Specific Activation of the Nuclear Receptors PPARg and RORA by the Antidiabetic Thiazolidinedione BRL 49653 and the Antiarthritic Thiazolidinedione Derivative CGP 52608

The thiazolidinedione BRL 49653 and the thiazolidinedione derivative CGP 52608 are lead compounds of two pharmacologically different classes of compounds. BRL 49653 is a high affinity ligand of peroxisome proliferator-activated receptor g (PPARg) and a prototype of novel antidiabetic agents, whereas CGP 52608 activates retinoic acid receptor-related orphan receptor a (RORA) and exhibits potent antiarthritic activity. Both receptors belong to the superfamily of nuclear receptors and are structurally related transcription factors. We tested BRL 49653 and CGP 52608 for receptor specificity on PPARg, RORA, and retinoic acid receptor a, a closely related receptor to RORA, and compared their pharmacological properties in in vitro and in vivo models in which these compounds have shown typical effects. BRL 49653 specifically induced PPARg-mediated gene activation, whereas CGP 52608 specifically activated RORA in transiently transfected cells. Both compounds were active in nanomolar concentrations. Leptin production in differentiated adipocytes was inhibited by nanomolar concentrations of BRL 49653 but not by CGP 52608. BRL 49653 antagonized weight loss, elevated blood glucose levels, and elevated plasma triglyceride levels in an in vivo model of glucocorticoidinduced insulin resistance in rats, whereas CGP 52608 exhibited steroid-like effects on triglyceride levels and body weight in this model. In contrast, potent antiarthritic activity in rat adjuvant arthritis was shown for CGP 52608, whereas BRL 49653 was nearly inactive. Our results support the concept that transcriptional control mechanisms via the nuclear receptors PPARg and RORA are responsible at least in part for the different pharmacological properties of BRL 49653 and CGP 52608. Both compounds are prototypes of interesting novel therapeutic agents for the treatment of non-insulin-dependent diabetes mellitus and rheumatoid arthritis. BRL 49653 [(6)-5-([4-[2-methyl-2(pyridylamino)ethoxy] phenyl]methyl)2,4-thiazolidinedione)] is a novel hypoglycemic and hypolipidemic agent used in animal models of NIDDM (Oakes et al., 1994). BRL 49653 and structurally related thiazolidinediones such as ciglitazone, pioglitazone, and troglitazone improve insulin resistance by enhancing insulin action in skeletal muscle, liver, and adipose tissue. Although the precise mechanism of action remains unknown, it has been recently shown that these thiazolidinediones, as well as the prostaglandin J2 metabolite 15d-PGJ2, are ligands of the PPARg (Forman et al., 1995; Lehmann et al., 1995; Lambe and Tugwood, 1996). Thiazolidinedioneinduced activation of PPARg correlates with the antidiabetic actions in vivo (Berger et al., 1996). PPARg is a member of the nuclear receptor superfamily of transcription factors (for a review, see Schoonjans et al., 1996). The three known PPAR subtypes (a, b, and g) exhibit typical tissue distribution in adult animals and during development (Braissant et al., 1996). The expression of PPARg is one of the earliest events during the differentiation of fibroblasts to adipocytes (Tontonoz et al., 1994a), and ectopic expression of PPARg promotes this conversion (Tontonoz et al., 1994b). PPARg regulates the transcription of several adipocyte-specific genes, including phosphoenolpyruvate carboxykinase, adipocyte fatty acid binding protein aP2, and leptin (Schoonjans et al., 1996). Antidiabetic thiazolidinediones and 15d-PGJ2 promote adipocyte differentiation and suppress leptin gene expression in concentrations similar to their Kd values for binding to PPARg (De Vos et al., 1996; Kallen and Lazar, 1996). These findings suggest a pivotal role for PPARg and its ligands in controlling adipocyte development and glucose homeostasis. ABBREVIATIONS: PPAR, peroxisome proliferator-activated receptor; ROR, retinoic acid receptor-related orphan receptor; RORA, retinoic acid receptor-related orphan receptor a, RAR, retinoic acid receptor; RXR, retinoid X receptor; RA, all-trans retinoic acid, GR, glucocorticoid receptor; CAT, chloramphenicol acetyltransferase; FCS, fetal calf serum; NIDDM, non-insulin-dependent diabetes mellitus; PGD2, prostaglandin D2; 15d-PGJ2, 15-deoxy-D12,14-prostaglandin J2. 0026-895X/98/061131-08$3.00/0 Copyright © by The American Society for Pharmacology and Experimental Therapeutics All rights of reproduction in any form reserved. MOLECULAR PHARMACOLOGY, 53:1131–1138 (1998). 1131 at A PE T Jornals on N ovem er 7, 2017 m oharm .aspeurnals.org D ow nladed from CGP 52608 [1-[3-allyl-4-oxo-thiazolidine-2-ylidene]-4methyl-thiosemicarbazone] is the lead compound of a structurally different class of thiazolidinedione derivatives with potent therapeutic effects in experimental arthritis models (Missbach et al., 1996). This compound has been shown to specifically activate RORA, another member of the nuclear receptor superfamily (Wiesenberg et al., 1995). RORA (RORa or RZRa) is one of three known subtypes (a, b, and g) of the ROR. Each subtype shows a characteristic tissue expression pattern (for a review, see Carlberg and Wiesenberg, 1995). In searching for a natural ligand, the pineal gland hormone melatonin was found to specifically activate RORA and to compete with CGP 52608 for binding (Wiesenberg et al., 1995). Structure-activity relationship studies with CGP 52608 analogues revealed a striking correlation between activation of RORA and inhibition of rat adjuvant arthritis, suggesting a key role of this receptor in mediating the antiarthritic effects of these compounds (Missbach et al., 1996). The identification of orphan receptor ligands is always an important step toward a better understanding of their regulatory functions during development and homeostasis. In the case of PPARs, structurally diverse compounds such as peroxisome proliferators, antidiabetic thiazolidinediones, fatty acids, prostaglandin and leukotriene derivatives, and the endogenous steroid dehydroepiandrosterone have been shown to activate PPARs, either directly as ligands or indirectly by as-yet-unknown mechanisms (Devchand et al., 1996; Peters et al., 1996; Forman et al., 1997; Kliewer et al., 1997). So far, known activators of RORs are the pineal gland hormone melatonin (Becker-Andre et al., 1994; Wiesenberg et al., 1995) and the antiarthritic thiazolidinedione derivatives (Missbach et al., 1996). Given the diversity of compounds capable of activating PPARs, the aim of this study was to investigate the specificity of PPARg and RORA activation by the antidiabetic thiazolidinedione BRL 49653 and the antiarthritic thiazolidinedione derivative CGP 52608 and to compare their effects in functional assays in which either compound had shown typical effects. The models used were leptin production in differentiated adipocytes and glucocorticoid-induced insulin resistance in rats (models for PPARg ligands) and rat adjuvant arthritis, an in vivo model in which CGP 52608 and analogues have shown high activity. Materials and Methods Compounds. All thiazolidinediones and derivatives were synthesized in the Department of Chemical Research (Novartis Pharma AG, Basel, Switzerland). The structures of CGP 52608 and BRL 49653 are given in Fig. 1. CGP 52608, CGP 55707, and CGP 55066 were synthesized as described by Missbach et al. (1996), and BRL 49653 (racemate) was synthesized as described by Cantello et al. (1994). Melatonin and PGD2 were obtained from Fluka (Buchs, Switzerland), and 15d-PGJ2 was from Cayman Chemical (Ann Arbor, MI). Thiazolidinedione derivatives and melatonin were dissolved in dimethylsulfoxide, and prostaglandins were dissolved in ethanol at 10 mM; dilutions were made in cell culture medium before use. Preparative separation of the enantiomers of BRL 49653. The two enantiomers of BRL 49653 were separated from the racemate by chiral high performance liquid chromatography (Abbott et al., 1994). A 500 3 50-mm column loaded with Chiracel absorbance was used. The eluent was 15% ethanol in n-heptane, and the flow rate was 150 ml/min. The enantiomers were detected by UV (254 nm). Retention times were 14.6 min for the R-(1)-enantiomer and 37.6 min for the S-(2)-enantiomer, yielding 99.6% pure R-(1)-enantiomer and 77% pure S-(2)-enantiomer. DNA constructs, transfection, and CAT assays. Materials and methods were described in detail by Wiesenberg et al. (1995). Briefly, we used the pBLCAT2-derived CAT reporter constructs, containing, in the XbaI site, natural response elements for ROR (CAAAATGGGTCA), identified in the human 5-lipoxygenase gene promoter (Steinhilber et al., 1995); for PPAR (AATGTAGGTAATAGTTCAATAGGTCA), found in the mouse bifunctional enzyme gene promoter (Bardot et al., 1993); or for RAR (AGGGTTCACCGAAAGTTCA), from the human RARb gene promoter (De The et al., 1990). The cDNAs of human RORA (Becker-Andre et al., 1993), Xenopus laevis PPARg (Dreyer et al., 1992), human RXRa, and human RARa have been subcloned into the expression vector pSG5 (Stratagene, La Jolla, CA). Drosophila SL-3 cells (2 3 10 cells per well in a six-well plate) were grown overnight in Schneider’s medium (Life Technologies, Grand Island, NY) without FCS. Liposomes were formed by incubating 2 mg of the reporter plasmid, 1 mg of receptor expression vector, and 1 mg of the reference plasmid pCH110 (Pharmacia, Piscataway, NJ) with 15 mg of N-[1-(2,3-dioleoyloxy)propyl]-N,N,N-trimethylammonium methylsulfate (Boehringer-Mannheim, Mannheim, Germany) for 15 min at room temperature in a total volume of 100 ml. After dilution with 0.9 ml of Schneider’s medium, the liposomes were added to the cells. Eight hours after transfection, 500 ml of Schneider’s medium supplemented with the indicated ligand was added. After an additional 16 hr, the cells were harvested, and CAT assays were performed. The CAT activities were normalized to b-galactosidase activity, and induction factors were calculated as the ratio of CAT activity of ligand-stimulated cells to that of mock-induced controls. Each condition was analyzed in triplicate, and data are shown as mean values with standard deviation. Secretion of leptin by fully differentiated 3T3-F442A adipocytes. The clonal cell line 3T3-F442A (Green and Kehinde, 1976) was obtained through the courtesy of Dr. B. Fève (INSERM Crétail, France). Differentiation was carried out as described previously (Dani et al., 1989). Cells were plated onto 12-well tissue culture plates (Falcon) at a density of 1.5 3 10 cells/cm in Dulbecco’s Fig. 1. Structures of CGP 52608 and BRL 49653. 1132 Wiesenberg et al. at A PE T Jornals on N ovem er 7, 2017 m oharm .aspeurnals.org D ow nladed from modified Eagle’s medium supplemented with 8% FCS, 200 units/ml penicillin, 50 mg/ml streptomycin, 33 mM biotin, and 17 mM pantothenate. Differentiation was initiated after the cells reached confluency by adding to the standard medium 2 nM triiodothyronine, 17 nM insulin, 100 nM dexamethasone, and 100 mM isobutylmethylxanthine (differentiation medium). After 3 days, the medium was replaced by differentiation medium without dexamethasone and isobutylmethylxanthine and changed thereafter every second day for 2 weeks until differentiation to adipocytes was complete. For measurement of leptin secretion, fully differentiated adipocytes were incubated for 72 hr in fresh medium containing test compounds or the vehicle. Immunoreactive leptin was quantified in the supernatants by radioimmunoassay using rabbit polyclonal antibodies against recombinant mouse leptin (Rentsch et al., 1995). Typically, 50-ml standards or samples were incubated for 18 hr at 4° with 50 ml of radiolabeled [I]leptin, 50 ml of antiserum (diluted 1:4000), and 50 ml of phosphate-buffered saline in the presence of 0.1% Tween-20 and 0.1% bovine serum albumin (final concentrations). Bound and free leptin were separated by centrifugation (30 min at 3000 3 g at 4°) after the addition of 600 ml of 20% polyethylene glycol and 50 ml of g-globulin (10 mg/ml). Pellets were counted in a gamma counter. The detection limit was 0.48 ng of leptin/ml. Experiments were performed in triplicate, and the results are given as mean values with standard deviation. Glucocorticoid-induced insulin resistance. Male Lewis rats (LEW/TIF; specific pathogen free, 250–270 g body weight, five animals/group; Ciba Animal Farm, Sisseln, Switzerland) received either dexamethasone alone (0.15 mg/kg p.o.) or a combination of dexamethasone (0.15 mg/kg p.o.) and CGP 52608 or BRL 49653 (0.01–10 mg/kg p.o.) for 9 days. Test compounds were dissolved in 0.5 ml of DMSO plus 4.5 ml of 0.75% methylcellulose and were administered in a volume of 5 ml/kg. Control animals received the vehicle. After the animals were fasted overnight, blood samples were taken from the orbital vein in isoflurane narcosis on days 4 and 9. Blood glucose and plasma triglyceride levels were estimated using commercially available test kits (Boehringer-Mannheim). Differences between groups were statistically evaluated by Student’s t test. Rat adjuvant arthritis. Adjuvant arthritis was induced as described previously (Wiesenberg et al., 1989). Briefly, male Lewis rats (LEW/TIF; SPF, 180–200 g of body weight) were immunized by an intraplantar injection of Freund’s complete adjuvant (0.2 mg of heatkilled Mycobacterium butyricum (Difco, Detroit, MI) in 0.05 ml of paraffin oil (Riedel de Haen, Seelze, Switzerland) into the left hind paw (day 0). This procedure induced arthritis in 100% of the animals. Disease progression was followed by plethysmographic edema measurements of the injected hind paw (primary lesion) and the noninjected hind paw (secondary lesion). CGP 52608, BRL 49653 and prednisolone (reference compound) were given orally to five animals/ group between 8.00 and 10.00 a.m. from day 0 to 30 in 10 ml/kg of 0.75% methylcellulose. Normal (n 5 5) and arthritic control animals (n 5 10) received the vehicle. Differences between groups were statistically evaluated by Student’s t test.