Original Article Halofenate Is a Selective Peroxisome Proliferator–Activated Receptor ␥ Modulator With Antidiabetic Activity Tamara Allen,1 Fang Zhang,2 Shonna A. Moodie,2 L. Edward Clemens,2 Aaron Smith,1 Francine Gregoire,2 Andrea Bell,2 George E.O. Muscat,1 and Thomas A. Gustafson2 Halofenate has been shown previously to lower triglycer- (1–4). Treatment of dyslipidemic type 2 diabetic patients ides in dyslipidemic subjects. In addition, significant de- also showed triglyceride lowering and, surprisingly, signif- creases in fasting plasma glucose were observed but only in icant reductions in plasma glucose and insulin (3). Subse- type 2 diabetic patients. We hypothesized that halofenate quent studies in diabetic patients confirmed the glucose- might be an insulin sensitizer, and we present data to and triglyceride-lowering effects of halofenate in combina- suggest that halofenate is a selective peroxisome prolifera- tion with oral hypoglycemic drugs and as monotherapy tor–activated receptor (PPAR)-␥ modulator (SPPAR␥M). We demonstrate that the circulating form of halofenate, (4–7). While the precise mechanism of halofenate’s poten- halofenic acid (HA), binds to and selectively modulates tiation of the glycemic effect of sulfonylureas was not PPAR-␥. Reporter assays show that HA is a partial PPAR-␥ understood, it was originally hypothesized that halofenate, agonist, which can antagonize the activity of the full ago- being highly plasma protein bound, might dislodge oral nist rosiglitazone. The data suggest that the partial ago- hypoglycemic compounds from serum binding proteins, nism of HA may be explained in part by effective thus increasing their efficacy (8). However, significant displacement of corepressors (N-CoR and SMRT) coupled decreases in glucose were also observed with halofenate with inefficient recruitment of coactivators (p300, CBP, monotherapy (5) showing that halofenate could function and TRAP 220). In human preadipocytes, HA displays weak independently of sulfonylureas. In analyzing these histor- adipogenic activity and antagonizes rosiglitazone-medi- ated adipogenic differentiation. Moreover, in 3T3-L1 adi- ical data, we noted that halofenate lowered glucose levels pocytes, HA selectively modulates the expression of in diabetic, but not normoglycemic, subjects and that the multiple PPAR-␥–responsive genes. Studies in the diabetic time course of the beneficial glycemic effects was similar ob/ob mouse demonstrate halofenate’s acute antidiabetic to that of the insulin-sensitizing thiazolidinediones (TZDs), properties. Longer-term studies in the obese Zucker (fa/ which possesses glucose- and insulin-lowering properties fa) rat demonstrate halofenate’s comparable insulin sensi- mediated via activation of peroxisome proliferator–acti- tization to rosiglitazone in the absence of body weight vated receptor (PPAR)-␥ (9). We hypothesized that the increases. Our data establish halofenate as a novel insulin-sensitizing effects of halofenate might similarly ␥ SPPAR M with promising therapeutic utility with the po- involve PPAR-␥; we carried out a series of experiments to tential for less weight gain. Diabetes 55:2523–2533, 2006 test this hypothesis. PPAR-␥ is a member of the NR1C subgroup, which includes PPAR-␣ and -␦. These receptors form het- erodimers with the retinoid X receptor and modulate the alofenate was tested clinically in the 1970s as a transcription of genes. PPAR-␥ is predominantly ex- hypolipidemic and hypouricemic agent. In sub- pressed in white and brown adipose tissue, with lower sequent investigator-led studies, halofenate expression in liver, muscle, and other tissues (10). PPAR-␥ Hwas shown to lower serum triglycerides and ligands include a surprisingly diverse set of natural ligands uric acid in patients with a variety of hyperlipidemias (11) such as linolenic, eicosapentaenoic, docohexaenoic, and arachidonic acid and synthetic ligands such as the TZDs, L-tyrosine–based compounds, several nonsteroidal From the 1Division of Molecular Genetics and Development, Institute for Molecular Bioscience, University of Queensland, St. Lucia, Australia; and the anti-inflammatory drugs, and a variety of new chemical 2Department of Biology, Metabolex, Hayward, California. classes (12,13). Originally identified as a regulator of Address correspondence and reprint requests to Thomas A. Gustafson, adipogenesis, PPAR-␥ was thought to mediate the actions Metabolex, 3876 Bay Center Pl., Hayward, CA 94545. E-mail: gus@ metabolex.com. of TZDs solely through its actions in adipose tissue. Received for publication 4 May 2006 and accepted in revised form 20 June However, subsequent studies utilizing tissue-specific 2006. PPAR-␥ gene knockouts have demonstrated a complex T.A., F.Z., and S.A.M. contributed equally to this work. ␥ Additional information for this article can be found in an online appendix at role for PPAR- in whole-body insulin sensitivity involving http://diabetes.diabetesjournals.org. multiple tissues, including liver and muscle (14–16). AUC, area under the curve; HA, halofenic acid; ID, interaction domain; Two TZDs, rosiglitazone and pioglitazone, are currently PPAR, peroxisome proliferator–activated receptor; SPPAR␥M, selective approved to treat type 2 diabetes. Despite their proven PPAR-␥ modulator; TZD, thiazolidinedione. DOI: 10.2337/db06-0618 efficacy, a number of deleterious side effects have been © 2006 by the American Diabetes Association. noted, including increased weight gain and edema (17). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “advertisement” in accordance Weight gain is likely due to both increased adiposity and with 18 U.S.C. Section 1734 solely to indicate this fact. fluid retention. Edema is particularly a problem in patients DIABETES, VOL. 55, SEPTEMBER 2006 2523 HALOFENATE IS A SELECTIVE PPAR-␥ MODULATOR who are also taking insulin or sulfonylureas, and TZD treatment has been linked to increased incidence of con- gestive heart failure (17,18). Accordingly, efforts have been mounted to generate novel PPAR-␥ modulators that retain the beneficial clinical effects while avoiding these side effects. A variety of new PPAR-␥ ligands that possess differential pharmacological affinities for PPAR-␥ and have been termed selective PPAR-␥ modulators (SPPAR␥Ms) have recently been reported (19). SPPAR␥Ms are believed to bind in distinct manners to the ligand binding pocket of PPAR-␥, leading to altered receptor conformational stabil- ity and resulting in distinct patterns of gene expression FIG. 1. Chemical structures of compounds. Halofenate (A) is a pro-drug (20–22). The molecular basis of this effect is thought to ester of HA (B). Halofenate is rapidly modified in vivo to HA, the involve differential cofactor displacement and recruitment circulating form of the drug. that regulates gene expression in a gene- and tissue- ␮ specific manner (23). Further characterization of (catalog no. AM-1; 100 nmol/l insulin and 1.0 mol/l dexamethasone) in the ␥ presence of 0.25 mmol/l isobutyl-methylxanthine and either rosiglitazone (1 SPPAR Ms will ideally yield agents for the treatment of ␮mol/l) or HA (100 or 200 ␮mol/l) for 3 days and then fed with adipocyte diabetes, which are as effective as current therapies but medium for additional 12 days. PPAR-␥–mediated ligand-induced differentia- reduce or eliminate the more deleterious side effects. This tion was assessed by Oil Red O staining and analysis of FABP4 gene study was initiated in an effort to elucidate the underlying expression by quantitative PCR. mechanism of the antidiabetic effects of halofenate. Our Quantitative fluorescent real-time PCR analysis. Total RNA was isolated data suggest that halofenate is a novel SPPAR␥M with using Trizol (InVitrogen). cDNA was synthesized and used for quantitative fluorescent real-time PCR amplification. Amplification of each target cDNA robust antidiabetic and insulin-sensitizing activity. Fur- was then performed with TaqMan PCR Reagent Kits according to the thermore, halofenate appears to enhance insulin sensitiv- manufacturer’s protocol using Assay-On-Demand (Applied Biosystems) as- ity in a manner that leads to less weight gain than the says. 36B4 RNA was used as the control. Data are expressed as the means Ϯ currently marketed compounds. SE. Statistical significance was determined using ANOVA. ob/ob mouse studies. Male 8-week-old ob/ob mice (Jackson Laboratories) were used. All animal procedures were approved by the local institutional animal care and use committee following guidelines issued by the U.S. RESEARCH DESIGNS AND METHODS Department of Agriculture Animal and Plant Health Inspection Services. Plasmids. pCMX-mouse PPAR-␥ and -␣ plasmids were gifts from R. Evans. Halofenate was formulated daily as a suspension by vortexing in a mixture of The ligand-binding domain of mouse PPAR-␥ (aa 174–475) and -␣ (aa 2% Tween 80 and 1.0% (wt/vol) methylcellulose. Vehicle and halofenate were 166–463) was generated by PCR and cloned into pGAL0. Full-length mouse administered by oral gavage. Blood samples were collected by tail nipping 3 h PPAR-␥ was generated by PCR and cloned into pVP16. The ligand-binding after dosing. Plasma glucose levels were measured using the method of domains for human PPAR-␣ (aa 166–469), -␦ (aa 135–442), and -␥ (aa 172– Trinder (Glucose Oxidase G7016, Peroxidase P8125; Sigma Chemical, St. 476) were generated by PCR and cloned into pFA-CMV plasmid (Stratagene). Louis, MO) and insulin by radioimmunoassay (Linco).