Peroxisome Proliferator-Activated Receptor-α,γ-Agonist Improves Insulin Sensitivity and Prevents Loss of Left Ventricular Function in Obese Dyslipidemic Mice
Objective— We investigated the effect of a dual peroxisome proliferator-activated receptor (PPAR)α,γ-agonist on atherosclerosis and cardiac function in mice with combined leptin and low-density lipoprotein receptor deficiency (DKO). In these mice, obesity, diabetes, and hyperlipidemia are associated with accelerated atherosclerosis and loss of cardiac function.
Methods and Results— We treated 12-week-old DKO mice with the PPARα,γ-agonist (S)-3-(4-(2-carbazol-9-yl-ethoxy) phenyl-2-ethoxy-propionic-acid) for 12 weeks. The agonist lowered free fatty acids with 42% and increased insulin sensitivity with 76%. It had no effect on plasma cholesterol and triglycerides. RT-PCR analysis showed that the agonist increased the expression of fatty acid transport protein-4, fatty acid binding protein-4, glucose transporter-4, hormone-sensitive lipase, and adiponectin in white adipose tissue that was associated with the increase in insulin sensitivity. At 24 weeks, the shortening fraction (SF) of placebo DKO mice was 30% lower than that of C57BL6 mice. The PPAR agonist increased PPARγ but not PPARα expression in the heart and prevented loss of left ventricular function. Adiponectin correlated positively with PPARγ in the heart and with SF. The agonist had no effect on atherosclerosis in the aortic arch of DKO mice.
Conclusions— The dual PPARα,γ-agonist improved insulin sensitivity without affecting cholesterol and triglycerides. This was associated with induction of PPARγ in the heart and prevention of loss of left ventricle function.
Cardiovascular diseases remain the leading cause of mortality in the Western societies. Several risk factors predispose to cardiovascular diseases, including the metabolic syndrome components diabetes, obesity, insulin resistance, dyslipidemia, and hypertension. It has been demonstrated that nuclear peroxisome proliferator-activated receptor (PPAR) deactivation (mainly obesity-related) is a key phase of metabolic syndrome initiation.1–3 The identification of the nuclear receptors PPARγ and PPARα as being the primary targets for the insulin-sensitizing thiazolidinediones and the lipid-lowering fibrates, respectively, has provided opportunities for the identification of novel compounds for the treatment of the metabolic syndrome.
Recently, we demonstrated that several metabolic syndrome components (obesity, dyslipidemia, diabetes, and hypertension) in mice with combined leptin and low-density lipoprotein receptor deficiency (DKO) are associated with increased oxidative stress, accelerated atherosclerosis, and impaired cardiovascular function.4,5 Compared with lean mice, PPARα and PPARγ expression were downregulated in DKO mice. Diet restriction-induced weight loss resulted in increased expression of PPARα and PPARγ in the adipose tissue (AT), the aortic arch, and the heart. These changes in expression were associated with increased insulin sensitivity, decreased hypertriglyceridemia, reduced mean 24-hour blood pressure and heart rate, restored circadian variations of blood pressure and heart rate, increased ejection fraction, and reduced atherosclerosis. To further study the role of PPARs in metabolism, atherosclerosis, and cardiovascular function in this mouse model, we analyzed the effect of the dual PPARα/γ agonist (S)-3-(4-(2-carbazol-9-yl-ethoxy) phenyl-2-ethoxy-propionic-acid) developed by Novo Nordisk on carbohydrate and lipid metabolism and on atherosclerosis and cardiac function. Previously, the structure, PPAR-activating properties, binding affinities to different PPAR isoforms, and pharmacokinetics of the dual PPARα/γ agonist have been described extensively.6 In vitro, this agonist has been shown to activate PPARα and PPARγ, but it had no effect on PPARδ activity. The agonist improved insulin sensitivity in diabetic db/db mice more than the PPARγ agonists pioglitazone and rosiglitazone. Moreover, the agonist lowered plasma triglycerides and cholesterol in high cholesterol-fed rats after 4 days of treatment, indicating in vivo PPARα activity.6
Breeding and genotyping of the mice, quantitative real-time RT-PCR and microarray analysis, and measurement of blood parameters, the extent of atherosclerosis in the aortic root and left ventricular (LV) function, were all performed according to previously described protocols. These protocols are available in the online supplement, available at http://atvb.ahajournals.org.
The PPARα/γ agonist (S)-3-(4-(2-carbazol-9-yl-ethoxy) phenyl-2-ethoxy-propionic-acid) was dosed as suspension in 0.2% carboxy methyl cellulose +0.4% Tween-80 in saline. Fresh suspensions were made for 7 days of dosing and kept at +4°C. The agonist was administered orally (10 mL/kg; 3 mg/kg per day) for 12 weeks starting at the age of 12 weeks (n=9). At this age, DKO mice have no detectable atherosclerosis, and their LV function is not different from that of lean mice. Placebo mice (n=10) received the suspension solution only. Experimental procedures in animals were performed in accordance with protocols approved by the institutional animal care and research advisory committee.
Real Time RT-PCR and Microarray Analysis
The level of mRNA expression for PPARα and PPARγ in different tissues from C57BL6, placebo, and PPAR agonist-treated DKO mice at 24 weeks was measured by quantitative real-time RT-PCR. Because PPARδ expression might compensate for variations in PPARα and PPARγ expression,7 we also measured PPARδ. These protocols are available in the online supplement. We confirmed the identity of the PPARα and PPARγ obtained by RT-PCR on cDNA from AT on 15% polyacrylamide gels before and after digestion with AleI, respectively BglII.
We searched for genes that could explain the effect of the PPAR agonist on free fatty acids (FFAs), glucose tolerance, and insulin sensitivity in the absence of an effect on triglycerides. To screen for candidate genes for quantitative real-time RT-PCR analysis, we used microarray analysis to compare gene expression in the AT from PPAR agonist-treated mice (n=3) with that in free-fed DKO mice (pooled RNA from 3 mice). We also compared gene expression in DKO mice after weight loss (n=3), resulting from restricted food intake, with the same free-fed mice.5 We selected these mice because weight loss was associated with a similar increase of the insulin sensitivity. In contrast to PPAR agonist treatment, weight loss was associated with a decrease in triglycerides. The selection of candidate genes and RT-PCR analysis of these selected genes was performed as described in the detailed protocol (see online supplement).
Groups were compared with Kruskal-Wallis test (Graph Pad Prism version 3.02) followed by Dunn multiple comparisons test. Linear regression analysis was performed using the Statistical Package for the Social Sciences (SPSS for Windows, release 10.0.5). A P value of <0.05 was considered statistically significant.
Blood Analysis and Weight
The PPAR agonist lowered FFAs with 42%, insulin with 56%, glucose with 45%, homeostasis model assessment with 76%, and increased glucose tolerance with 55%. The PPAR agonist had no effect on weight, cholesterol, and triglycerides (Table 1). We excluded that the lack of effect on cholesterol and triglycerides was attributable to a too low dose. A higher dose of 10 mg/kg per day for 12 weeks (n=6), which decreased FFA levels to 0.13±0.032 mmol/L, had no effect on cholesterol and triglycerides. Plasma adiponectin levels increased 4.3-fold after 12-week treatment with the agonist (52±14 μg/mL versus 12±4.8 μg/mL) but were not affected by placebo (13±8.2 μg/mL versus 15±6.0 μg/mL).
RT-PCR Analysis of PPAR Expression
Compared with placebo DKO mice, PPARγ expression in the heart of PPAR agonist-treated DKO mice was 1.7-fold higher (Figure 1A); that in skeletal muscle was 1.5-fold higher (2.8±0.6 versus 1.8±0.5; agonist versus placebo treated). The agonist had no effect on PPARγ expression in both AT (Figure 1B) and aortic arch (Figure 1C) and decreased PPARγ expression in the liver by 38% (Figure 1D). PPARα expression was 2.1-fold higher in both AT (Figure 1B) and aortic arch (Figure 1C) of PPAR agonist-treated compared with placebo DKO mice. The agonist had no effect on PPARα expression in skeletal muscle (1.2±0.3 versus 1.3±0.3), heart (Figure 1A), or liver (Figure 1D).
We confirmed the identity of the PPARα and PPARγ in the AT by polyacrylamide gel electrophoresis of the RT-PCR product (Figure 2). As expected, the PPARα cDNA migrated as a 65-bp fragment, whereas the PPARγ cDNA migrated as a 106-bp fragment. Digestion of PPARα cDNA with AleI yielded a 39- and a 26-bp fragment. AleI did not cleave PPARγ cDNA. Digestion of PPARγ cDNA with BglII yielded a 61- and a 45-bp fragment. BglII did not cleave PPARα cDNA.
Compared with C57BL6 mice, PPARδ expression in DKO mice was 50% lower (ratio is 0.50±0.16) in AT and 31% lower in the aortic arch (ratio is 0.69±0.15), whereas in the skeletal muscle, heart and liver PPARδ expression was comparable (ratio is &1). In all tissues, the agonist had no effect on PPARδ expression, which is in agreement with the lack of in vitro PPARδ-activating capacity of the agonist. The PPAR agonist had no effect on the expression of the housekeeping gene β-actin.
Microarray and RT-PCR Analysis
We searched for genes that were differentially expressed in the AT of PPAR agonist-treated mice and could explain the effect of the PPAR agonist on FFAs, glucose tolerance, and insulin sensitivity in the absence of an effect on triglycerides and cholesterol, with microarray analysis of RNA extracts from AT. Supplemental Table I (see online supplement) shows that 6387 genes were differentially expressed in the AT of PPAR agonist-treated mice compared with free-fed, placebo DKO mice. The expression of these genes in diet-restricted mice is also shown. Using the Gene Ontology database, we searched for genes that could explain the difference in fatty acids and glucose tolerance and insulin sensitivity between PPAR agonist-treated and placebo mice. We identified 4 genes: the fatty acid transport protein-4 (FATP-4), hormone sensitive lipase (HSL), fatty acid binding protein-4 (FABP-4), and glucose transporter-4 (GLUT-4; Table 2). All 4 genes were similarly differentially expressed in diet-restricted compared with the same free-fed mice. We then compared the expression of these 4 genes with RT-PCR analysis. Although adiponectin was not selected as a candidate gene by microarray analysis, we included adiponectin in the RT-PCR analysis because PPAR agonist treatment increased plasma adiponectin level. RT-PCR showed increased expression of FATP-4, HSL, FABP-4, GLUT-4 (Table 2), and adiponectin (0.46±0.16 versus 0.24±0.15; PPAR agonist versus placebo; P<0.05) in agonist-treated DKO mice. PPARα was also not among the differentially expressed genes, although RT-PCR analysis performed before microarray analysis showed significant upregulation of PPARα in the AT of PPAR agonist-treated mice.
The expression of FATP-4, FABP-4, GLUT-4, adiponectin and HSL correlated inversely with homeostasis model assessment (R=−0.74, −0.77, −0.65, −0.65 and −0.62, respectively; P<0.05 for all). Adiponectin expression also correlated with plasma adiponectin concentrations (R=0.55, P<0.05). We also searched for genes that could explain the difference in triglycerides between PPAR agonist-treated and diet-restricted mice. Microarray analysis showed a higher expression of LPL and caveolin-1, and a lower expression of phospholipid transfer protein (PLTP) in the AT of DKO mice after weight-loss (Table 2). The PPAR agonist had no effect on LPL expression (confirmed by RT-PCR), lowered caveolin-1 expression to some extent (effect not observed in RT-PCR), and increased even PLTP expression (confirmed with RT-PCR; Table 2).
Microarray analysis did not show differences in PPARγ (0.83±0.27) and PPARδ (1.3±0.09) between agonist-treated and placebo mice. Microarray analysis did also not show differences between PPAR agonist-treated and placebo-treated mice for genes that play a role in cholesterol metabolism: stearoyl-coenzyme A desaturase (0.84±0.02), 3-hydroxy-3-methylglutaryl-coenzyme A reductase (1.3±0.03), 7-dehydrocholesterol reductase (0.93±0.23), lecithin:cholesterol acyl transferase (1.0±0.08), or phospholipase A2 (0.96±0.18), among others.
At 24 weeks of age, heart rate was comparable in C57BL6 and in placebo-treated and PPAR agonist-treated DKO mice. There was no difference in both septum and posterior wall thickness between the 3 groups. However, LV end diastolic, LV end systolic (LVES) diameter, and LV mass of placebo DKO mice was significantly higher than that of C57BL6 mice (supplemental Table II). The shortening fraction (SF) was significantly lower in placebo DKO mice than in C57BL6 mice (Figure 3A). The PPAR agonist increased SF (Figure 3A) as a result of a lower LVES diameter (Figure 3B). A higher dose of 10 mg/kg per day for 12 weeks (n=6) had a similar effect on SF. The agonist had no effect on LV mass (supplemental Table II).
There was no significant difference in plaque volume when we compared the PPAR agonist-treated mice with placebo-treated mice (0.095±0.070 mm3 versus 0.088±0.038 mm3, respectively). The higher dose also had no effect.
We investigated the effect of a dual PPARα/γ agonist on atherosclerosis and cardiovascular function in DKO mice. The dual agonist lowered FFAs and improved glucose tolerance and insulin sensitivity in the absence of an effect on cholesterol and triglycerides. In addition to GLUT-4 and HSL, we identified FATP-4 and FABP-4 as significantly upregulated genes in the AT of PPAR-agonist treated mice that were associated with the improved insulin sensitivity. The fact that these genes were also upregulated after weight loss, which was associated with similar changes in insulin sensitivity, further underscored their role. The PPAR agonist prevented loss of LV function in the absence of an inhibition of atherosclerosis. The agonist increased PPARγ expression in the heart and adiponectin in the AT. Both PPARγ and adiponectin correlated with LV function and with each other. Our data suggest a relationship between insulin sensitivity, adiponectin in AT and PPARγ in the heart, and cardiac function, independent of cholesterol and triglycerides.
Effect of the Agonist on FFAs, Glucose Tolerance, and Insulin Sensitivity in the Absence of an Effect on Triglycerides
The general concept is that the ability of PPAR agonists to upregulate lipid metabolism in liver (PPARα) and white AT (PPARγ, and to a lesser extent PPARα), thereby decreasing circulating lipids, is central to their insulin-sensitizing effects within skeletal muscle and heart.8–10 Here, the PPAR agonist had no effect on triglycerides and total cholesterol but lowered FFA levels and improved glucose tolerance and insulin sensitivity.
In the AT, FATP-4, FABP-4, GLUT-4, HSL, and adiponectin are known regulators of glucose tolerance and insulin sensitivity that are under the transcriptional control of PPARs.11–16 Reducing the availability of FFAs from the AT to liver and muscles is a pivotal component of the insulin-sensitizing mechanism of PPAR agonists in the AT. The improved insulin sensitivity in agonist-treated DKO mice can thus result from a decrease in circulating FFAs by increased uptake in the AT, supported by increased expression of FATP-4 and FABP-4. The increase in FATP-4 and FABP-4 was associated with an increase in PPARα expression, supporting its role in regulating FFA uptake in the AT.
Like FFAs, adiponectin is an important signaling molecule regulating insulin sensitivity in muscle and liver by increasing FFA oxidation. The agonist increased the expression of adiponectin in the AT, and this was associated with an increase of plasma adiponectin. The increase in adiponectin also indicates that the insulin sensitivity of the AT itself is improved, supported by increased expression of GLUT-4 in the AT of agonist-treated mice. Interestingly, in diet-restricted mice, a comparable upregulation of FATP-4, HSL, FABP-4, and GLUT-4 and an increase in adiponectin levels was also associated with improved insulin sensitivity.
Increased PPARγ expression in the liver of ob/ob mice compensated for the impaired insulin resistance in these mice.17 In the present study, we found a similar overexpression of PPARγ in the liver of DKO as in ob/ob mice. The increased insulin sensitivity in agonist-treated DKO mice was associated with a 38% lower liver expression of PPARγ. This suggests that the increase in insulin sensitivity in agonist-treated mice caused the liver to stop compensating for the insulin resistance by decreasing PPARγ expression. Also, the increase in adiponectin could lead to improved hepatic fat mobilization and hepatic insulin sensitivity, as was seen in patients with type 2 diabetes.18
PPARα expression in the liver of placebo DKO mice was comparable to that in C57BL6 mice (ratio is &1) indicating that obesity, dyslipidemia, and insulin resistance in DKO mice was not associated with a change in liver PPARα expression. Thus, PPARα in the liver is most likely not responsible for the observed dyslipidemia in DKO mice, and lack of effect of the agonist on PPARα expression in the liver is thus most likely not the cause of the persistent dyslipidemia in agonist-treated DKO mice.
In diet-restricted mice, increased expression of LPL and of caveolin-1 and decreased expression of PLTP were associated with an improvement of dyslipidemia. Lack of effect of the agonist on dyslipidemia could partially be explained by the lack of effect on the expression of PPARγ and LPL and the somewhat lower expression of caveolin-1 (certainly compared with diet-restricted mice) and the even higher expression of PLTP in PPAR agonist-treated mice.
We also showed that PPARδ, a key regulator of fat burning in the AT that opposes the fat-storing function of PPARγ (and PPARα),19 is downregulated in the AT of placebo-treated DKO mice, indicating that PPARδ might also play an important role in the observed dyslipidemia in DKO mice. The agonist had no effect on PPARδ expression, which is in agreement with the lack of effect on dyslipidemia.
Direct Effect of the Agonist on LV Function: Central Role of PPARγ
Muscle-specific disruption of PPARγ results in a state of severe insulin resistance,20 and activation of PPARγ corrects impaired muscle insulin action.21,22 The higher PPARγ expression in the heart of agonist-treated DKO mice was observed in association with improved insulin sensitivity. Our data thus suggest that the agonist improved the insulin resistance in the heart of DKO mice via upregulation of PPARγ. In other studies, PPARγ also improved LV function in diabetic rats23 and improved recovery of LV function after regional ischemia in pigs.24
Gilde et al25 showed that PPARα and PPARβ/δ, but not PPARγ, modulate the expression of genes involved in cardiac lipid metabolism. Both PPARα and PPARδ expression in the heart of placebo-treated DKO mice were not different from that in lean C57BL6 mice, and the agonist had no effect on their expression in the heart of DKO mice despite the agonist improved cardiac function, suggesting that they are not important for regulating cardiac function in DKO.
Possible Indirect Effect of the Agonist on LV Function: Central Role of Adiponectin
Insulin sensitivity correlated with PPARγ expression in the heart. Adiponectin enhances insulin sensitivity and increases FFA oxidation in muscles.11,26 In the present study, PPAR agonist treatment resulted in an increase in plasma adiponectin levels that was associated with improved insulin sensitivity. Adiponectin levels also correlated positively with PPARγ expression in the heart, indicating that the increase in PPARγ in the heart of agonist-treated DKO mice could also be the result of the positive effect of the agonist on plasma adiponectin levels. Our data are thus in agreement with data in man, monkey, and mouse, demonstrating that adiponectin is an insulin-sensitizing adipocytokine.27–30 The increase in adiponectin in agonist-treated mice suggested a link between insulin sensitivity in the AT and insulin sensitivity in the heart, and could therefore have an effect on LV function. Indeed, the increase in adiponectin correlated with improved LV function.
Lack of Effect on Atherosclerosis
Previously, we showed that weight loss in DKO mice resulted in lower plasma triglycerides that were associated with lower plaque volume.5 In the present study, the agonist had no effect on triglycerides and plaque volume. In aggregate, the 2 sets of data suggest that the triglyceride level is the predominant determinant of plaque volume in DKO mice. Recently, Hennuyer et al31 also showed that improvement of insulin sensitivity in the absence of an effect on dyslipidemia in mice treated with a PPARγ agonist was not sufficient to inhibit atherosclerosis. However, use of a PPARα agonist did improve dyslipidemia and reduced atherosclerosis.
Previously, weight loss induced an increase in both PPARα and PPARγ in the aortic arch of DKO mice. In contrast, the agonist only increased PPARα. Thus, the lack of effect of the agonist on PPARγ expression in the aortic arch could also explain the lack of its effect on atherosclerosis. The role of PPARδ in the aortic arch is poorly understood. In DKO mice, atherosclerosis was clearly associated with a downregulation of PPARδ, supporting its antiatherogenic action. The agonist had no effect on PPARδ expression in the aortic arch.
Here we show that next to PPARα and PPARγ, PPARδ is also downregulated in the AT and the aortic arch of the DKO mice. Because PPARα/γ agonist treatment had no effect on PPARδ expression, we were unable to determine its role in the development of obesity, dyslipidemia, and insulin resistance and the associated atherosclerosis and impaired LV function in DKO mice. In the future, this mouse model might prove to be a useful tool to further investigate the role of PPARδ in the development of the metabolic syndrome. Microarray analysis on a small number of mice was performed to search for genes that could explain changes in metabolic components and that had to be confirmed with RT-PCR analysis. The fact that important genes (eg, PPARα and adiponectin) were not picked up in microarray analysis, possibly because of the wide variation in signal, underscores the limitation of this approach. On the other hand, the fact that we identified FATP-4 and FABP-4 as crucial genes supports its relevance.
PPAR agonist-induced lowering of FFAs and improvement in glucose tolerance and insulin sensitivity in the absence of an effect on cholesterol and triglycerides prevented loss of ventricle function but did not inhibit atherosclerosis in obese, dyslipidemic mice. Interestingly, 4 genes that were upregulated in the AT of PPAR agonist-treated mice (FATP-4, FABP-4, HSL, and GLUT-4) and that were associated with the improved insulin sensitivity were also upregulated in the AT of DKO mice after weight loss that resulted in a similar improvement of insulin sensitivity. An important finding is that this improvement was obtained in the absence of a decrease in triglycerides in PPAR agonist-treated mice, in contrast with diet-restricted mice. Furthermore, our data suggest a relationship between adiponectin and PPARγ in the heart, insulin sensitivity, and cardiac function, independent of cholesterol and triglycerides.
This study was supported in part by the Fonds voor Wetenschappelijk Onderzoek-Vlaanderen (Program G027604) and Interuniversity Attraction Poles Program—Belgian Science Policy (P5/02). The dual PPARα,γ agonist 3q was a kind gift from Per Sauerberg, Novo Nordisk A/S. We thank Hilde Bernar, Els Deridder, and Michèle Landeloos for excellent technical assistance.
- Received March 29, 2005.
- Accepted January 17, 2006.
Ferre P. The biology of peroxisome proliferator-activated receptors: relationship with lipid metabolism and insulin sensitivity. Diabetes. 2004; 53: S43–S50.
Mertens A, Verhamme P, Bielicki JK, Phillips MC, Quarck R, Verreth W, Stengel D, Ninio E, Navab M, Mackness B, Mackness M, Holvoet P. Increased low-density lipoprotein oxidation and impaired high-density lipoprotein antioxidant defense are associated with increased macrophage homing and atherosclerosis in dyslipidemic obese mice: LCAT gene transfer decreases atherosclerosis. Circulation. 2003; 107: 1640–1646.
Verreth W, De KD, Pelat M, Verhamme P, Ganame J, Bielicki JK, Mertens A, Quarck R, Benhabiles N, Marguerie G, Mackness B, Mackness M, Ninio E, Herregods MC, Balligand JL, Holvoet P. Weight loss-associated induction of peroxisome proliferator-activated receptor-alpha and peroxisome proliferator-activated receptor-gamma correlate with reduced atherosclerosis and improved cardiovascular function in obese insulin-resistant mice. Circulation. 2004; 110: 3259–3269.
Sauerberg P, Pettersson I, Jeppesen L, Bury PS, Mogensen JP, Wassermann K, Brand CL, Sturis J, Woldike HF, Fleckner J, Andersen AS, Mortensen SB, Svensson LA, Rasmussen HB, Lehmann SV, Polivka Z, Sindelar K, Panajotova V, Ynddal L, Wulff EM. Novel tricyclic-alpha-alkyloxyphenylpropionic acids: dual PPARalpha/gamma agonists with hypolipidemic and antidiabetic activity. J Med Chem. 2002; 45: 789–804.
Muoio DM, MacLean PS, Lang DB, Li S, Houmard JA, Way JM, Winegar DA, Corton JC, Dohm GL, Kraus WE. Fatty acid homeostasis and induction of lipid regulatory genes in skeletal muscles of peroxisome proliferator-activated receptor (PPAR) alpha knock-out mice. Evidence for compensatory regulation by PPAR delta. J Biol Chem. 2002; 277: 26089–26097.
Wolf G. Adiponectin: a regulator of energy homeostasis. Nutr Rev. 2003; 61: 290–292.
Karlsson M, Thorn H, Parpal S, Stralfors P, Gustavsson J. Insulin induces translocation of glucose transporter GLUT4 to plasma membrane caveolae in adipocytes. FASEB J. 2002; 16: 249–251.
Miura T, Suzuki W, Ishihara E, Arai I, Ishida H, Seino Y, Tanigawa K. Impairment of insulin-stimulated GLUT4 translocation in skeletal muscle and adipose tissue in the Tsumura Suzuki obese diabetic mouse: a new genetic animal model of type 2 diabetes. Eur J Endocrinol. 2001; 145: 785–790.
Bajaj M, Suraamornkul S, Piper P, Hardies LJ, Glass L, Cersosimo E, Pratipanawatr T, Miyazaki Y, DeFronzo RA. Decreased plasma adiponectin concentrations are closely related to hepatic fat content and hepatic insulin resistance in pioglitazone-treated type 2 diabetic patients. J Clin Endocrinol Metab. 2004; 89: 200–206.
Zhu P, Lu L, Xu Y, Schwartz GG. Troglitazone improves recovery of left ventricular function after regional ischemia in pigs. Circulation. 2000; 101: 1165–1171.
Gilde AJ, van der Lee KA, Willemsen PH, Chinetti G, van der Leij FR, van d, V, Staels B, van BM. Peroxisome proliferator-activated receptor (PPAR) alpha and PPARbeta/delta, but not PPARgamma, modulate the expression of genes involved in cardiac lipid metabolism. Circ Res. 2003; 92: 518–524.
Hotta K, Funahashi T, Bodkin NL, Ortmeyer HK, Arita Y, Hansen BC, Matsuzawa Y. Circulating concentrations of the adipocyte protein adiponectin are decreased in parallel with reduced insulin sensitivity during the progression to type 2 diabetes in rhesus monkeys. Diabetes. 2001; 50: 1126–1133.
Yamauchi T, Kamon J, Waki H, Terauchi Y, Kubota N, Hara K, Mori Y, Ide T, Murakami K, Tsuboyama-Kasaoka N, Ezaki O, Akanuma Y, Gavrilova O, Vinson C, Reitman ML, Kagechika H, Shudo K, Yoda M, Nakano Y, Tobe K, Nagai R, Kimura S, Tomita M, Froguel P, Kadowaki T. The fat-derived hormone adiponectin reverses insulin resistance associated with both lipoatrophy and obesity. Nat Med. 2001; 7: 941–946.
Hennuyer N, Tailleux A, Torpier G, Mezdour H, Fruchart JC, Staels B, Fievet C. PPARalpha, but not PPARgamma, activators decrease macrophage-laden atherosclerotic lesions in a nondiabetic mouse model of mixed dyslipidemia. Arterioscler Thromb Vasc Biol. 2005; 25: 1897–1902.