Direct Thiazolidinedione Action on Isolated Rat Skeletal Muscle Fuel Handling Is Independent of Peroxisome Proliferator–Activated Receptor-␥؊Mediated Changes in Gene Expression Barbara Brunmair, Florian Gras, Susanne Neschen, Michael Roden, Ludwig Wagner, Werner Waldha¨usl, and Clemens Fu¨ rnsinn

Thiazolidinediones (TZDs) are believed to induce insu- lin sensitization by modulating gene expression via agonistic stimulation of the nuclear peroxisome prolif- nsulin resistance is a common metabolic abnormal- erator–activated receptor-␥ (PPAR-␥). We have shown ity associated with obesity, hypertension, and type 2 earlier that the TZD troglitazone inhibits mitochondrial diabetes (1). Thiazolidinediones (TZDs) are a class fuel oxidation in isolated rat skeletal muscle. In the Iof oral antidiabetic agents that improve insulin sen- present study, rat soleus muscle strips were exposed to sitivity and glucose homeostasis in type 2 diabetic patients TZDs to examine whether the inhibition of fuel oxida- (2–4) as well as in various animal models of diabetes and tion is mediated by PPAR-␥ activation. Our findings obesity (2–9). The TZDs troglitazone, rosiglitazone, and ,consistently indicated direct, acute, and PPAR-␥؊inde- pioglitazone have already been used in clinical practice pendent TZD action on skeletal muscle fuel metabolism. but the mechanisms by which TZDs improve insulin ␮ Rapid stimulation of lactate release by 20 mol/l trogli- sensitivity as well as deranged glucose and lipid metabo- tazone within 30 min suggested that direct TZD action lism are not yet fully understood (2–4,10). on skeletal muscle in vitro does not rely on changes in –gene expression rates (12.6 ؎ 0.6 [control] vs. 16.0 ؎ 0.8 TZDs are agonistic ligands of peroxisome proliferator ␥ ␥ ؊ ؊ ␮mol ⅐ g 1 ⅐ h 1 [troglitazone]; P < 0.01). This conclu- activated receptor- (PPAR- ), which belongs to the sion was supported by the failure of actinomycin D and nuclear hormone receptor superfamily of transcription cycloheximide to block the effects of troglitazone. Mi- factors (5,6,10–12). Upon stimulation, PPAR-␥ binds to tochondrial fuel oxidation was consistently inhibited by responsive elements located in the promotor regions of six different TZDs (percent inhibition of CO2 produc- many genes and modulates their transcriptive activities tion from palmitate after 25 h: troglitazone, ؊61 ؎ 2%; (2,10). Convincing evidence for an important role of -pioglitazone, ؊43 ؎ 7% ; rosiglitazone, ؊22 ؎ 6%; PPAR-␥ in TZD-induced insulin sensitization includes in BM13.1258, ؊47 ؎ 9%; BM15.2054, ؊51 ؎ 4%; and sulin sensitization by PPAR-␥ agonists that do not belong T-174, ؊59 ؎ 4% [P < 0.005 each]), but not by PPAR-␥ to the TZD class (13–16) and by LG-100.268, an agonist of agonistic compounds not belonging to the TZD class retinoid X receptor (RXR), which is the heterodimeric ؎ ؎ ؊ (JTT-501, 5 7% [NS]; prostaglandin J2,17 7% [P < partner of PPAR-␥ (17). Furthermore, a strong correlation 0.05]), which further argues against dependence on of antidiabetic efficacies of PPAR-␥ agonists in vivo with PPAR-␥ activation. In summary, our findings provided their respective potentials to bind and activate PPAR-␥ in good evidence that direct inhibition of mitochondrial vitro has been demonstrated (6,12,16). fuel oxidation in isolated skeletal muscle is a group- ␥ specific effect of TZDs and is independent of PPAR-␥؊ Although there is evidence to support PPAR- having mediated gene expression. Diabetes 50:2309–2315, 2001 an important role in TZD-induced insulin sensitization, TZDs also seem to address other mechanisms that do not involve PPAR-␥. That metabolic responses to TZDs in several experimental setups are independent of PPAR-␥– induced gene transcription is indicated by their rapid occurrence (5,7,18–24) and by a failure of the amplitudes of such responses to reflect the PPAR-␥–activating effica- cies of different TZDs (5,20). The relevance of PPAR-␥ is From the Department of Medicine III, Division of Endocrinology & Metabo- specifically unclear in skeletal muscle, which quantita- lism, University of Vienna, Austria. Address correspondence and reprint requests to Clemens Fu¨ rnsinn, Ph.D., tively is the most important target tissue for insulin and Department of Medicine III, Division of Endocrinology & Metabolism, Wa¨hr- plays a predominant role in TZD-induced improvement of inger Gu¨ rtel 18-20, A-1090 Vienna, Austria. E-mail: clemens.fuernsinn@ ␥ akh-wien.ac.at. glucose homeostasis (25,26). Although PPAR- mRNA is Received for publication 15 June 2000 and accepted in revised form 27 June abundant in fat and hardly found in skeletal muscle 2001. (10,27,28), the abundance of PPAR-␥ protein in muscle BSA, bovine serum albumin; [3H]2DG, 2-deoxy-D-[2,6-3H]glucose; PPAR, peroxisome proliferator–activated receptor; RXR, retinoid X receptor; TZD, was recently reported to be 67% of that in fat (29). thiazolidinedione. However, clear experimental evidence for any functional

DIABETES, VOL. 50, OCTOBER 2001 2309 PPAR␥-INDEPENDENT TZD ACTION IN VITRO

PPAR-␥ signaling in skeletal muscle has never been pro- measurement period, the media alternatively contained trace amounts of 14 14 3 3 vided. Because relevance of direct TZD effects on skeletal D-[U- C]glucose, [U- C]palmitic acid, or 2-deoxy-D-[2,6- H]glucose ([ H]2DG) plus [U14C]sucrose (all from Amersham, Amersham, U.K.), and if not stated muscle for antidiabetic action is still being debated, we otherwise, 100 nmol/l insulin. At the end of the measurement period, muscles examined the interaction of TZDs with isolated skeletal were quickly removed, blotted, and frozen in liquid nitrogen. Later, muscle muscle fuel metabolism. strips were lysed in 1 mol/l KOH at 70°C; the lysate was then used for further We recently reported that troglitazone had rapid and analytical procedures, as described below. The method used for determining CO production was based on the direct action on fuel metabolism of freshly isolated rat 2 skeletal muscle in vitro, action that was characterized by addition of radiolabeled palmitate or glucose to the incubation medium. Hence the data reflect CO production from the extracellular substrate only, distinct inhibition of CO production from palmitate and 2 2 but neglect CO2 production from intracellular lipid or glycogen pools. Some glucose (19). Troglitazone shifted glycolytic flux from the experiments were designed to determine CO2 production from intracellular aerobic toward the anaerobic pathway simultaneous with lipid and glycogen stores prelabeled with [U-14C]palmitic acid or D-[U- glycogen depletion, which was marked after exposure of 14C]glucose, respectively. In these experiments, trace amounts of the radio- muscle specimens to troglitazone for 25 h (19). In the active compounds were added to the medium during a 24-h pretreatment period. Subsequently, CO2 production was measured without radioactive present study, we aimed to investigate the mechanisms tracers in the medium. responsible for direct interaction of TZDs with in vitro fuel Short-term muscle incubation (30 or 60 min). Muscles from SD rats were handling in skeletal muscle using specimens of isolated rat put into coated 25-ml Erlenmeyer flasks that were placed into the waterbath soleus muscle. In particular, the studies were designed to immediately after preparation (one strip per flask). Each flask contained a continuous atmosphere of 95% O /5% CO and 3 ml Krebs-Ringer buffer provide evidence for or against the hypothesis that TZDs 2 2 solution (pH 7.35) supplemented with 5.5 mmol/l glucose, 0.3% (wt/vol) BSA, directly affects muscle fuel metabolism independent of 300 ␮mol/l palmitate, 0.5% (vol/vol) ethanol, and 0.1% (vol/vol) DMSO. PPAR-␥Ϫinduced modulation of gene expression. In the short-term experiments, the pretreatment period lasted for 30 min, after which muscles were immediately transferred into another set of flasks. Then muscles were incubated for another 30 min in 3 ml of identical buffer RESEARCH DESIGN AND METHODS solution, which was supplemented with the above-described radioactive Rats. Male SD rats were purchased from the breeding facilities of the tracers and 30 nmol/l insulin (measurement period). A total of 20 ␮mol/l University of Vienna (Himberg, Austria) and were used at a body weight of troglitazone was added to the incubation medium during the measurement ϳ140 g. Obese Zucker rats (HsdHlr: fa/fa) were obtained from Harlan- period only or during both pretreatment and measurement periods, resulting Winkelmann (Borchen, Germany) and were used at age ϳ5 months (body in troglitazone exposure periods of 30 and 60 min, respectively. Finally, weight ϳ750 g). Rats were kept in an artificial 12-h light/dark cycle at constant muscles were quickly removed, blotted, frozen, and lysed in KOH for further room temperature. Conventional laboratory diet and tap water were provided analytical procedures. ad libitum until the evening before rats were killed, when only food was Analytical procedures. Net uptake of [3H]2DG was determined using withdrawn. Rats were killed by cervical dislocation between 8:30 and 9:30 A.M. [14C]sucrose as a marker of extracellular space by previously described All experiments were performed according to local law and the principles of methods (19). Under the applied experimental conditions, insulin-stimulated good laboratory animal care. [3H]2DG uptake did not reach saturation within the measurement period of Compounds. The PPAR-␥ agonistic TZDs troglitazone, pioglitazone, and 60 min (data not shown). The net rate of glucose incorporation into glyco- rosiglitazone, as well as the RXR agonist LG-100.268 (17), were generously gen, referred to as glycogen synthesis, was determined by measuring the provided by Sankyo (Tokyo, Japan); the TZDs BM13.1258 and BM15.2054 (5) conversion of [14C]glucose to [14C]glycogen, as previously described (31). were generously provided by Boehringer-Mannheim/LaRoche (Mannheim, 14 Rates of CO2 production were calculated from the conversion of [ C]glucose Germany); the TZD T-174 (8) was generously provided by Tanabe (Saitama, or [14C]palmitate into 14CO , which was trapped with a solution containing ␥ ␣ 2 Japan); the non-TZD PPAR- plus PPAR- agonist JTT-501 (13,14) was methanol and phenethylamine (1:1) (33). Rates of lactate release were calcu- ␥ generously provided by Tobacco (Osaka, Japan); and the natural PPAR- lated from the amount of lactate accumulated in the incubation medium ⌬12,14 ligand 15-deoxy- -prostaglandin J2 (30) and vitamin E succinate were during the measurement period; this concentration was determined enzymat- purchased from Calbiochem (La Jolla, CA). ically using the lactate dehydrogenase method (34). For the determination of Long-term muscle incubation (5 or 25 h). Immediately after rats were muscle glycogen content prevailing at the end of the experiment, glycogen in killed, two (SD rats) or three (Zucker rats) longitudinal soleus muscle strips the muscle lysate was completely degraded to glucose with amyloglucosidase per leg (i.e., four or six strips per rat) were prepared, weighed (ϳ25 mg/strip), (33). Glucose was then measured enzymatically by a commercial kit (Human, and tied under tension on stainless steel clips, as previously described (31). Taunusstein, Germany). According to procedures used earlier (19,32), muscles were immediately put Positive control experiments on cycloheximide and actinomycin D. To into 50-ml Erlenmeyer flasks coated with BlueSlick solution (Serva, Heidel- confirm that cycloheximide efficiently blocked protein synthesis in our berg, Germany) and placed into a shaking waterbath (four or six strips per experimental setup, muscle strips were incubated in the absence or presence flask, 37°C, 130 cycles/min). Each flask contained 20 ml Cell Culture Medium of 1 mg/l cycloheximide under the described conditions, except that culture 199 (5.5 mmol/l glucose [pH 7.35]; cat. no. M-4530; Sigma, St. Louis, MO) supplemented with 0.3% (wt/vol) fatty acidϪfree bovine serum albumin (BSA), medium devoid of methionine was used (modified Eagle’s medium; cat. no. 5 mmol/l HEPES, 25,000 units/l penicillin G, and 25 mg/l streptomycin. Palmi- 31900-020, Life Technologies, Paisley, U.K.), which was supplemented with all ␮ 35 tate was dissolved in ethanol and added to the medium to give final concen- ingredients as listed above for Medium 199 plus 0.25 Ci/ml [ S]methionine. trations of 300 ␮mol/l palmitate and 0.5% ethanol (vol/vol). An atmosphere of The effects of troglitazone and cycloheximide on soleus muscle fuel handling were not influenced by the medium used (Medium 199 versus modified Eagle’s 95% O2/5% CO2 was continuously provided within the flasks. TZDs, JTT-501, prostaglandin J , LG-100.268, and vitamin E succinate were medium; data not shown). After 4 or 24 h, muscles were quickly removed, 2 35 dissolved in DMSO (Sigma) and added to the medium to give the final blotted, and frozen for the determination of net [ S]methionine incorporated indicated drug concentrations. In some experiments, the medium also con- into protein. Later, muscle specimens were thawed and immediately lysed in tained 1 mg/l of the protein synthesis inhibitor cycloheximide or 1 mg/l of the 0.5 ml NaOH (1 mol/l), and the protein was precipitated with 0.6 ml perchloric nucleic acid synthesis inhibitor (i.e., gene transcription inhibitor) actinomycin acid (1 mol/l). After centrifugation, the supernatant was discarded and the D. The final DMSO concentration in incubation media, including controls, was pellet was redissolved in 0.5 ml NaOH. After this procedure was repeated 0.1% (vol/vol) in all experiments. In experiments designed for direct compar- twice, 0.6 ml perchloric acid were added and the sample was counted for 35S ison of the various agents, the first of four muscle strips from the same SD rat radioactivity. was incubated in the absence of any drug (control), the second was exposed To provide evidence that 1 mg/l actinomycin D inhibited transcription in to troglitazone, and the third and fourth were exposed to other compounds. our experimental setup, muscle strips were homogenized immediately after Hence an intraindividual control was available, and troglitazone action was incubation for 24 h in Medium 199, as described above. Total RNA was confirmed for each rat. isolated from muscle homogenates with RNAzol B following the instructions After pre-exposure periods of 4 or 24 h, muscles were immediately of the manufacturer (Tel-Test, Friendswood, TX), and the RNA content of the transferred into 25-ml flasks and incubated in 3 ml of identical buffer solution extracts was determined photometrically (ratio 260:280 nm Ͼ1.8). (one strip per flask). Muscles were then incubated for another 60 min, during Statistics. All results are given as means Ϯ SE. P values were calculated by which metabolic rates were determined (measurement period). During the two-tailed paired Student’s t test, and P Ͻ 0.05 was considered significant.

2310 DIABETES, VOL. 50, OCTOBER 2001 B. BRUNMAIR AND ASSOCIATES

TABLE 1 Rates of insulin-stimulated (100 nmol/l) fuel metabolism in soleus muscle strips from SD rats exposed to 0 (control) or 5 ␮mol/l troglitazone for 25 h Parameter Units n Control Troglitazone P ⅐ Ϫ1 ⅐ Ϫ1 Ϯ Ϯ Ͻ CO2 production from palmitate nmol palmitate g h 60 76 3292 0.0001 ⅐ Ϫ1 ⅐ Ϫ1 Ϯ Ϯ Ͻ CO2 production from glucose nmol glucose g h 56 1,920 66 727 24 0.0001 Lactate release ␮mol ⅐ gϪ1 ⅐ hϪ1 70 24.4 Ϯ 0.5 52.6 Ϯ 1.0 Ͻ0.0001 Glucose transport cpm ⅐ mgϪ1 ⅐ hϪ1 70 868 Ϯ 22 1,133 Ϯ 28 Ͻ0.0001 Glycogen synthesis ␮mol glucose ⅐ gϪ1 ⅐ hϪ1 58 2.30 Ϯ 0.10 1.30 Ϯ 0.04 Ͻ0.0001 Glycogen content ␮mol glucosyl units/g 58 12.1 Ϯ 0.4 9.1 Ϯ 0.3 Ͻ0.0001 Data are means Ϯ SE.

RESULTS of protein synthesis and transcription, respectively, in our Effects of 25-h troglitazone exposure. In agreement specific experimental setup. Cycloheximide blocked pro- with our previous report (19), the exposure of rat soleus tein synthesis—that is, [35S]methionine incorporation into muscle strips to 5 ␮mol/l troglitazone for 25 h distinctly protein—in isolated rat soleus muscle to 17 Ϯ 3 and 13 Ϯ modulated insulin-stimulated fuel metabolism (Table 1). 2% of control over 4 and 24 h, respectively (P Ͻ 0.0001 ϭ Rates of CO2 production from both palmitate and glucose each; n 6 each). After exposure to actinomycin D for were inhibited by Ϫ62%, whereas lactate release more 24 h, the amount of total RNA extractable from soleus than doubled (ϩ116%). In parallel, [3H]2DG transport was muscle strips was reduced by 28% (0.87 Ϯ 0.06 vs. 1.20 Ϯ enhanced by ϩ34%, and the rate of net glucose incorpora- 0.17 mg/g wet wt; n ϭ 12; P Ͻ 0.05), indicating inhibition tion into glycogen was decreased by Ϫ43%. Compared of transcriptional activity. with that of control specimens, the glycogen content of Marked effects of both cycloheximide and actinomycin muscle strips was Ϫ25% lower after incubation with D on fuel handling of isolated rat muscle were observed troglitazone (all troglitazone effects, P Ͻ 0.0001) (Table 1). after 25 h, whereas only minor effects of actinomycin D In parallel to the reduced conversion of extracellular but no effects of cycloheximide were observed after 5 h. substrates to CO2 (Table 1), troglitazone inhibited CO2 All results were consistent, in that they clearly demonstrat- production from prelabeled intracellular substrate stores ed the failure of cycloheximide and actinomycin D to under basal conditions (troglitazone-induced change as block troglitazone’s inhibitory action on fuel oxidation and Ϫ Ϯ ϭ percent of control: CO2 from palmitate, 33 4%, n 6, glycogen synthesis. Thus, the troglitazone-induced reduc- Ͻ Ϫ Ϯ ϭ Ͻ P 0.0001; CO2 from glucose, 40 6%, n 6, P 0.001) tions in glycogenesis or CO2 production from palmitate and and after stimulation with 100 nmol/l insulin (CO2 from glucose were not affected by inhibition of gene expression Ϫ Ϯ ϭ Ͻ palmitate, 45 6%, n 5, P 0.001; CO2 from glucose, (Fig. 1). Ϫ40 Ϯ 10%, n ϭ 5, P Ͻ 0.005). Troglitazone action on insulin-resistant muscle. Fol- Time dependence of troglitazone action. Exposure of lowing the same protocol used for muscle strips obtained SD rat muscle to 20 ␮mol/l troglitazone for 30 min signif- from SD rats (Table 1), the dosage-dependent effects of icantly increased the rate of insulin-stimulated lactate troglitazone exposure in vitro were tested in soleus muscle release (ϩ 27%; P Ͻ 0.01). After 60 min, the troglitazone- strips obtained from genetically obese Zucker rats (fa/fa), induced increase in lactate release was accompanied by which exhibit severe insulin resistance (31). Using concen- significant reductions in mitochondrial fuel oxidation and trations of 0.63Ϫ10 ␮mol/l troglitazone, a dose-dependent glycogen storage (Table 2). inhibition of fuel conversion to CO2 was observed in Interaction of troglitazone with cycloheximide and association with increased anaerobic glycolysis and glyco- ␮ actinomycin D. Positive control experiments confirmed gen depletion (Fig. 2). At 5 mol/l troglitazone, CO2 pro- that cycloheximide and actinomycin D acted as inhibitors duction from palmitate and glucose changed Ϫ48% and

TABLE 2 Rates of insulin-stimulated (30 nmol/l) fuel metabolism in soleus muscle strips from SD rats exposed to 0 (control) or 20 ␮mol/l troglitazone for 30 or 60 min Parameter Units min Control Troglitazone P ⅐ Ϫ1 ⅐ Ϫ1 Ϯ Ϯ CO2 production from palmitate nmol palmitate g h 30 109 9 105 6NS 60 114 Ϯ 690Ϯ 6 Ͻ0.002 ⅐ Ϫ1 ⅐ Ϫ1 Ϯ Ϯ CO2 production from glucose nmol glucose g h 30 213 8 208 16 NS 60 237 Ϯ 13 201 Ϯ 16 Ͻ0.05 Lactate release ␮mol ⅐ gϪ1 ⅐ hϪ1 30 12.6 Ϯ 0.6 16.0 Ϯ 0.8 Ͻ0.01 60 14.2 Ϯ 0.6 17.4 Ϯ 0.6 Ͻ0.001 Glucose transport cpm ⅐ gϪ1 ⅐ hϪ1 30 675 Ϯ 35 678 Ϯ 44 NS 60 664 Ϯ 24 653 Ϯ 45 NS Glycogen synthesis ␮mol glucose ⅐ gϪ1 ⅐ hϪ1 30 6.03 Ϯ 0.30 5.60 Ϯ 0.30 NS 60 6.57 Ϯ 0.45 5.23 Ϯ 0.36 Ͻ0.005 Glycogen content ␮mol glucosyl units/g 30 14.2 Ϯ 0.6 13.4 Ϯ 1.0 NS 60 15.0 Ϯ 0.7 13.2 Ϯ 0.6 Ͻ0.001 Data are means Ϯ SE; n ϭ 12 each. NS, nonsignificant.

DIABETES, VOL. 50, OCTOBER 2001 2311 PPAR␥-INDEPENDENT TZD ACTION IN VITRO

FIG. 1. Interactions of troglitazone with cycloheximide (A) and actinomycin D (B) on skeletal muscle fuel metabolism. Soleus muscle strips from SD rats were exposed to 1 mg/l cycloheximide or 1 mg/l actinomycin D for 5 or 25 h, and the effects of concomitant exposure to troglitazone (f; 10 ␮mol/l for5hor5␮mol/l for 25 h) or no exposure to troglitazone (Ⅺ) were determined. During the last hour of the incubation, ؎ insulin-stimulated (100 nmol/l) rates of glycogen synthesis and CO2 production from palmitate and glucose were measured. Data are means SE. *P < 0.05, †P < 0.025, ‡P < 0.02, §P < 0.01, ʈP < 0.001, ¶P < 0.0001 without versus with troglitazone; #P < 0.05, **P < 0.02, ††P < 0.01, ‡‡P < 0.001 without versus with cycloheximide/actinomycin D.

Ϫ66%, respectively; lactate release, ϩ234%; glucose trans- port, ϩ110%; glycogen synthesis, Ϫ27%; and glycogen content, Ϫ64%. Comparison of troglitazone to other compounds. Muscle specimens were exposed for 25 h to 5 ␮mol/l of the various agents listed in Fig. 3, which depicts the percent of metabolic rates found in intraindividual control muscle, which was incubated in the absence of any drug (absolute rates for control and troglitazone-exposed muscle strips given in Table 1). All six PPAR-␥ agonistic TZDs tested shifted fuel utili- zation from aerobic toward anaerobic pathways, as indi- cated by blunted rates of CO2 production from palmitate (at least P Ͻ 0.005) and by increased rates of lactate release (P Ͻ 0.0001 each). Quantitative efficacies of the re- spective TZDs, however, varied considerably. Troglitazone triggered a very distinct response, whereas the potent PPAR-␥ agonist rosiglitazone was revealed to be the weak- est TZD in this setup; significant decreases in glucose oxidation (P Ͻ 0.0001 each), glycogen synthesis (at least P Ͻ 0.001), and glycogen content (at least P Ͻ 0.05) were obvious for most TZDs, but not for rosiglitazone. Regarding the non-TZD PPAR-␥ agonists prostaglandin FIG. 2. Insulin-stimulated (100 nmol/l) fuel metabolism in soleus J and JTT-501, 5 ␮mol/l of these substances did not muscle strips from obese Zucker rats (fa/fa) age ϳ5 months exposed to ؎ various concentrations of troglitazone for 25 h (f). Data are means 2 modulate fuel metabolism in the manner observed in SE. *P < 0.05, †P < 0.02, ‡P < 0.01, §P < 0.005, ʈP < 0.001, ¶P < 0.0001 response to TZDs. JTT-501 failed to affect any parameter versus without troglitazone (Ⅺ).

2312 DIABETES, VOL. 50, OCTOBER 2001 B. BRUNMAIR AND ASSOCIATES

FIG. 3. Rates of insulin-stimulated (100 nmol/l) fuel metabolism in soleus muscle strips from SD rats exposed for 25 h to 5 ␮mol/l each of the following: the PPAR-␥ agonistic TZDs troglitazone, pioglitazone, rosiglitazone, BM13.1258, BM15.2054, and T-174 (f); the non-TZD PPAR-␥ 3 o Ⅺ agonists prostaglandin J2 and JTT 501 ( ); the RXR agonist LG-100.268 ( ); and vitamin E succinate ( ). Data are given as percent of an ,intraindividual control value as determined in the absence of the respective compound. Data are means ؎ SE. *P < 0.05, †P < 0.01, ‡P < 0.005 §P < 0.002, ʈP < 0.001, ¶P < 0.0001 versus control. measured, and prostaglandin J2 moderately stimulated have provided clear evidence that this interaction is not Ͻ glycogenesis and palmitate oxidation to CO2 (P 0.05 based on the common property of TZDs to modulate gene each). transcription and protein synthesis via the activation of Parallel experiments were performed using LG-100.268 PPAR-␥. and vitamin E succinate, which are not ligands for PPAR-␥. First, the experiments on time dependence and inter- Both compounds failed to influence glucose transport or action with actinomycin D and cycloheximide argue glycogen metabolism. LG-100.268 inhibited CO2 produc- against any role for PPAR-␥Ϫmediated transcription or tion from palmitate (P Ͻ 0.0001) and marginally increased translation in troglitazone’s direct effects on muscle fuel lactate release (P Ͻ 0.05), and vitamin E moderately metabolism in vitro. Thus distinct responses to troglita- decreased CO2 production from both palmitate and glu- zone occurred within 30 min and became marked within cose (P Ͻ 0.05 each). 60 min, a finding that would argue against the established view that triggering metabolic effects via the modulation of DISCUSSION gene expression requires a more prolonged time period. In There is evidence available to indicate that the antidiabetic addition, actinomycin D and cycloheximide, currently action of TZDs in vivo relies on delayed and PPAR- regarded as potent blockers of gene transcription and ␥Ϫdependent mechanisms, and that TZDs may indirectly protein synthesis, respectively, clearly failed to counteract act on skeletal muscle via PPAR-␥ activation in adipose troglitazone action in our experimental setup. tissue (2,11,35). However, the potential role and mecha- Second, our experiments revealed that direct troglita- nism of direct and acute interaction of TZDs with skeletal zone action on isolated skeletal muscle is independent of muscle are still unclear. We demonstrated in an earlier concomitant insulin stimulation (see also 19), and that study that troglitazone distinctly affects fuel handling of muscles isolated from obese Zucker rats, which exhibit isolated rat soleus muscle strips in vitro. Thus troglitazone lower rates of glucose metabolism than muscles from lean rapidly shifts fuel utilization from aerobic toward anaero- littermates (31), had similar dose-dependent responses to bic pathways, increases the rate of glucose transport, and troglitazone as did muscle strips from normally insulin- inhibits glycogen storage (19). The present study extended sensitive SD rats (detailed dose-response curve for SD rat our previous investigations by demonstrating that CO2 muscle in 19). These findings indicate major differences production is reduced not only from extracellular, but also between TZD in vivo versus in vitro action, because the from intracellular, substrates. Furthermore, we demon- metabolic effects of TZDs in vivo are seen only in the strated that other TZDs share troglitazone’s ability to presence of insulin, and TZD action is hardly detectable in directly interact with isolated rat skeletal muscle, and healthy rodents but is very distinct in insulin-resistant

DIABETES, VOL. 50, OCTOBER 2001 2313 PPAR␥-INDEPENDENT TZD ACTION IN VITRO animals (7,9). Furthermore, the inhibition of fuel oxidation muscle. Hence independent potentials to inhibit fuel oxi- observed in obese rat muscle exposed to TZDs in vitro dation to CO2 may be intrinsic to both parts of the contrasts with the increased glucose oxidation and insulin troglitazone molecule, and may interact in a synergistic sensitization of the glycogenic pathway prevailing in skel- manner to render troglitazone with extraordinary potency etal muscle isolated from insulin-resistant rodents orally to directly influence fuel metabolism of isolated muscle. treated with TZDs in vivo (5,36,37). Given the widely The different pharmacokinetic properties of the specific accepted assumption that TZD-induced insulin sensitiza- TZD, however, may also be of relevance. tion in vivo relies on PPAR-␥ activation (2,6,10–12,16), The marked reduction of CO2 production from palmitate such divergences in TZD in vivo versus in vitro action observed in TZD-exposed isolated rat muscle is in line support our conclusion that a different—i.e., a PPAR- with findings from several other studies likewise reporting ␥Ϫindependent—mechanism must underlie modulation of that TZD can inhibit oxidative steps of lipid metabolism in fuel metabolism in isolated muscle. vitro. Thus TZD exposure has been demonstrated to Third, we compared the efficacies of different PPAR-␥ distinctly reduce ketone body production from oleate in agonistic compounds and found further evidence against rat hepatocytes (21), progesterone production in porcine the involvement of PPAR-␥. Our results clearly indicated granulosa cells (22), and cholesterol synthesis in several that the observed shift in fuel utilization from aerobic cell lines related to adipose tissue, liver, and muscle (20). toward anaerobic pathways is specific for drugs belonging Notably, many parallels exist with our findings, including to the TZD class. This conclusion is based on the finding 1) superiority of troglitazone over pioglitazone and rosigli- that all six TZDs tested markedly stimulated lactate re- tazone (20–22), 2) the occurrence of responses within a lease and inhibited CO production, whereas the two short time range (20–22), 3) the lack of modulation by 2 4 non-TZD PPAR-␥ agonists, JTT-501 and prostaglandin J , cycloheximide and actinomycin D (20), and ) the inde- 2 pendence of the concomitant presence of insulin (19–22). failed to trigger such responses. Considering the inability In conclusion, our results demonstrated that TZDs di- of non-TZD PPAR-␥ agonists to elicit responses as trig- rectly and rapidly affect fuel metabolism of isolated native gered by the same concentration of a TZD, it is of note that rat skeletal muscle independent of PPAR-␥ activation and the RXR agonist LG-100.268 inhibited the conversion of gene expression. The observed shift of fuel utilization palmitate to CO2. Although final conclusions cannot be from aerobic to anaerobic pathways obviously has no drawn from the present study, it is possible that different functional consequence for the ability of TZDs to act as mechanisms mediate the actions of TZDs and LG-100.268, insulin sensitizers in vivo, a finding that agrees with the as RXR is the heterodimeric partner for a number of other suggested importance of PPAR-␥ for antidiabetic TZD ␥ receptors beside PPAR- (17). action. Nevertheless, the obtained data suggest a TZD- Comparison of the relative potencies of different com- specific potential to affect fuel metabolism via PPAR- pounds belonging to the TZD class further corroborates ␥Ϫindependent mechanisms of action. The contribution of our interpretation that the observed responses to TZDs in such PPAR-␥Ϫindependent actions to the beneficial vitro were not related to PPAR-␥ activation. Focusing on and/or unwanted actions of TZDs in vivo, however, is the three TZDs that have been used for regular patient unclear and may differ considerably among the various treatment, the relative PPAR-␥ agonistic efficacy of rosigli- antidiabetic agents belonging to the TZD class. tazone is greater than that of pioglitazone, which in turn is greater than that of troglitazone (5,6,12,16,20). This rank- ACKNOWLEDGMENTS ing, however, is reversed by these TZDs’ respective poten- This work was supported by the Austrian Science Fund tials to inhibit mitochondrial fuel oxidation in vitro, in (grant no. P13049-MED). B.B. was supported in part by an which case troglitazone is ranked higher than pioglitazone, institutional grant from Sankyo to the Division of Endo- which in turn is ranked higher than rosiglitazone. The crinology and Metabolism. relative efficacies of TZDs to modulate fuel metabolism in We thank Sankyo (Tokyo, Japan), Tobacco (Osaka, isolated muscle therefore clearly fail to reflect their Japan), Boehringer-Mannheim/LaRoche (Mannheim, Ger- ␥ PPAR- agonistic potentials. In this context, it should be many), and Tanabe (Saitama, Japan) for generously pro- ␥ emphasized that the strong correlation of PPAR- agonism viding the compounds. We also thank the staff at the in vitro with insulin-sensitizing efficacy in vivo is often Biomedical Research Center, University of Vienna, for ␥ regarded as the most convincing evidence for PPAR- taking care of the rats. activation being an early and essential step in TZD-induced insulin sensitization (6,12,16). In contrast, we conclude REFERENCES that the different rankings of efficacy in our experimental 1. DeFronzo RA, Ferrannini E: Insulin resistance: a multifaceted syndrome setting provide good evidence that PPAR-␥ activation is responsible for NIDDM, obesity, hypertension, dyslipidemia, and athero- not relevant for direct interaction of TZDs with skeletal sclerotic cardiovascular disease. Diabetes Care 14:173–194, 1991 muscle fuel handling in vitro. 2. Saltiel AR, Olefsky JM: Thiazolidinediones in the treatment of insulin The fact that all TZDs share the ability to inhibit aerobic resistance and type 2 diabetes. Diabetes 45:1661–1669, 1996 3. Plosker GL, Faulds D: Troglitazone: a review of its use in the management fuel metabolism in vitro suggests a role for the TZD ring of type 2 diabetes mellitus. Drugs 57:409–438, 1999 structure common to this group of molecules (2,12). If so, 4. Balfour JA, Plosker GL: Rosiglitazone. Drugs 57:921–930, 1999 differences in efficacy among the TZDs used are likely to 5. Fu¨ rnsinn C, Brunmair B, Meyer M, Neschen S, Furtmu¨ ller R, Roden M, rely on other structures attached to the TZD ring. In the Ku¨ hnle HF, Nowotny P, Schneider B, Waldha¨usl W: Chronic and acute effects of thiazolidinediones BM13.1258 and BM15.2054 on rat skeletal case of troglitazone, the TZD ring is attached to a molec- muscle glucose metabolism. Br J Pharmacol 128:1141–1148, 1999 ular structure also found in vitamin E, which we showed to 6. Berger J, Bailey P, Biswas C, Cullinan CA, Doebber TW, Hayes NS, be a moderate inhibitor of CO2 production in isolated Saperstein R, Smith RG, Leibowitz MD: Thiazolidinediones produce a

2314 DIABETES, VOL. 50, OCTOBER 2001 B. BRUNMAIR AND ASSOCIATES

conformational change in peroxisomal proliferator–activated receptor-␥: 21. Fulgencio J-P, Kohl C, Girard J, Pe´gorier J-P: Troglitazone inhibits fatty binding and activation correlate with antidiabetic actions in db/db mice. acid oxidation and esterification, and gluconeogenesis in isolated hepato- Endocrinology 137:4189–4195, 1996 cytes from starved rats. Diabetes 45:1556–1562, 1996 7. Fujiwara T, Yoshioka S, Yoshioka T, Ushiyama I, Horikoshi H: Character- 22. Gasic S, Bodenburg Y, Nagamani M, Green A, Urban RJ: Troglitazone ization of new oral antidiabetic agent CS-045: studies in KK and ob/ob mice inhibits progesterone production in porcine granulosa cells. Endocrinol- and Zucker fatty rats. Diabetes 37:1549–1558, 1988 ogy 139:4962–4966, 1998 8. Arakawa K, Ishihara T, Aoto M, Inamusa M, Saito A, Ikezawa K: Actions of 23. Lee MK, Olefsky JM: Acute effects of troglitazone on in vivo insulin action novel antidiabetic thiazolidinedione, T-174, in animal models of non- in normal rats. Metabolism 44:1166–1169, 1995 insulin-dependent diabetes mellitus (NIDDM) and in cultured muscle cells. 24. Ranganathan S Kern PA: Thiazolidinediones inhibit lipoprotein lipase Br J Pharmacol 125:429–436, 1998 activity in adipocytes. J Biol Chem 273:26117–26122, 1998 9. Hofmann C, Lorenz K, Colca JR: Glucose transport deficiency in diabetic 25. Inzucchi SE, Maggs DG, Spollett GR, Page SL, Rife FS, Walton V, Shulman animals is corrected by treatment with the oral antihyperglycemic agent pioglitazone. Endocrinology 129:1915–1925, 1991 GI: Efficacy and metabolic effects of metformin and troglitazone in type II 10. Auwerx J: PPAR␥, the ultimate thrifty gene. Diabetologia 42:1033–1049, diabetes mellitus. N Engl J Med 338:867–872, 1998 1999 26. Baron AD, Brechtel G, Wallace P, Edelman SV: Rates and tissue sites of 11. Lehmann JM, Moore LB, Smith-Oliver TA, Wilkinson WO, Willson TM, non-insulin and insulin-mediated glucose uptake in humans. Am J Physiol Kliewer SA: Thiazolidinedione is a high affinity ligand for peroxisome 255:E769–E774, 1988 proliferator-activated receptor ␥ (PPAR␥). J Biol Chem 270:12953–12956, 27. Auboeuf D, Rieusset J, Fajas L, Vallier P, Frering V, Riou JP, Staels B, 1995 Auwerx J, Laville M, Vidal H: Tissue distribution and quantification of the 12. Willson TM, Cobb JE, Cowan DJ, Wiethe RW, Correa ID, Prakash SR, Beck expression of mRNAs of peroxisome proliferator-activated receptors and KD, Moore LB, Kliewer SA, Lehmann JM: The structure-activity relation- liver X receptor-␣ in humans: no alteration in adipose tissue of obese and ship between peroxisome proliferator-activated receptor ␥ agonism and NIDDM patients. Diabetes 46:1319–1327, 1997 the antihyperglycemic activity of thiazolidinediones. J Med Chem 39:665– 28. Gorla-Bajszczak A, Siegrist-Kaiser C, Boss O, Burger AG, Meier CA: 668, 1996 Expression of peroxisome proliferator-activated receptors in lean and 13. Shibata T, Matsui K, Yonemori F, Wakitani K: JTT-501, a novel antidiabetic obese Zucker rats. Eur J Endocrinol 142:71–78, 2000 agent, improves insulin resistance in genetic and non-genetic insulin- 29. Loviscach M, Rehman N, Carter L, Mudaliar S, Mohadeen P, Ciaraldi TP, resistant models. Br J Pharmacol 125:1744–1750, 1998 Veerkamp JH, Henry RR: Distribution of peroxisome proliferator-activated 14. Maegawa H, Obata T, Shibata T, Fujita T, Ugi S, Morino K, Nishio Y, Kojima receptors (PPARs) in human skeletal muscle and adipose tissue: relation H, Hidaka H, Haneda M, Yasuda H, Kikkawa R, Kashiwagi A: A new to insulin action. Diabetologia 43:304–311, 2000 antidiabetic agent (JTT-501) rapidly stimulates glucose disposal rates by 30. Forman BM, Tontonoz P, Chen J, Brun RP, Spiegelman BM, Evans RM: enhancing insulin signal transduction in skeletal muscle. Diabetologia 12,14 15-deoxy-D -prostaglandin J2 is a ligand for the adipocyte determination 42:151–159, 1999 factor PPAR␥. Cell 83:803–812, 1995 15. Brown KK, Henke BR, Blanchard SG, Cobb JE, Mook R, Kaldor I, Kliewer 31. Crettaz M, Prentki M, Zanietti D, Jeanrenaud B: Insulin resistance in soleus SA, Lehmann JM, Lenhard JM, Harrington WW, Novak PJ, Faison W, Binz muscle from obese Zucker rats: involvement of several defective sites. JG, Hashim MA, Oliver WO, Brown HR, Parks DJ, Plunket KD, Tong W-Q, Biochem J 186:525–534, 1980 Menius JA, Adkinson K, Noble SA, Willson TM: A novel N-aryl tyrosine 32. Stace PB, Marchington DR, Kerbey AL, Randle PJ: Long term culture of rat activator of peroxisome proliferator-activated receptor-␥ reverses the soleus muscle in vitro: its effects on glucose utilization and insulin diabetic phenotype of the Zucker diabetic fatty rat. Diabetes 48:1415–1424, sensitivity. FEBS Lett 273:91–94, 1990 1999 33. Fu¨ rnsinn C, Ebner K, Waldha¨usl W: Failure of GLP-1(7–36)amide to affect 16. Berger J, Leibowitz MD, Doebber TW, Elbrecht A, Zhang B, Zhou G, Biswas glycogenesis in rat skeletal muscle. Diabetologia 38:864–867, 1995 C, Cullinan CA, Hayes NS, Li Y, Tanen M, Ventre J, Wu MS, Berger GD, 34. Engel RC, Jones JB: Causes and elimination of erratic blanks in enzymatic Mosley R, Marquis R, Santini C, Sahoo SP, Tolman RL, Smith RG, Moller ϩ DE: Novel peroxisome proliferator-activated receptor (PPAR) ␥ and metabolic assays involving the use of NAD in alkaline hydrazine buffers: PPAR␦ ligands produce distinct biological effects. J Biol Chem 274:6718– improved conditions for the assay of L-glutamate, L-lactate and other 6725, 1999 metabolites. Anal Biochem 88:475–484, 1978 17. Mukherjee R, Davies PJA, Crombie DL, Bischoff ED, Cesario RM, Jow L, 35. Okuno A, Tamemoto H, Tobe K, Ueki K, Mori Y, Iwamoto K, Umesono K, Hamann LG, Boehm MF, Mondon CE, Nadzan AM, Paterniti JR, Heyman Akanuma Y, Fujiwara T, Horikoshi H, Yazaki Y, Kadowaki T: Troglitazone RA: Sensitization of diabetic and obese mice to insulin by retinoid X increases the number of small adipocytes without the change of white receptor agonists. Nature 386:407–410, 1997 adipose tissue mass in obese Zucker rats. J Clin Invest 101:1354–1361, 18. El-Kebbi IM, Roser S, Pollet RJ: Regulation of glucose transport by 1998 pioglitazone in cultured muscle cells. Metabolism 43:953–958, 1994 36. Sreenan S, Keck S, Fuller T, Cockburn B, Burant CF: Effects of troglitazone 19. Fu¨ rnsinn C, Brunmair B, Neschen S, Roden M, Waldha¨usl W: Troglitazone on substrate storage and utilization in insulin-resistant rats. Am J Physiol

directly inhibits CO2 production from glucose and palmitate in isolated rat 276:E1119–E1129, 1999 skeletal muscle. J Pharmacol Exp Ther 293:487–493, 2000 37. Stevenson RW, Hutson NJ, Krupp MN, Volkmann RA, Holland GF, Eggler 20. Wang M, Wise SC, Leff T, Su T-Z: Troglitazone, an antidiabetic agent, JF, Clark DA, McPherson RK, Hall KL, Danbury BH, Gibbs EM, Kreutter inhibits cholesterol biosynthesis through a mechanism independent of DK: Actions of novel antidiabetic agent englitazone in hyperglycemic peroxisome proliferator-activated receptor-␥. Diabetes 48:254–260, 1999 hyperinsulinemic ob/ob mice. Diabetes 39:1218–1227, 1990

DIABETES, VOL. 50, OCTOBER 2001 2315