- animals is not due to peripheral insulin resistance.

[0062] Beta cell function may be assessed by measuring plasma insulin levels in Abcal -/- and control mice before and after glucose challenge. The net change in plasma insulin after glucose challenge was significantly less in mice lacking ABCAl compared to controls (Figure 3b), indicating that these mice have impaired glucose stimulated insulin release in vivo. Glucose-stimulated insulin release is also impaired in islets from abcal -/- mice (Figure 4).

[0063] ABCAl beta cell specific knockout mice were developed and used to illustrate the beta-cell specific role of ABCAl in glucose homeostasis. Such mice (both heterozygous and homozygous) may be generally referred to as ABCAl-BSKO (ABCAl-beta cell specific knockout). Abcal +/-P denotes a heterozygous mouse, Abcal -P/-P denotes a homozygous mouse.

[0064] ABCAl-BSKO (Abcal +/-P and Abcal -PAP) mice display normal levels of plasma total cholesterol and HDL. Without wishing to be bound by theory, beta-cell ABCAl may not play a role in plasma HDL metabolism (Figure 5).

[0065] Gene dosing of beta cell ABCAl may be important for islet function, and in a cell having marginal ABCAl biological activity, even a modest increase such as that observed in a homozygous normal or wild type compared to a heterozygous subject, may improve islet function.

[0066] Cholesterol efflux from pancreatic islets is also ABCAl -dependent, as demonstrated in Abcal-P/-P mice. Exposure to higher circulating cholesterol levels, as occurs to the islets from Abcal-P/-P mice, (Example 9) results in a greater degree of cholesterol accumulation and more severe impairment in glucose tolerance compared to mice with global deletion of ABCAl.

[0067] In mice with normal beta-cell ABCAl expression, rosiglitazone restores glucose tolerance (Figure 7B). In beta-cell specific knockout mice (Abcal-P/-P), dietary rosiglitazone has no significant effect (Figure 7C).

ABCAl Agonists and ABCAl biological activity

[0068] In some embodiments of the invention, methods of identifying an agonist of an ABC transporter are provided. The ABC transporter may be ABCAl, ABCGl, ABCG5, ABCG8, ABCG2, ABCA7 or ABCA4. Preferably, the ABC transporter may be ABCAl and the agonist specifically increases expression or biological activity of ABCAl.

[0069] The term "agonist" as used herein refers to a composition capable of contacting or combining with a receptor on a cell and initiating or enhancing the same reaction or activity otherwise produced by the binding of an endogenous composition. An agonist may increase the expression of a gene product or protein, or may increase the biological activity of a gene product or protein (a 'direct agonist'). An agonist may also act indirectly on a gene product or protein (an 'indirect agonist'), for example by altering transcription or translation of a transcription factor or translation factor that in turn acts on the gene product or protein.

[0070] The term "antagonist" as used herein refers to a composition that acts to reduce the physiological activity of another compound or composition, for example by combining with and blocking the receptor of the endogenous composition. An antagonist may decrease the expression of a gene product or protein, or may decrease the biological

activity of a gene product or protein (a 'direct antagonist'). An antagonist may also act indirectly on a gene product or protein (an 'indirect antagonist'), for example by altering transcription or translation of a transcription factor or translation factor that in turn acts on the gene product or protein.

[0071] An ABCAl agonist refers generally to a composition that affects ABCAl biological activity, or improve the stability or half-life of ABCAl in pancreatic beta-cells.

Increasing ABCAl biological activity or stability, or increasing the half-life of ABCAl in the pancreatic beta-cell may prevent the onset of symptomatic type 2 DM. Compositions according to various embodiments of the invention may comprise a direct, or indirect, ABCAl agonist, and may be generally referred to as 'compositions'. Examples of ABCAl agonists include, for example thiazolidinediones such as rosiglitizone, pioglitazone, troglitazone, MCC-555, ciglitazone and the like.

[0072] A 'test compound' or 'test agonist' or 'candidate agonist' as used herein refers to a compound or composition that has, or is suspected of having, ABCAl agonist activity.

[0073] The terms "subject" and "patient" may be used interchangeably. A "subject" refers to an animal, or a mammal, including, but not limited to, a rodent, mouse, rat, dog, cat, pig, or primate, including but not limited to a monkey, chimpanzee or human. A subject may be a transgenic animal, such as a transgenic mouse.

[0074] The term "ABC transporter" as used herein refers to a transporter that hydrolyzes ATP and transports a substance across a membrane. An ABC transporter may include an ATP binding cassette and a transmembrane region. Examples of ABC transporters may include ABCAl, ABCGl, ABCA4, ABCA7, ABCG5, ABCG8 or ABCG2.

[0075] The term "ABCAl biological activity" as used herein refers to binding or hydrolysis of ATP, transport of cholesterol, phospholipids, lipids or triglycerides across a membrane. Biological activity may be, for example, a result of, or influenced by, an increase or decrease in ABCAl protein expression or an increase or decrease in ABCAl gene transcription. ABCAl biological activity in pancreatic beta cells may enhance the health and/or function of the beta cells. ABCAl biological activity may therefore, be assessed indirectly by measuring the health and/or function of beta cells.

[0076] In some embodiments of the invention, pancreatic islet cells heterozygous for a deletion or mutation in a gene encoding ABCAl may be used in an assay or screen to identify an agonist of ABCAl. Such an agonist may, for example, increase ABCAl gene transcription, expression, biological activity, stability or half-life in a pancreatic beta cell. Part or all of the gene may be deleted, for example in a 'knockout' mutant, or a mutation that affects ABCAl biological activity may be present. Examples of mutations that have been identified in human subjects, and may be associated with alterations in ABCAl biological activity may be found in, for example, Brunham et al 2006. Annu Rev. Nutr 26:105-129, herein incorporated by reference. Homologues of the mutations identified in human subjects may be incorporated into the ABCAl gene of another subject, such as a mouse, using known methods, including those exemplified herein, and described by references cited herein.

[0077] The pancreatic beta cells may be primary cells that are cultured ex vivo, or may be transformed or immortalized cells or cell lines maintained in vitro using standard methods. INS-I cells are an example of such a cell line (Asfari M et al 1992. Endocrinology 130:167-178).

[0078] Alternatively, pancreatic islet cells may be obtained from an Abcal+/P- mouse.

[0079] To identify an ABCAl agonist, a pancreatic islet cell as described is exposed to a test compound, for example by inclusion of such a test compound in the cell culture medium or buffer. Alternatively, a test compound may be administered to an animal having ABCAl expression in a pancreatic beta cell. For example, the test compound may be administered in the diet. ABCAl expression, biological activity, regulated catabolism or stability may be measured and compared to that of a control compound. Examples of control compounds include those known to have ABCAl agonist activity, such as a thiazolidinedione, such as rosiglitazone. Therefore, the invention provides a method for identification of an ABCAl agonist, providing a pancreatic beta cell capable of expressing ABCAl, contacting the pancreatic beta cell with a candidate compound and measuring ABCAl biological activity in the pancreatic beta cell.

[0080] Isolated pancreatic beta-cells from heterozygous ABCAl-BSKO (Abcal +/P-) mice may be used in vitro for screening or identifying ABCAl agonists. An ABCAl-

BSKO heterozygous mouse was prepared and may be used for identification of ABCAl agonists, The pancreatic beta-cells of the ABCAl-BSKO mouse (in vitro, ex vivo or in vivo) may be exposed to candidate compounds, and the ABCAl biological activity measured, for example, by a cholesterol efflux assay. Alternatively, expression of ABCAl RNA or protein may be measured by Northern blot, western blot, or other methods well- known to those of skill in the art.

[0081] In some embodiments of the invention, an ABCAl-BSKO mouse, when treated with a carrier or a test compound that does not have ABCAl agonist activity, may have Abcal transcription or protein expression levels from about 10 to about 80% relative to that of a wild-type control mouse. Preferably, the ABCAl-BSKO mouse may have Abcal transcription or protein expression levels of about 50% , relative to that of a wild-type control mouse.

[0082] Alternatively, a test compound may be administered to an ABCAl-BSKO mouse, for example an Abcal +/P- mouse, and the pancreatic islets of the mouse assessed for ABCAl transcription, expression, biological activity or stability, or glucose and insulin tolerance. Alternately, agonists of ABCAl may affect ABCAl biological activity in pancreatic beta cells without any effect on transcription or expression in pancreatic beta cells, and instead increase the lipid metabolism, such as cholesterol efflux in pancreatic beta cells.

[0083] Increasing transcription of an ABCAl gene may result in increased translation and expression of ABCAl protein. ABCAl transcription may be measured, for example, by standard Northern blot analysis using an ABCAl nucleic acid sequence or fragment thereof as a hybridization probe. Transcription may also be measured by quantitative or semi-quantitative RT-PCR, or by other sequence-specific mRNA detection methods. The conditions and methods used in these techniques are known to those skilled in the art of molecular biology, and examples of them may be found, for example, in Ausubel et al. Current Protocols in Molecular Biology, John Wiley & Sons, New York, N.Y., 1988, incorporated herein by reference.

[0084] Expression of ABCAl protein may be measured by immunological techniques such as ELISA or Western blot using an anti-ABCAl antibody and techniques known in

the art. The conditions and methods used in these techniques are known to those skilled in the art of molecular biology, and examples of them may be found, for example, in Ausubel et al. Current Protocols in Molecular Biology, John Wiley & Sons, New York, N.Y., 1988 incorporated herein by reference. The level of ABCAl expression in the presence of the candidate agonist is compared to the level measured for the same cells in the same culture medium or in a parallel set of test animals, but in the absence of the candidate agonist.

[0085] ABCAl biological activity in pancreatic beta-cells may be measured using a cholesterol efflux assay. A cholesterol efflux assay measures the ability of cells to transfer cholesterol to an extracellular acceptor molecule and is dependent on ABCAl function. In this procedure, cells are loaded with radiolabeled cholesterol by any of several biochemical pathways (Marcil, M. et al., 1999. Arterioscler. Thromb. Vase. Biol. 19: 159-169; incorporated herein by reference). Cholesterol efflux is then measured after incubation for various times (typically 0 to 24 hours) in the presence of HDL or purified apoprotein, for example apoAI, apoC or apoE. Cholesterol efflux is determined as the percentage of total cholesterol in the culture medium after various times of incubation. ABCl expression levels and/or biological activity are associated with increased efflux while decreased levels of ABCl are associated with decreased cholesterol efflux.

[0086] The beta-cell protective effect of increased ABCAl biological activity may also be assessed by measuring the change in levels of apoptosis in pancreatic beta cells under lipotoxic conditions.

[0087] Apoptosis occurs in tissues during normal development and in response to environmental insult. Apoptotic death is hallmarked by several unique features including DNA fragmentation, nuclear enlargement, alterations in nuclear membrane permeability to certain dyes or stains, alterations in cell membrane biochemistry, alterations in cell surface proteins, and lipid markers or cell shrinkage and the appearance of 'blebs' of cell membranes (Watanabe M, et al. 2002. Microsc Microanal. 8(5):375-91, incorporated herein by reference). Most apoptosis assays detect these cellular changes, which occur late in apoptosis. Apoptosis may be detected by any of several methods that serve to monitor these unique changes. DNA fragmentation may be detected by agarose gel electrophoresis of cellular DNA (Zhang, J.H. & Xu, M. 2000.CeIl Res 10, 205-211 ; incorporated herein

by reference) or by a terminal dUTP nick end-labelling (TUNEL) assay (Modak, S. P. & Bollum, FJ. 1972. Exp Cell Res 75, 307-313.; Gavrieli, Y. et al. 1992. J Cell Biol 119, 493-501 ; both of which are incorporated herein by reference). PCR-based methods to fluorescently label and detect specific transcripts of pro-apoptotic genes may also be used, and have the advantage of providing very early signals of apoptotic events. Alteration in death protein expression is another early detection method, and can be observed by ELISA, western blot or other immunological techniques, using methods in protocols known to those in the art. Examples of such 'pro-apoptotic' gene products include caspases, such as caspase-3. Cell membrane biochemistry and cell surface markers are altered as cells initiate and progress through apoptosis - for example, the phosphatidylserine residues exposed to the external plasma membrane surface of apoptotic cells may be detected by the binding of annexin V in the presence of calcium (Van den Eijnde, et al. 1997. Cell Death Differ 4, 311-316; incorporated herein by reference).

Compositions

[0088] ABCAl agonists, or suspected ABCAl agonists (collectively referred to as an ABCAl agonist' or a 'test compound') may be included in a composition for administration to a subject to assess the effect on ABCAl biological activity.

[0089] A "pharmacologically effective amount" or an "effective amount" of a composition as used herein refers to using an amount of a composition present in such a concentration to result in a therapeutic level of drug delivered over the term that the drug is used. This may be dependent on mode of delivery, time period of the dosage, age, weight, general health, sex and diet of the subject receiving the composition.

[0090] Compositions according to various embodiments of the invention may be formulated with any of a variety of pharmaceutically acceptable excipients, frequently in an aqueous vehicle such as Water for Injection, Ringer's lactate, isotonic saline or the like. Pharmaceutically acceptable excipients include, for example, salts, buffers, antioxidants, complexing agents, tonicity agents, cryoprotectants, lyoprotectants, suspending agents, emulsifying agents, antimicrobial agents, preservatives, chelating agents, binding agents, surfactants, wetting agents, solvents, dispersion media, coatings, antibacterial, antimicrobial or antifungal agents, isotonic and absorption delaying agents, non-aqueous

vehicles such as fixed oils, or polymers for sustained or controlled release, and the like that are physiologically compatible. See, for example, Berge et al. (1977. J. Pharm Sci. 66: 1- 19), or Remington- The Science and Practice of Pharmacy, 21 ^st edition. Gennaro et al editors. Lippincott Williams & Wilkins Philadelphia (both of which are herein incorporated by reference).

[0091] Compositions according to various embodiments of the invention may be administered by any of several routes, including, for example, subcutaneous injection, intraperitoneal injection, intramuscular injection, intravenous injection, epidermal or transdermal administration, mucosal membrane administration, orally, nasally, rectally, or vaginally. See, for example, Remington- The Science and Practice of Pharmacy, 21 ^st edition. Gennaro et al editors. Lippincott Williams & Wilkins Philadelphia (incorporated herein by reference). Carrier formulations may be selected or modified according to the route of administration.

[0092] Compositions according to various embodiments of the invention may be provided in a unit dosage form, or in a bulk form suitable for formulation or dilution at the point of use.

[0093] Compositions according to various embodiments of the invention may be administered to a subject in a single-dose, or in several doses administered over time. Dosage schedules may be dependent on, for example, the subject's condition, age, gender, weight, route of administration, formulation, or general health. Dosage schedules may be calculated from measurements of adsorption, distribution, metabolism, excretion and toxicity in a subject, or may be extrapolated from measurements on an experimental animal, such as a rat or mouse, for use in a human subject. Optimization of dosage and treatment regimens are discussed in, for example, Goodman & Gilman's The Pharmacological Basis of Therapeutics 11 ^th edition. 2006. LL Brunton, editor. McGraw- Hill, New York, or Remington- The Science and Practice of Pharmacy, 21 ^st edition. Gennaro et al editors. Lippincott Williams & Wilkins Philadelphia (both of which are incorporated herein by reference).

[0094] The terms "treatment," , "treating", "therapeutic use," or "treatment regimen" as used herein may be used interchangeably are meant to encompass prophylactic, palliative,

and therapeutic modalities of administration of the compositions of the present invention, and include any and all uses of the presently claimed compounds that remedy a disease state, condition, symptom, sign, or disorder caused by an lipotoxicity-based pathology, a hyper or hypoglycemic-based pathology, or a pathology related to dysfunction or dysregulation of pancreatic islet cells, or pancreatic beta cells. Thus, any prevention, amelioration, alleviation, reversal, or complete elimination of an undesirable disease state, symptom, condition, sign, or disorder associated with such pathologies is encompassed by the present invention. A treatment may comprise administration of an effective amount of a composition as described herein, alone or in combination with another treatment or therapeutic modality as is known in the art.

Nucleic acid compositions

[0095] Nucleic acid sequences include those which encode a polypeptide, amino acid sequence or protein, and are capable of at least being transcribed. Thus, sequences encoding mRNA, tRNA and rRNA are included within this definition. The sequences may be in the sense or antisense orientation with respect to the promoter. Antisense constructs can be used to inhibit the expression of a gene in a cell according to well-known techniques. Sequences encoding mRNA may optionally include some or all of 5' and/or 3' transcribed but untranslated flanking sequences naturally, or otherwise, associated with the translated coding sequence. It may optionally further include the associated transcriptional control sequences normally associated with the transcribed sequences, for example transcriptional stop signals, polyadenylation sites and downstream enhancer elements.

[0096] In some embodiments of the invention, the nucleic acid comprises a sequence encoding ABCAl. The sequence encoding ABCAl may be, for example, a human, rat, mouse sequence, or a sequence substantially identical thereto. An exemplary nucleic acid sequence encoding human ABCAl is listed in GenBank under accession number NM_005502 (incorporated herein by reference).

[0097] Two nucleic acid sequences may be "substantially identical" if they hybridize under high stringency conditions. In some embodiments, high stringency conditions are, for example, conditions that allow hybridization comparable with the hybridization that occurs using a DNA probe of at least 500 nucleotides in length, in a buffer containing 0.5

M NaHPO ₄ , pH 7.2, 7% SDS, 1 mM EDTA, and 1% BSA (fraction V), at a temperature of 65°C, or a buffer containing 48% formamide, 4.8x SSC, 0.2 M Tris-Cl, pH 7.6, Ix Denhardt's solution, 10% dextran sulfate, and 0.1% SDS, at a temperature of 42°C. (These are typical conditions for high stringency northern or Southern hybridizations.) Hybridizations may be carried out over a period of about 20 to 30 minutes, or about 2 to 6 hours, or about 10 to 15 hours, or over 24 hours or more. High stringency hybridization is also relied upon for the success of numerous techniques routinely performed by molecular biologists, such as high stringency PCR, DNA sequencing, single strand conformational polymorphism analysis, and in situ hybridization. In contrast to northern and Southern hybridizations, these techniques are usually performed with relatively short probes (e.g., usually about 16 nucleotides or longer for PCR or sequencing and about 40 nucleotides or longer for in situ hybridization). The high stringency conditions used in these techniques are well known to those skilled in the art of molecular biology, and examples of them can be found, for example, in Ausubel et al., Current Protocols in Molecular Biology, John Wiley & Sons, New York, N.Y., 1998, which is incorporated herein by reference.

[0098] The degree of sequence identity can be quantified using publicly available sequence analysis software (e.g., Sequence Analysis Software Package of the Genetics Computer Group, University of Wisconsin Biotechnology Center, 1710 University Avenue, Madison, Wis. 53705, or BLAST software available from the National Library of Medicine, or as described herein). Examples of useful software include the programs Pile- up and PrettyBox. Such software matches similar sequences by assigning degrees of homology to various substitutions, deletions, substitutions, and other modifications.

[0099] Isolated nucleic acids corresponding to the ABCAl gene, or a fragment thereof, may be used as a tool to express the ABCAl protein in an appropriate cell in vitro or in vivo. Expression systems which may be employed include baculovirus, herpes virus, adenovirus, adeno-associated virus, coxsackievirus, reovirus. retrovirus, lentivirus, baculovirus, bacterial systems and eukaryotic systems such as CHO cells. Naked DNA and DNA liposome complexes may also be used. Specific proteins or peptides may be produced using molecular biology techniques or methods ("recombinant" proteins or peptides). Conventional techniques or methods used in recombinant molecular biology are described in, for example, Molecular Cloning: a Laboratory Manual 3 ^rd edition. Sambrook

and Russell. CSHL Press, Cold Spring Harbour, New York; Current Protocols in Molecular Biology, 2007 Ausubel et al editors. Wiley InterScience, New York; Current Protocols in Immunology, 2006 Coligan et al editors. Wiley InterScience, New York (all of which are incorporated herein by reference).

[00100] The term 'gene therapy' as used herein refers to the insertion of exogenous nucleic acid into cells to provide a specialized function such as increasing expression, or providing the means for expression in a cell where the gene is damages or dysfunctional. The exogenous nucleic acid, when expressed, provides the gene product, usually a protein that has normal or enhanced biological activity in the cell.

[00101] The term 'expression' as used herein refers to the transcription or translation of a gene or nucleic acid. The gene may be endogenous to the cell or animal, or may be supplied exogenously, such as in gene therapy.

[00102] The term 'vector' as used herein refers to an agent that transfers nucleic acid from one location to another. A vector may itself be an isolated nucleic acid such as a plasmid or artificial chromosome, or may be a virus. The nucleic acid may encode a gene of interest and may be operably linked to a control sequence or other regulatory sequence.

[00103] The term "operably linked" as used herein refers to a juxtaposition wherein the components described are in a relationship permitting them to function in their intended manner. For example, a control sequence operably linked to a coding sequence is ligated in such a way that expression of the coding sequence is achieved under conditions compatible with the control sequence.

[00104] A 'control sequence' as used herein comprises a regulatory sequence allowing expression of the heterologous gene and a signal for termination of transcription. The control sequence may be selected from those which are functional in mammalian cells. The control sequence may be derived from promoter sequences of eukaryotic genes. For example, it may be a promoter, or a fragment of a promoter,derived from the genome of a cell in which expression of the heterologous gene is to occur, preferably a pancreatic beta cell. With respect to promoters of eukaryotic genes, it may be a promoter, or fragment of a promoter, that functions in a ubiquitous manner (such as promoters of .beta-

actin, or tubulin) or, alternatively, a tissue-specific manner. In another example, the control sequence may be a promoter, or fragment of a promoter, that responds to specific stimuli, for example binding of a steroid hormone receptors. A viral promoter or fragment of a viral promoter may also be used, for example the Moloney murine leukemia virus long terminal repeat (MMLV LTR) promoter, a promoter of HSV, or a promoter or fragment of a promoter of a CMV gene such as VP 16.

[00105] In some embodiments, nucleic acid encoding abcal or a fragment thereof may be used to express the ABCAl protein in an appropriate cell in vitro, in vivo or ex vivo. Alternatively, the nucleic acid may be cloned into an expression vector suitable for production of useful quantities of ABCAl protein in vitro. The resulting ABCAl protein may be used in assays or screens to identify agonists of ABCAl activity, or may be administered to a cell in vitro, in vivo or ex vivo in combination with a protein transduction domain, as described below.

Gene therapy

[00106] Gene therapy approaches to enhancing ABCAl expression in pancreatic beta cells may be facilitated by the use of viral-mediated or synthetic gene delivery systems. Exogenous nucleic acid may be introduced into pancreatic beta cells in vivo, in vitro or ex vivo by conventional transformation or transfection techniques. "Transfection" or "transformation" as used herein refers to techniques and methods for introducing exogenous nucleic acid into a host cell. Such methods and techniques may include calcium phosphate or calcium chloride co-precipitation, electroporation, microinjection, viral mediated transfection or transfection with polymeric gene delivery systems such as poly-L- lysine, dendrimers, liposomes, or targeted polymers. Examples of polymeric gene delivery systems are known in the art, and are described in, for example, Anwer, K. et al., 2003. Crit Rev Ther Drug Carriers. 20:249-293. Methods for transfecting or transforming cells are known in the art, and are generally described in, for example, Ausubel, et al. Current Protocols in Molecular Biology, John Wiley & Sons, New York, N.Y., 1988-2007. Additionally, a multipart expression system suitable for large membrane- spanning proteins may be used, for example, as described in WO 03/093468. All of the preceding citations are incorporated herein by reference.

[00107] Pancreatic beta-cells may be transformed by viral-mediated introduction of exogenous nucleic acid in vitro, in vivo or ex vivo. A suitable vector may comprise nucleic acid encoding ABCAl or a fragment thereof and appropriate control sequences, preferably specific for expression in pancreatic beta-cells. Beta-cell specific expression may be driven by beta-cell specific control sequences such as the rat, human or mouse insulin promoter, the PDX-I promoter or the islet amyloid polypeptide promoter (IAPP). Examples of viral vectors may include adenovirus, adeno- associated virus, retrovirus, lentivirus, reovirus, baculovirus or coxsackievirus. Suitable viral vectors may be modified for targeting terminally differentiated or non-dividing cells or specific cell types. Examples of suitable viral vectors may include, for example, AAV2, AA V5 or AAV8 (Wang et al., 2002. Hum Gene Ther. 15:405-13; Loiler et al., 2003. Gene Therapy 10:1551-1558; Rehman et al , 2005. Gene Therapy 1-11; all of which are incorporated herein by reference).

[00108] ABCAl protein may be delivered to pancreatic beta-cells in vitro, in vivo or ex vivo by protein transduction. The ABCAl protein may be fused to a cationic cell- penetrating peptide comprising a protein transduction domain. Examples of protein transduction domains and delivery techniques may be found in, for example, Wadia, J.. S. and Dowdy, S. F. 2002. Curr. Opin Biotechnol. 13:52-56 (incorporated herein by reference) and references therein.

[00109] Pancreatic beta-cells that have been transformed or transfected in vitro or ex vivo may be transplanted as an autograft or allograft into a host mammal. Techniques for selective isolation, handling and transplantation of pancreatic islets may be found, for example, in Shapiro et al. 2000. NEJM 343:230-238 or Stock, P.G. and Bluestone, J.A. 2004. Ann. Rev. Med 55:133-156 (incorporated herein by reference).

Kits

[00110] In accordance with another aspect of the invention, there is provided a kit for screening for an ABCAl agonist.. The kit may include a composition comprising a compound known to have ABCAl biological activity as a control, together with instructions for using a compound in a screen for an ABCAl agonist. The instructions may

include, for example, dose concentrations, dose intervals, preferred administration methods, preferred ABCAl biological activity assays (direct or indirect), or the like.

Methods

Cell Culture

[00111 ] INS-I cells (rat beta-cell line) were cultured in RPMI 1640 media

(Invitrogen) supplemented with pencillin, streptomycin, HEPES, sodium pyruvate, beta mercaptoethanol, and with or without 5% fetal calf serum. Cells were treated with lμM rosiglitazone in ethanol or control (ethanol alone) for 48 hours in media without serum. Western blotting was performed as previously described (Wellington, C. L. et al., 2002. J. Lipid Res. 43, 1939-1949)

Generation of ABCAl beta-cell specific knock-out mice (Abcal+/P- and Abcal-P/-P)

[00112] ABCAl floxed mice have been described by Timmins 2005 (Timmins, J.

M. et al., 2005. J Clin. Invest 115, 1333-1342). ABCAl floxed mice were crossed to mice transgenic for Cre under the control of the rat inulin promoter obtained from the Jackson Labs, Inc (Postic C et al. 1999. J. Biol Chem 274:305-315). Heterozygous offspring were intercrossed to generate the +/+, +/- and -/- mice used to study. All studies were performed using age-matched littermate controls.

ABCAl -/- mice

[00113] ABCAl global knock out mice (ABCAl-/-) were obtained from Jackson

Labs Bar Harbour, Maine. ABCAl-/- mice were generated on a mixed C57B16/J 129/Sv strain (McNeish, J et al., 2000. Proc. Nat. Acad. Sci 97:4245-4250.).

Mouse maintenance

[00114] Mice received a standard laboratory chow, or a high-fat diet (Harlan Teklad TD 93075) with or without 0.36% rosiglitazone (to yield a dose of 30 mg/kg/d based on measured food intake), as indicated. All mice were maintained under 12 hour light/dark

cycles and received standard lab chow and water ad libitum. All animal procedures were approved by the Animal Care Committee of the University of British Columbia.

Glucose Tolerance Test

[00115] Mice fasted for 4 hours were administered 2 mg / kg body weight anhydrous glucose by intraperitoneal injection. Blood glucose concentration was determined using a ONE TOUCH Ultra glucometer at 0, 15, 30, 60 and 90 minutes. Blood was collected at time 0 and 15 minutes and serum insulin concentration was determined by ELISA (Alpco).

Insulin Tolerance Test

[00116] Mice fasted for 4 hours were administered 1 U / kg body weight recombinant human insulin (Novo Nordisk) by intraperitoneal injection. Blood glucose concentration was determined using a ONE TOUCH Ultra glucometer at 0, 15, 30, 60 and 90 minutes. To determine insulin secretion in vivo, we injected 4-h-fasted mice with 3 mg/kg glucose or 0.3 mg/kg L-arginine and measured plasma insulin by ELISA (Crystal Chem).

Islet isolation and ex- vivo Glucose-Stimulated Insulin Release

[00117] Mouse islets were isolated using the method of Marzban 2005, or Plesner.

(Marzban, L. et ah, 2005. . Endocrinology 146, 1808-1817; Plesner et al 2005. Diabetes 54:2533-2540). After an overnight culture, approximately 40 islets per well were incubated in media comprising Kreb's Ringer bicarbonate buffer (KRB-BSA) containing 1.67 mM glucose for 2 h, then incubated with buffer containing 1.67 mM glucose, 20 mM glucose or 1.67 mM glucose plus 25 mM KCl.. After a one hour incubation supernatant was collected and islets were lysed in buffer containing 50 mM Tris-HCl (pH 8.0), 150 mM NaCl, 0.02% sodium azide, 0.1% SDS, 1% NP-40, 0.5% sodium deoxycholate and complete protease inhibitor (Roche), or 1 M glacial acetic acid. Insulin concentration in supernatant and cell lysate was deterimed by ELISA (Alpco). Glucose-stimulated insulin release is expressed as the fraction of total insulin released into media and is the mean of 3 wells containing islets from 3 mice per genotype. Insulin secretion in isolated mouse islets was also studied using the perifusion technique, essentially as described by Johnson 2003

(Johnson JD et al 2003. J. Clin Invest 111:1147-1160). Temperature- and CO2-controlled perifusion columns were loaded with B 150 size-matched islets from a given mouse and allowed to equilibrate to a 0.3 ml/min perifusion flow for 1 h, after which islets were subjected to a stepwise increase from 3 mM to 20 mM glucose. We assayed insulin in the samples using Linco's Rat Insulin RIA Kit. Data are normalized to the pretreatment values (first seven time points) to control for uneven loading of columns. Area-under-curve values were calculated as the cumulative percent pretreatment.

Plasma Lipid Analysis

[00118] Plasma cholesterol and HDL cholesterol (on polyethylene glycol- precipitated plasma) by enzymatic assay according to the manufacturer's instructions (Thermo Electron Corporation). Cholesterol levels were determined by gas-liquid chromatography (Rudel et al 1998. Arterioscler.Thromb. Vase. Biol 18:1818-1827) and tissue triglyceride content by enzymatic assay (Carr et al 1993. clin biochem 26:39-42). Plasma free fatty acids were determined by NEFA C kit (Wako), plasma b- hydroxybutyrate and Cortisol levels by enzymatic assay (Stanbio and Alpco).

Real-time PCR, western blotting and immunofluorescence.

[00119] Real-time PCR was performed as described (Brunham et al 2006. J.

Clin. Invest 116: 1052-1062). Briefly, we extracted total RNA from isolated islets using the RNeasy kit (Life Technologies) and reverse-transcribed Dnase treated RNA using Superscript II (Life Technologies). We used Rnase- treated DNA for real-time PCR using SYBR Green PCR Master Mix (Applied Biosystems) in an ABI Prism 7700 Sequence Detection System. GAPDH was used as the invariant control. mRNA levels in control mice were arbitrarily set at 1.

[00120] Western blotting was performed as previously described (Wellington et al 2002. Lab. Invest. 82:273-283). Briefly, tissues were homogenized and sonicated in 20 mM HEPES, 5 mM KCl, 5 mM MgC12, 0.5% (vol/vol) Triton X-100 and complete protease inhibitor (Roche). Protein concentration was determined by the Lowry assay. We separated equivalent amounts of total protein by SDS-PAGE, transferred these to

polyvinyldene difluoride (PVDF) membranes and probed them with antibodies to Abcal (Wellington, supra) or antibodies to GAPDH (Chemicon).

[00121] Immunofluorescence was performed with antibodies to insulin and glucagon (DAKO) on paraffin-embedded sections. Immunofluorescence for Abcal was performed as previously described (Brunham et al 2006. J. Clin Invest 116:1052-1062) on 3-month-old mice. For b-cell mass measurements, the percentage of insulin-positive surface area was determined from ten evenly spaced slides per pancreas. The mean insulinpositive area of the ten slides was then multiplied by pancreatic weight to estimate b-cell mass. We performed filipin (Sigma) staining on cryostat sections as previously described (Heaps et al 2005. Am J Physiol Heart Circ. Physiol 288:H568-H576) and counterstained with propidium iodide (Sigma). Statistical analysis. Data are presented as mean ± s.e.m. Differences between groups were calculated by Student's t-test (for two groups) or one-way analysis of variance (ANOVA) with Newman- Keuls post-hoc test (for three groups). A P- value of 0.05 was considered significant.

[00122] ABCGl mRNAs were quantified as described (Brunham et al 2007. Nature

Medicine 13:340-347).

Human studies

[00123] Oral glucose tolerance tests were performed according to standard clinical chemistry protocols. Following an overnight fast, subjects received an oral 75 g glucose challenge. A hand vein was cannulated retrogradely and kept in a thermoregulated box at 55 ⁰ C to obtain arterialized blood samples. Blood was collected at -15, 0, 30, 60 and 120 min in fluoride containing collection tube to inhibit glycolysis and the samples were centrifuged immediately after collection. Blood glucose levels were determined using a high volume chemistry analyzer. Three patients heterozygous for the C1477R mutation were studied along with 3 unaffected relatives, matched for age, gender and body mass index.

Example 1 ABCAl is expressed in beta cells of the endocrine pancreas

[00124] Murine pancreas, specific cell types and INS-I cell lysates were immunoblotted and probed using a polyclonal antibody directed against ABCAl, and against GAPDH which served as a loading control (Figure 1). The majority of pancreatic tissue is exocrine, and beta cells represent only a small fraction of the endocrine islet tissue. ABCAl expression is undetectable in whole pancreas lysates and islet cell lysates isolated from an ABCAl knockout mouse, but is strongly detected in islet cell lysates isolated from a wildtype littermate control and INS-I cells. Detection of ABCAl in normal mouse islets and INS-I cell lysates but not abcal-/- mouse islets or whole pancreas confirms the specificity of the polyclonal antibody used.

Example 2

ABCAl expression is regulated by PPARgamma agonists

[00125] Rat INS-I cells were treated as described with 1 uM rosiglitazone in ethanol - a PPAR gamma agonist - or vehicle control (ethanol only). Cell lysates were prepared as described and immunoblotted using an ABCAl -specific polyclonal antibody, or a GAPDH- specific polyclonal antibody as a loading control. ABCAl expression in INS-I cells is induced in the presence of rosiglitazone (Figure 2).

Example 3 ABCAl expression is required for glucose tolerance

[00126] Mice lacking expression of the abcal gene (abcal-/-) demonstrate impaired glucose tolerance. Mice lacking abcal are hyperglycemic during glucose tolerance testing (Figure 3a), but have normal insulin sensitivity as assessed by intraperitoneal insulin tolerance testing (Figure 3b). The acute insulin response to glucose (change in serum insulin divided by change in blood glucose 15 minutes after IP glucose challenge) is decreased in abcal-/- mice (Figure 3c) (*p<0.01).

[00127] Glucose stimulated insulin secretion is lacking in islets isolated from abcal-

/- mice. Isolated islets cultured under high and low glucose conditions, and insulin in islets and secreted into the media was determined by ELISA (Figure 4). Islets from control mice increase the secretion of insulin in response to high glucose, while this effect is abolished in islets from abcal-/- mice.

[00128] The impaired glucose tolerance phenotype may be due to either a beta-cell defect or impaired insulin sensitivity. The normal insulin sensitivity observed in both

control and knockout mice indicates that this impaired glucose tolerance is due to a beta- cell defect. The ratio of change in serum insulin to change in serum glucose illustrates that the abcal-7- mice are defective in glucose-stimulated insulin secretion in the beta-cells, and that ABCAl expression is essential for normal insulin secretion ex vivo.

Example 4

ABCAl-BSKO mice have normal plasma cholesterol levels

[00129] Abcal beta-cell specific knockout mice (ABCAl-BSKO: Abcal-P/-P,

Abcal +/-P) were generated and characterized. Mice lacking abcal expression specifically in beta-cells have total cholesterol (Figure 5a) and HDL cholesterol (Figure 5b) levels indistinguishable from littermate controls.

[00130] The finding that there is no effect on lipid levels illustrates that beta-cell

ABCAl has no effect on plasma HDL metabolism. Beta-cells in the ABCAl-BSKO mice are exposed to a much higher concentration of circulating plasma cholesterol than beta- cells in the global knock-out, which has very low circulating plasma cholesterol levels.

Example 5

Glucose tolerance is altered in ABCAl-BSKO heterozygote mice

[00131] Mice deficient in ABCAl expression specifically in the beta-cells (Abcal -

P/-P) display profound hyperglycemia during glucose tolerance testing, indicating that beta-cell ABCAl is crucial for beta-cell function (Figure 6). ABCAl-BSKO heterozygote mice have profoundly impaired glucose tolerance. This is especially significant as these are heterozygotes. The ABCAl-BSKO heterozygote mice also demonstrate greater impairment of glucose tolerance than the abcal-/- mice, illustrating a crucial role for ABCAl in beta cell function in vivo. The greater impairment of glucose tolerance in the heterozygote mice may relate to the higher plasma cholesterol levels observed in the ABCAl-BSKO mice.

Example 6

Glucose tolerance is altered in ABCAl-BSKO homozygote mice on a high-fat diet. [00132] Abcal -P/-P mice and controls were fed either a high-fat (50% fat by weight) diet alone or in combination with 30 mg/kg/day rosiglitazone as diet admixture. Glucose tolerance testing as performed prior to the initiation of the feeding and after 2

weeks on the diet. Figure 7B shows glucose tolerance in wildtype mice prior to feeding, and after 2 weeks of feeding high-fat diet or high-fat diet plus rosiglitazone. High-fat feeding resulted in significantly impaired glucose tolerance in wildtype mice, and concurrent rosiglitazone treatment prevented this impairment in glucose tolerance. Abcal- P/-P mice fed a high-fat diet also developed worse glucose tolerance relative to prior feeding, but rosiglitazone treatment had no effect on glucose tolerance in these mice. The total area under the glucose tolerance curve (AUC _ipgtt ) is therefore increased in Abcal+/+ mice fed the high-fat diet, and returned to levels not significantly difference from baseline by rosiglitazone treatment (Figure 7D). In contrast, the AUCi _Pgtt for Abcal-P/-P was significantly increased by high-fat feeding, but did not change in response to rosiglitazone (Figure 7D).

[00133] Figure 7F shows fasting insulin levels in Abcal+/+ and Abcal-P/-P mice on chow, and after high-fat or high-fat plus rosiglitazone feeding. Fasting insulin levels were significantly increased in Abcal+/+ mice fed a high-fat diet, and rosiglitazone returned fasting insulin to baseline levels. Abcal-P/-P mice showed no change in fasting insulin levels on either treatment. Figure 7F shows insulin sensitivity as determined by i.p. insulin tolerance test. Rosiglitazone treatment improved insulin sensitivity to a similar extent in both genotypes. Therefore, the difference in glucose tolerance observed could not be explained by a difference in the effect of rosiglitazone on insulin sensitivity between Abcal+/+ and Abcal-P/-P mice. Plasma cholesterol levels were increased in both genotypes by high-fat feeding, and did not differ significant between genotypes.

[00134] Pancreatic islets from Abcal+/+ and Abcal-P/-P mice fed either a high-fat or high-fat plus rosiglitazone diet were also examined. Abcal+/+ mice fed a high-fat diet had disaggregated islets with alpha cells within the islet (Figure 7H). Rosiglitazone treatment improved islet morphology in these mice (Figure 71). Disaggregated islets were also observed in Abcal-P/-P mice receiving a high-fat diet (Figure 7J), as well as on rosiglitazone (Figure 7K). Rosiglitazone treatment therefore resulted in qualitative morphological changes to islets in Abcal+/+ mice, but not in Abcal-P/-P mice.

[00135] These data indicate that the beneficial effect of rosiglitazone on beta-cell structure and function is at least partially dependent on beta-cell ABCAl. These data

further indicate that compounds which specifically upregulate beta-cell ABCAl would be expected to produce the beneficial effects of TZDs on beta-cell function, while avoiding the undesirable effect of these drugs on adiposity.

Example 7

Characterization of homozygous ABCAl-BSKO mice

[00136] Intraperitoneal glucose tolerance testing was performed on homozygous control (Abcal ^+/+ ), heterozygous (Abcal ^+/'p ) and homozygous knockout (Abcal ^{~p/ p} ) mice. Heterozygous 2-month old (Figure 8A) and 4-month old (Figure 8B) and homozygous 2-month old (Figure 8C) and 4-month old (Figure 8D) ABCAl beta-cell specific knock-out mice demonstrate progressive and marked impairment in glucose tolerance in the absence of beta-cell ABCAl . Heterozygous Abcal +/-P mice display significantly impaired glucose tolerance— The progressive nature of glucose intolerance in these mice is consistent with increased islet cholesterol in the absence of beta-cell ABCAl leading to impairment in insulin secretion that worsens with time.

[00137] Plasma insulin levels during the glucose tolerance tests are significantly lower in Abcal -P/-P mice 15 minutes following the i.p. glucose challenge (Figure 9) indicating that absence of beta-cell ABCAl impairs glucose stimulated insulin secretion. A first-phase insulin response 2 minutes following a glucose (3 mg/g) challenge was also examined. Plasma insulin levels are lower in Abcal -P/-P mice compared to controls at 2 minutes following the glucose challenge (Figure 10, P<0.01), indicating that ABCAl is essential for first-phase glucose stimulated insulin secretion in vivo. Similarly, insulin secretion in response to L-arginine (0.3 mg/g) is significantly lower in Abcal-P/-P mice compared to controls (Figure 11, P<0.05).

[00138] The reduction in insulin secretion observed in Abcal-P/-P mice could be due to a reduction in β-cell mass, an impairment of β-cell function, or a combination of both. Quantitative histo-morphometrical analysis on evenly-spaced pancreas sections from Abcal -P/-P mice and controls, and observed no difference in β-cell mass among genotypes (Figure 12a). Similarly, gross morphology of islets is not altered (Figure 12b), nor is the overall abundance or size of islets (Figure 12c). The reduction in insulin secretion

observed in vivo can therefore not be explained by a reduction in β-cell mass, and more likely reflects an impairment in islet function.

[00139] Figure 13a shows that insulin secretion in response to 24 mM glucose is significantly blunted in islets lacking Abcal. Incubation of islets with KCl, which directly depolarizes the cell membrane, leads to significant stimulation of insulin secretion in wildtype but not Abcal-P/-P islets (Figure 13a). This finding suggests that the impairment in insulin secretion in Abcal -deficient islets lies downstream of nutrient sensing and secondary signal generation by the β-cell. Insulin content is increased in islets lacking Abcal (Figure 13b), as well as in whole pancreata from Abcal-P/-P mice (Figure 13c), while static insulin mRNA levels are not significantly altered (Figure 13d), consistent with a defect in the exocytosis of insulin-containing secretory granules. Examination of the dynamics of glucose-stimulated insulin secretion by perifusion of isolated islets demonstrates a defect in first-phase glucose-stimulated insulin release in mice lacking β- cell Abcal (Figure 13e,f).

[00140] Cellular cholesterol levels in sections of pancreas from Abcal ^~pι~p mice was examined using the cholesterol probe filipin. Islets from Abcal ^~PI~P mice displayed a marked increase in the intensity of filipin staining (Figure 13g,h), consistent with our quantitative analysis of islet cholesterol content. Filipin staining appeared to be specifically enhanced at the plasma membrane of islet cells from AbcaF ^p/~p mice. These data suggest that defective Abcal -mediated efflux from β-cells leads to cholesterol accumulation at the plasma membrane which is associated with impaired insulin secretion.

Example 8

ABCAl Mediates cholesterol Efflux from Islets

[00141] Islet cells were isolated from littermate control (Abcal f./fl) and beta-cell specific Abcal knockout (Abcal -P/-P) mice, and the cholesterol efflux determined. Islets isolated from control mice demonstrate a significant (p=0.03) efflux of cholesterol to apoAl relative to experimental control. Islets from knockout mice do not demonstrate a significant difference (p=0.2) in cholesterol efflux to apoAl relative to experimental control.

Example 9

Role of beta-cell ABCAl in Islet Cholesterol Metabolism

[00142] Figure 15a shows that absence of beta-cell ABCAl results in a significant increase in the concentration of islet total and free cholesterol and cholesterol ester (P<0.05). Islet TG levels are not significantly altered. Measurement of islet cholesterol content in mice with global deletion of ABCAl reveals no increase in islet total or free cholesterol or cholesterol ester levels.

[00143] Figure 15b shows the the expression of various sterol-sensitive mRNAs in isolated islets. The expression of HMG-CoA reducatase (Hmgcr) and LDL receptor (LdIr) mRNA, both of which are downregulated by cholesterol, is significantly reduced in islets lacking ABCAl. Abcgl mRNA is readily detectable in isolated mouse islets, where it has not previously been reported to be expressed, and that the expression of Abcgl is increased in islets lacking beta-cell ABCAl (Figure 15b). This suggests that ABCGl, may increase to compensate for the loss of beta-cell ABCAl and the resultant increased intracellular cholesterol content. ABCGl may be important for preventing excess islet lipid accumulation, and that it would therefore be beneficial to upregulate the expression of these transporters to preserve islet function.

Example 10

Upregulation of beta-cell ABCAl using Thiazolidinediones Lowers Islet Cholesterol and Prevents Glucose Intolerance

[00144] The role of beta-cell ABCAl in the response to treatment with a representative thiazolidinedione, rosiglitazone was examined. We fed Abcal-P/-P and control mice a high-fat diet (30% fat by weight) alone or in combination with 30 mg/kg/day rosiglitazone. Glucose tolerance tests were performed prior to and 2 weeks after initiation of high-fat feeding. High-fat feeding resulted in significantly impaired glucose tolerance in wildtype mice, and rosiglitazone restored glucose tolerance (Figure 7 B). High-fat feeding also worsened glucose tolerance in Abcal-P/-P mice, but rosiglitazone had no effect on glucose tolerance in these animals (Figure 7C). The total area under the IPGTT curve (AUCipgtt) is increased in Abcal+/+ mice fed the high-fat diet, and returns to levels not

significantly different from that of chow-fed mice following rosiglitazone (Figure 7D). In contrast, the AUCipgtt for Abcal-P/-P mice is significantly increased by high-fat feeding, and does not improve in response to rosiglitazone (Figure 7D). Comparisons between regular, high fat (HF) and high fat + rosiglitazone in wild type (+/+) and Abcal -P/-P mice (-P/-P) demonstrate no significant difference in mice fed a regular vs a high fat + rosiglitazone diet (Table 1).

Table 1 Treatment comparisons between regular, high fat (HF) and high fat + rosiglitazone (HF Rosi) in wild type (+/+) and Abcal -P/-P mice (-P/-P)

[00145] Fasting insulin levels are significantly increased in Abcal +/+ mice fed a high-fat diet and return to baseline levels by rosiglitazone. Abcal-P/-P mice show no change in fasting insulin levels in response to either high-fat diet or rosiglitazone Insulin sensitivity is improved to a similar extent in both genotypes in response to rosiglitazone, indicating that the difference in glucose tolerance observed between Abcal +/+ and Abcal - P/-P mice can not be explained by different effects of rosiglitazone on insulin sensitivity.

[00146] Treatment with rosiglitazone lowers the free cholesterol concentration in islets isolated from wildtype mice (Figure 16). Free cholesterol levels are higher in islets isolated from Abcal -P/-P mice and are unchanged by rosiglitazone. Notably, rosiglitazone lowers islet TG concentration equivalently in both genotypes of mice. These data show that a significant part of the beneficial effect of rosiglitazone is due to upregulation of ABCAl -dependent cholesterol efflux from islets which protects against diet-induced impairment in glucose tolerance.

Example 11

Validation of the Role of ABCAl in Islet Function in Humans

[00147] To investigate whether ABCAl has a role in islet function in humans detailed measurements of glucose homeostasis were performed in humans with validated, disease-causing mutations in the ABCAl gene (Figure 17). Three patients heterozygous for the C1477R mutation were studied along with 3 unaffeted relatives, matched for age, gender and body mass index. Compared to unaffected relatives (controls), humans with ABCAl mutations (heterozygotes or "hets") have higher blood glucose levels and lower insulin levels after a 75 g oral glucose challenge. The acute insulin response to glucose, a surrogate measure of islet function, shows a significant reduction in the subjects carrying ABCAl mutations (Figure 20).

[00148] All citations are incorporated herein by reference.

[00149] One or more currently preferred embodiments have been described by way of example. It will be apparent to persons skilled in the art that a number of variations and modifications can be made without departing from the scope of the invention as defined in the claims.