Uncoupling protein (UCP) 3 and UCP2, mitochondrial carrier proteins dissipating electrochemical gradient across the mitochondrial inner membrane, have been implicated in the regulation of energy metabolism. The UCP3 gene is expressed abundantly in the skeletal muscle, while the UCP2 gene is detected in the white adipose tissue (WAT) with diffuse localization throughout the body. Uncoupling of electron transport and ATP synthesis has been reported to increase glucose uptake, suggesting that UCP may be involved in glucose metabolism. Thiazolidinediones (TZDs), which are insulin-sensitizing agents for NIDDM, have been reported to increase energy expenditure. To elucidate the pathophysiologic significance of UCP3 and UCP2 in the effect of TZDs on glucose metabolism and energy expenditure, we examined their basal mRNA levels in the WAT, brown adipose tissue (BAT), and skeletal muscle from Wistar fatty rats, a rat model of NIDDM and obesity with leptin receptor defect, and investigated expression of the genes encoding UCP3 and UCP2 in Wistar fatty rats and in Wistar lean rats with 2-week oral administration of 3 mg x kg(-1) x day(-1) pioglitazone, a TZD derivative. Basal UCP3 mRNA levels were significantly lower (38 +/- 8, 45 +/- 13, and 76 +/- 6%) in the retroperitoneal WAT, BAT, and skeletal muscle from Wistar fatty rats than in those from Wistar lean rats, while basal UCP2 mRNA levels were significantly higher by 2.1-, 1.8-, and 2.5-fold in the subcutaneous WAT, retroperitoneal WAT, and BAT from Wistar fatty rats, respectively, than in those from Wistar lean rats. In pioglitazone-treated Wistar fatty rats, UCP3 mRNA levels were significantly increased by 2.1-, 2.0-, and 1.6-fold in the epididymal WAT, retroperitoneal WAT, and BAT, respectively, as compared with those in nontreated fatty rats. In pioglitazone-treated lean rats, UCP3 mRNA levels were significantly increased by 1.3-fold in the BAT as compared with those in nontreated lean rats. No significant change of UCP2 mRNA levels was observed in pioglitazone-treated fatty and lean rats. In addition, to examine the direct effect of TZDs on adipocytes, we examined the regulation of UCP3 and UCP2 gene expression using the primary culture of rat mature adipocytes from Sprague-Dawley rats. In rat cultured mature adipocytes, UCP3 mRNA levels were increased in a dose-responsive manner by 10(-5) to 10(-4) mol/l pioglitazone, while there was no significant change of UCP2 mRNA levels. These results clearly demonstrate that UCP3 gene expression is upregulated by TZDs in the WAT and BAT in Wistar fatty rats, an obese model with leptin receptor defect, and that adipose UCP3 gene expression is increased in response to TZDs in vitro. The present study suggests the involvement of UCP3 in the effects of TZDs on energy and glucose metabolism.