The effect of pioglitazone (PIO) on plasma adiponectin concentration, endogenous glucose production (EGP), and hepatic fat content (HFC) was studied in 11 type 2 diabetic patients (age, 52 ؎ 2 yr; body mass index, 29.6 ؎ 1.1 kg/m 2 ; HbA 1c , 7.8 ؎ 0.4%). HFC (magnetic resonance spectroscopy) and basal plasma adiponectin concentration were quantitated before and after PIO (45 mg/d) for 16 wk. Subjects received a 3-h euglycemic insulin (100 mU/m 2 ⅐min) clamp combined with 3-[ 3 H] glucose infusion to determine rates of EGP and tissue glucose disappearance (Rd) before and after PIO. PIO reduced fasting plasma glucose (10.0 ؎ 0.7 to 7.2 ؎ 0.6 mmol/liter, P < 0.01) and HbA 1c (7.8 ؎ 0.4 to 6.5 ؎ 0.3%, P < 0.01) despite increased body weight (83.0 ؎ 3.0 to 86.4 ؎ 3.0 kg, P < 0.01). PIO improved Rd (6.6 ؎ 0.6 vs. 5.2 ؎ 0.5 mg/kg⅐min, P < 0.005) and reduced EGP (0.23 ؎ 0.04 to 0.05 ؎ 0.02 mg/kg⅐min, P < 0.01) during the 3-h insulin clamp. After PIO treatment, HFC decreased from 21.3 ؎ 4.2 to 11.0 ؎ 2.4% (P < 0.01), and plasma adiponectin increased from 7 ؎ 1 to 21 ؎ 2 g/ml (P < 0.0001). Plasma adiponectin concentration correlated negatively with HFC (r ‫؍‬ ؊0.60, P < 0.05) and EGP (r ‫؍‬ ؊0.80, P < 0.004) and positively with Rd before (r ‫؍‬ 0.68, P < 0.02) pioglitazone treatment; similar correlations were observed between plasma adiponectin levels and HFC (r ‫؍‬ ؊0.65, P < 0.03) and Rd after (r ‫؍‬ 0.70, P ‫؍‬ 0.01) pioglitazone treatment. EGP was almost completely suppressed after pioglitazone treatment; taken collectively, plasma adiponectin concentration, before and after pioglitazone treatment, still correlated negatively with EGP during the insulin clamp (r ‫؍‬ ؊0.65, P < 0.001). In conclusion, PIO treatment in type 2 diabetes causes a 3-fold increase in plasma adiponectin concentration. The increase in plasma adiponectin is strongly associated with a decrease in hepatic fat content and improvements in hepatic and peripheral insulin sensitivity. The increase in plasma adiponectin concentration after thiazolidinedione therapy may play an important role in reversing the abnormality in hepatic fat mobilization and the hepatic/muscle insulin resistance in patients with type 2 diabetes. (J Clin Endocrinol Metab 89: 200 -206, 2004)

The effect of pioglitazone on splanchnic glucose uptake (SGU), endogenous glucose production (EGP), and hepatic fat content was studied in 14 type 2 diabetic patients (age 50 ؎ 2 years, BMI 29.4 ؎ 1.1 kg/m 2 , HbA 1c 7.8 ؎ 0.4%). Hepatic fat content (magnetic resonance spectroscopy) and SGU (oral glucose load-insulin clamp technique) were quantitated before and after pioglitazone (45 mg/day) therapy for 16 weeks. Subjects received a 7-h euglycemic insulin (100 mU ⅐ m ؊2 ⅐ min ؊1 ) clamp, and a 75-g oral glucose load was ingested 3 h after starting the insulin clamp. Following glucose ingestion, the steady-state glucose infusion rate during the insulin clamp was decreased appropriately to maintain euglycemia. SGU was calculated by subtracting the integrated decrease in glucose infusion rate during the 4 h after glucose ingestion from the ingested glucose load. 3-[ 3 H]glucose was infused during the initial 3 h of the insulin clamp to determine rates of EGP and glucose disappearance (R d ). Pioglitazone reduced fasting plasma glucose (10.0 ؎ 0.7 to 7.5 ؎ 0.6 mmol/l, P < 0.001) and HbA 1c (7.8 ؎ 0.4 to 6.7 ؎ 0.3%, P < 0.001) despite increased body weight (83 ؎ 3 to 86 ؎ 3 kg, P < 0.001). During the 3-h insulin clamp period before glucose ingestion, pioglitazone improved R d (6.9 ؎ 0.5 vs. 5.2 ؎ 0.5 mg ⅐ kg ؊1 ⅐ min ؊1 , P < 0.001) and insulinmediated suppression of EGP (0.21 ؎ 0.04 to 0.06 ؎ 0.02 mg ⅐ kg ؊1 ⅐ min ؊1 , P < 0.01). Following pioglitazone treatment, hepatic fat content decreased from 19.6 ؎ 3.6 to 10.4 ؎ 2.1%, (P < 0.005), and SGU increased from 33.0 ؎ 2.8 to 46.2 ؎ 5.1% (P < 0.005). Pioglitazone treatment in type 2 diabetes 1) decreases hepatic fat content and improves insulin-mediated suppression of EGP and 2) augments splanchnic and peripheral tissue glucose uptake. Improved splanchnic/peripheral glucose uptake and enhanced suppression of EGP contribute to the improvement in glycemic control in patients with type 2 diabetes. Diabetes 52:1364 -1370, 2003 T he splanchnic tissues play a pivotal role in the maintenance of normal glucose homeostasis (1). Hyperglycemia, plasma free fatty acid (FFA) concentration, and route of glucose administration all exert independent effects on splanchnic glucose uptake (SGU). When glucose is administered intravenously, the resultant hyperglycemia enhances SGU in proportion to the increase in plasma glucose concentration such that the splanchnic glucose clearance remains unchanged (2,3). This mass-action effect of hyperglycemia to augment SGU is dependent upon maintained portal insulin levels (2-5,8). Insulin per se does not increase SGU (2,5). Studies by DeFronzo and colleagues (3,5) in humans and by Cherrington and colleagues (6,7) in dogs have shown that the gastrointestinal/portal route of glucose administration has a specific enhancing effect on SGU. Thus, following glucose ingestion, the fractional, as well as absolute rate of glucose uptake by the splanchnic tissues is significantly greater than the combined effects of hyperinsulinemia plus hyperglycemia created...

To investigate the effect of a sustained (7-day) decrease in plasma free fatty acid (FFA) concentrations on insulin action and intramyocellular long-chain fatty acyl-CoAs (LCFA-CoAs), we studied the effect of acipimox, a potent inhibitor of lipolysis, in seven type 2 diabetic patients (age 53 M ultiple disturbances in free fatty acid (FFA) metabolism, including daylong elevated plasma FFA levels and accelerated rates of lipolysis, are a characteristic feature of type 2 diabetes (1-4). Elevated plasma FFA concentrations impair glucose metabolism by inhibiting the more proximal steps of insulin action in muscle (5-11) as well by augmenting basal hepatic gluconeogenesis and impairing the suppression of hepatic glucose production by insulin (7,12,13). In addition to having FFAs circulating in plasma in increased amounts, type 2 diabetic and obese patients have increased stores of triglycerides in muscle (14) and liver (15), which correlates closely with the presence of insulin resistance in these tissues. It is now recognized that the triglycerides in liver and muscle are in a state of constant turnover and that the metabolites of intracellular FFA metabolism (i.e., cytosolic long-chain fatty acyl-CoAs [LCFA-CoAs]) can impair insulin action in both liver and muscle (16,17). Cytosolic LCFA-CoA esters are intermediates in lipid synthesis/oxidation and are primarily derived from circulating fatty acids or intramuscular lipid sources such as triglycerides and phospholipids. With respect to insulin action, rodents fed a high-fat diet manifest increased intramuscular LCFA-CoA content, which is associated with insulin resistance (18). In contrast, weight loss in morbidly obese humans is associated with a reduction in intramuscular LCFA-CoA levels and enhanced insulin action (14,19).؎Adipocytes function not only as fat depots that release FFAs but also as endocrine organs that release hormones and cytokines in response to specific extracellular stimuli or changes in metabolic status. These secreted proteins, which include tumor necrosis factor-␣, interleukin 6, leptin, resistin, adiponectin, and others, perform a variety of diverse functions and have been referred to collectively as "adipocytokines." Plasma levels of adiponectin are reFrom the

Aims/hypothesis The aim of the study was to examine the effects of pioglitazone (PIO), a peroxisome proliferatoractivated receptor (PPAR)-γ agonist, and fenofibrate (FENO), a PPAR-α agonist, as monotherapy and in combination on glucose and lipid metabolism. Subjects and methods Fifteen type 2 diabetic patients received FENO (n=8) or PIO (n=7) for 3 months, followed by the addition of the other agent for 3 months in an openlabel study. Subjects received a 4 h hyperinsulinaemiceuglycaemic clamp and a hepatic fat content measurement at 0, 3 and 6 months. Results Following PIO, fasting plasma glucose (FPG) (p<0.05) and HbA 1c (p<0.01) decreased, while plasma adiponectin (AD) (5.5±0.9 to 13.8±3.5 μg/ml [SEM], p<0.03) and the rate of insulin-stimulated total-body glucose disposal (R d ) (23.8 ± 3.8 to 40.5 ± 4.4 μmol kg −1 min −1 , p < 0.005) increased. After FENO, FPG, HbA 1c , AD and R d did not change. PIO reduced fasting NEFA (784±53 to 546± 43 μmol/l, p<0.05), triacylglycerol (2.12±0.28 to 1.61± 0.22 mmol/l, p<0.05) and hepatic fat content (20.4±4.8 to 10.2±2.5%, p<0.02). Following FENO, fasting NEFA and hepatic fat content did not change, while triacylglycerol decreased (2.20± 0.14 to 1.59± 0.13 mmol/l, p<0.01).Addition of FENO to PIO had no effect on R d , FPG, HbA 1c , NEFA, hepatic fat content or AD, but triacylglycerol decreased (1.61±0.22 to 1.00±0.15 mmol/l, p<0.05). Addition of PIO to FENO increased R d (24.9±4.4 to 36.1± 2.2 μmol kg −1 min −1 , p<0.005) and AD (4.1±0.8 to 13.1± 2.5 μg/ml, p<0.005) and reduced FPG (p<0.05), HbA 1c (p<0.05), NEFA (p<0.01), hepatic fat content (18.3±3.1 to 13.5±2.1%, p<0.03) and triacylglycerol (1.59±0.13 to 0.96±0.9 mmol/l, p<0.01). Muscle adenosine 5′-monophosphate-activated protein kinase (AMPK) activity did not change following FENO; following the addition of PIO, muscle AMPK activity increased significantly (phosphorylated AMPK:total AMPK ratio 1.2±0.2 to 2.2±0.3, p<0.01). Conclusions/interpretation We conclude that PPAR-α therapy has no effect on NEFA or glucose metabolism and that addition of a PPAR-α agonist to a PPAR-γ agent causes a further decrease in plasma triacylglycerol, but has no effect on NEFA or glucose metabolism.

OBJECTIVES:To study the effect of pioglitazone (PIO) on plasma resistin concentration, endogenous glucose production (EGP), and hepatic fat content (HFC) in patients with type II diabetes (T2DM). SUBJECTS: A total of 13 T2DM patients (age ¼ 5172 y, BMI ¼ 29.771.1 kg/m 2 , HbA 1c ¼ 8.070.5%). METHODS: HFC (magnetic resonance spectroscopy) and basal plasma resistin concentration were quantitated before and after PIO treatment (45 mg/day) for 16 weeks. Subjects received a 3 h euglycemic insulin (100 mU/m 2 /min) clamp with 3-[ 3 H] glucose to determine rates of EGP and tissue glucose disappearance (Rd) before and after PIO. RESULTS: PIO reduced fasting plasma glucose (10.370.7 to 7.670.6 mmol/l, Po0.001) and HbA 1c (8.070.4 to 6.870.3%, Po0.001) despite increased body weight (83.273.4 to 86.373.4 kg, Po0.001). PIO improved Rd (4.970.4 to 6.670.5 mg/ kg/min, Po0.005) and reduced EGP (0.2270.04 to 0.0670.02 mg/kg/min, Po0.01) during the insulin clamp. Following PIO, HFC decreased from 21.173.5 to 11.272.1% (Po0.005), and plasma resistin decreased from 5.370.6 to 3.570.3 ng/ml (Po0.01). Plasma resistin concentration correlated positively with HFC before (r ¼ 0.58, Po0.05) and after (r ¼ 0.55, Po0.05) PIO treatment. Taken collectively, plasma resistin concentration, before and after PIO treatment, correlated positively with hepatic fat content (r ¼ 0.66, Po0.001) and EGP during the insulin clamp (r ¼ 0.41, Po0.05). However, the plasma resistin concentration did not correlate with whole body glucose disposal (Rd) during the insulin clamp either before (r ¼ À0.18, P ¼ NS) or after (r ¼ À0.13, P ¼ NS) PIO treatment. CONCLUSIONS: PIO treatment in T2DM causes a significant decrease in plasma resistin concentration. The decrease in plasma resistin is positively correlated with the decrease in hepatic fat content and improvement in hepatic insulin sensitivity.