Identification of a Novel Selective Peroxisome Proliferator- Activated Receptor Agonist, 2-Methyl-2-(4-{3-[1- (4-methylbenzyl)-5-oxo-4,5-dihydro-1H-1,2,4-triazol- 3-yl]propyl}phenoxy)propanoic Acid (LY518674), That Produces Marked Changes in Serum Lipids and Apolipoprotein A-1 Expression

Low high-density lipoprotein-cholesterol (HDL-c) is an important risk factor of coronary artery disease (CAD). Optimum therapy for raising HDL-c is still not available. Identification of novel HDL-raising agents would produce a major impact on CAD. In this study, we have identified a potent (IC50 24 nM) and selective peroxisome proliferator-activated receptor (PPAR ) agonist, 2-methyl-2-(4-{3-[1-(4-methylbenzyl)-5-oxo4,5-dihydro-1H-1,2,4-triazol-3-yl]propyl}phenoxy)propanoic acid (LY518674). In human apolipoprotein A-1 (apoA-1) transgenic mice, LY518674 produced a dose-dependent increase in serum HDL-c, resulting in 208 15% elevation at optimum dose. A new synthesis of apoA-1 contributed to the increase in HDL-c. LY518674 increased apoA-1 mRNA levels in liver. Moreover, liver slices from animals treated with LY518674 secreted 3to 6-fold more apoA-1 than control liver slices. In cultured hepatocytes, LY518674 produced 50% higher apoA-1 secretion, which was associated with increase in radiolabeled methionine incorporation in apoA-1. Thus, LY518674 is a potent and selective PPAR agonist that produced a much greater increase in serum HDL-c than the known fibrate drugs. The increase in HDL-c was associated with de novo synthesis of apoA-1. Low plasma high-density lipoprotein-cholesterol (HDL-c) and its associated apolipoprotein A-1 (apoA-1) are known risk factors for coronary artery disease (CAD) (Gordon and Rifkind, 1989). Approximately 40 to 50% of the patients with established CAD have cholesterol levels that are considered in the normal range ( 200 mg/dl). In a large number of these patients, low HDL-c is a common risk factor (Genest et al., 1991; Rubins et al., 1995). Moreover, low HDL-c is associated with a substantial increase in risk of CAD in diabetics. Several pharmacological strategies are currently being investigated to raise HDL-c including the inhibition of cholesterol ester transfer protein (CETP), activation of peroxisome proliferator-activated receptors (PPARs) and induction of ABCA1 (Rader, 2002; Brewer, 2004). The nuclear hormone receptor PPAR subfamily plays a major role in the regulation of lipids and glucose metabolism. PPAR , PPAR , and PPAR , encoded by three distinct genes, are members of the PPAR subfamily (Willson et al., 2000). PPAR is predominantly expressed in tissues that metabolize high amounts of fatty acids such as liver, kidney, heart, and muscle (Braissant et al., 1995; Auboeuf et al., 1997). High level of PPAR is expressed in adipose tissues where its activation is associated with adipocyte differentiation and insulin sensitivity (Tontonoz et al., 1995). PPAR is widely expressed in tissues Article, publication date, and citation information can be found at http://molpharm.aspetjournals.org. doi:10.1124/mol.105.010991. ABBREVIATIONS: HDL-c, high-density lipoprotein-cholesterol; CAD, coronary artery disease; LDL-c, low-density lipoprotein-cholesterol; VLDL, low-density lipoprotein; apoA-1, apolipoprotein A-1; CETP, cholesterol ester transport protein; PPAR, peroxisome proliferator-activated receptor; PPRE, peroxisome proliferator-activated receptor-response element; LY518674, 2-methyl-2-(4-{3-[1-(4-methylbenzyl)-5-oxo-4,5-dihydro-1H1,2,4-triazol-3-yl]propyl}phenoxy)propanoic acid; FPLC, fast protein liquid chromatography; PAGE, polyacrylamide gel electrophoresis; CHAPS, 3-[(3-cholamidopropyl)dimethylammonio]-1-propanesulfonic acid; FPLC, fast-performance liquid chromatography; LY487, 2-methyl-2-(4-{3-[1(4-methyl-benzyl)-5-oxo-4-propyl-4,5-dihydro-1H-[1,2,4]triazol-3-yl]-propyl}-phenoxy)-propionic acid. 0026-895X/05/6803-763–768$20.00 MOLECULAR PHARMACOLOGY Vol. 68, No. 3 Copyright © 2005 The American Society for Pharmacology and Experimental Therapeutics 10991/3046847 Mol Pharmacol 68:763–768, 2005 Printed in U.S.A. 763 at A PE T Jornals on July 1, 2017 m oharm .aspeurnals.org D ow nladed from (Braissant et al., 1995). Recent studies suggest that PPAR also plays a role in lipid metabolism, presumably through mechanisms different from PPAR and PPAR (Oliver et al., 2001). Several synthetic ligands of PPAR , including fenofibrate, gemfibrozil, bezafibrate, and clofibrate have been developed as therapy for dyslipidemia (Heller and Harvengt, 1983; Robins, 2001). However, fibrates are neither potent nor selective PPAR agonists. In rodents, fibrates lower HDL-c by reducing apoA-1 synthesis and also by inducing changes in HDL-modifying enzymes such as lecithin:cholesterol acyltransferase (Schoonjans et al., 1995, 1996). In humans, fibrates reduce triglycerides and produce a moderate elevation of HDL-c (Robins, 2001). Creation of a human apoA-1 transgenic mouse has provided a useful model for evaluation of HDL-c–raising agents and the investigation of the mechanism of HDL-c elevation. In this model, the human apoA-1 gene expression is driven by its homologous promoter, which includes the PPAR response element (PPRE). Therefore, fibrates produce a decrease in mouse apoA-1 and an increase in human apoA-1 expression (Berthou et al., 1996, Vu-Dac et al., 1998). Using this model, we show that a highly potent and selective PPAR agonist, LY518674, produced a large increase in HDL-c. Using ex vivo studies with liver from transgenic animals dosed with LY518674 and isolated hepatocytes, we have shown that a de novo induction of apoA-1 synthesis is an important mechanism contributing to the HDL-c elevation. Thus, the highly potent and selective PPAR agonist modulating HDL-c by enhancing apoA-1 synthesis may represent important pharmacological means for the treatment of patients with dyslipidemia, particularly those with low HDL-c. Materials and Methods LY518674 and LY487, shown in Fig. 1, were synthesized at Eli Lilly and Company. Fenofibrate was obtained from Sigma (St. Louis, MO) . Animal Dosing. Human apoA-1 transgenic mice from Jackson ImmunoResearch Laboratories Inc. (West Grove, PA) were maintained on a chow diet. Animals were dosed orally with vehicle [1% (w/v) carboxymethylcellulose and 0.25% Tween 80], fenofibrate, or LY518674 for 1 week. Three hours after the final dose, animals were sacrificed by CO2 asphyxiation, blood was drawn by cardiac puncture, and livers were excised. Blood samples were allowed to clot at room temperature, and serum was prepared by centrifugation. Fast-Performance Liquid Chromatography. Lipoproteins were separated by FPLC and quantified using an in-line detection system as described by Kieft et al. (1991). Pooled serum sample (25 l) was applied to a Superose 6HR 10 30 cm column (Amersham Biosciences, Piscataway, NJ) and eluted with phosphate-buffered saline, pH 7.4, containing 5 mM EDTA at 0.5 ml/min. The area under the curve corresponding to the elution of very low density lipoprotein (VLDL), LDL, and HDL was calculated using PerkinElmer Turbochrom (version 4.12F12) software (PerkinElmer Life and Analytical Sciences, Boston, MA). Serum cholesterol and triglycerides were determined by standard enzymatic methods. Liver Slicing. Each liver lobe was cored and then sliced using Krebs-Henseleit buffer supplemented with 20 mM glucose and mannitol. The slices were pooled and four slices per vial were incubated in hepatocyte maintenance medium (Cambrex Bio Science Walkersville, Walkersville, MD) for 2 h in a shaking water bath maintained at 29°C under 95% O2 and 5% CO2. Aliquots of the medium were collected at the indicated time intervals. RNA Extraction and Northern Blot Analysis. Total RNA was extracted with RNeasy reagent (QIAGEN, Valencia, CA), electrophoresed on a 1% agarose-formaldehyde gel, and then pressure-blotted (PosiBlot 30–30 Pressure Blotter; Stratagene, La Jolla, CA) onto BrightStar-Plus nylon membrane (Ambion, Austin, TX). Membranes were UV-cross-linked and hybridized in NorthernMax Prehyb/Hyb buffer for 18 h at 42°C with a P-labeled cDNA probe for apoA-1. Membranes were washed with NorthernMax High Stringency wash solution before exposure to Hyperfilm MP (Amersham Biosciences UK, Ltd., Little Chalfont, Buckinghamshire, UK). Expression levels were normalized with 18S RNA. Western Blotting. Protein concentration in the liver slice incubation medium collected at 30, 60, and 120 min was determined using Bradford reagent (Bio-Rad, Hercules, CA). Equal amounts of protein (235 g) were loaded onto a 4 to 20% Tris-Gly gel (Invitrogen, Carlsbad, CA), separated by SDS-polyacrylamide gel electrophoresis (PAGE), and transferred to polyvinylidene difluoride membrane. The membranes were blocked by incubation with 5% nonfat dry milk in Tris-buffered saline with Tween 20 and then blotted with human apoA-1 monoclonal antibody (Biodesign International, Kennebunk, ME). Bands were visualized using ECL detection system (Amersham Biosciences) and quantified by scanning densitometry. ApoA-1 Production in Monkey Hepatocytes. Cryopreserved hepatocytes from cynomologus monkeys (Cambrex Bio Science Walkersville) were cultured in hepatocyte maintenance medium for 24 h. Cells were then treated with the compounds. After 48 h, culture supernatants were collected and subjected to SDSPAGE followed by Western blotting. For quantification, the apoA-1 bands were normalized to transferrin band in each sample. For [S]methionine labeling of proteins, cells were incubated in methionine-free RPMI 1640 medium for 24 h in the presence or absence of the compound. [S]methionine (30 Ci) was then added, and the cultures were further incubated for 48 h. ApoA-1 was immunoprecipitated from the supernatants (150 g of protein) using monoclonal antibodies coupled to protein G-Sepharose (Biodesign International). The immunoprecipitated fractions were washed and electrophoresed. The bands were quantified by scanning densitometry. Binding Assay. DNA-dependent binding was performed using Scintillation Proximity Assay (Amersham Biosciences). Biotinylated oligonucleotide 5 -TAATGTAGGTAATAGTTCAATAGGTCAAAGGG-3 was used for binding receptor dimers to Yttrium silicate streptavidin-coated beads in the presence of the H-labeled corresponding ligand for each receptor. Competition binding reactions were carried out in 10 mM HEPES buffer, pH 7.8, 80 mM KCl, 0.5 mM MgCl2, 1 mM dithiothreitol, 0.5% CHAPS, and 14% glycerol using 2.5 g each of PPAR , PPAR , or PPAR and retinoid X receptor , 0.1 nM to 10 M concentrations of compounds, and 30,000 cpm of the corresponding labeled ligand. Cotransfection Assay. Cotransfection assays were performed in CV-1 cells using calcium phosphate coprecipitation method as described previously (Mukherjee et al., 1997). The reporter plasmid consisted of the acyl-CoA oxidase PPRE and thymidine kinase promoters upstream of the luciferase reporter cDNA. For PPAR , interference by endogenous PPAR was eliminated by using a GAL4 chimera in which the DNA binding domain of the transfected PPAR was replaced by that of GAL4 and the GAL4 response element was used in place of the acyl-CoA oxidase PPRE. After an overnight incubation, transfected cells were trypsinized and plated in 96-well Fig. 1. Structures of LY518674 and LY487. 764 Singh et al. at A PE T Jornals on July 1, 2017 m oharm .aspeurnals.org D ow nladed from dishes in Dulbecco’s modified Eagle’s medium containing 10% charcoal-stripped fetal bovine serum. After 6 h, cells were exposed to a 0.1 nM to 10 M concentration of test compounds. Cotransfection efficacy was determined relative to reference agonists for each receptor.