Leptin is a circulating hormone that is expressed abundantly and specifically in the adipose tissue. It is involved in the regulation of energy homeostasis, as well as the neuroendocrine and reproductive systems. Here, we demonstrate production of leptin by nonadipose tissue, namely, placental trophoblasts and amnion cells from uteri of pregnant women. We show that pregnant women secrete a considerable amount of leptin from the placenta into the maternal circulation as compared with nonpregnant obese women. Leptin production was also detected in a cultured human choriocarcinoma cell line, BeWo cells, and was augmented during the course of forskolin-induced differentiation of cytotrophoblasts into syncytiotrophoblasts. Plasma leptin levels were markedly elevated in patients with hydatidiform mole or choriocarcinoma and were reduced after surgical treatment or chemotherapy. Leptin is also produced by primary cultured human amnion cells and is secreted into the amniotic fluid. The present study provides evidence for leptin as a novel placenta-derived hormone in humans and suggests the physiologic and pathophysiologic significance of leptin in normal pregnancy and gestational trophoblastic neoplasms.

Excess of body fat, or obesity, is a major health problem and confers a higher risk of cardiovascular and metabolic disorders such as diabetes, hypertension, and coronary heart disease. Leptin is an adipocyte-derived satiety factor that plays an important role in the regulation of energy homeostasis, and its synthesis and secretion are markedly increased in obese subjects. To explore the metabolic consequences of an increased amount of leptin on a long-term basis in vivo, we generated transgenic skinny mice with elevated plasma leptin concentrations comparable to those in obese subjects. Overexpression of leptin in the liver has resulted in complete disappearance of white and brown adipose tissue for a long period of time in mice. Transgenic skinny mice exhibit increased glucose metabolism accompanied by the activation of insulin signaling in the skeletal muscle and liver. They also show small-sized livers with a marked decrease in glycogen and lipid storage. The phenotypes are in striking contrast to those of recently reported animal models of lipoatrophic diabetes and patients with lipoatrophic diabetes with reduced amount of leptin. The present study provides evidence that leptin is an adipocyte-derived antidiabetic hormone in vivo and suggests its pathophysiologic and therapeutic implications in diabetes.

In order to elucidate energy balance in the skeletal muscle, we cloned cDNA of a homologue of uncoupling protein (UCP) from rat skeletal muscle. We also cloned rat UCP-2 cDNA from rat brown adipose tissue (BAT). The UCP cloned from rat skeletal muscle showed 57% and 72% identity with rat UCP-1 and UCP-2. The mRNA was expressed abundantly in the skeletal muscle, moderately in the BAT, and slightly in the white adipose tissue (WAT) with a major band at 2.5 kb and a minor band at 2.8 kb, while the UCP-2 gene expression was widely detected in the whole body with substantial levels in the WAT and with slight levels in the skeletal muscle and BAT. The rat UCP cloned in the present study showed 86% identity with the recently cloned human UCP-3, which was also expressed abundantly in the skeletal muscle with a signal of 2.4 kb. Therefore, the rat UCP was considered to be rat UCP-3. In rats fed high-fat diet the UCP-3 gene expression was augmented 2-fold in the skeletal muscle while UCP-2 mRNA levels were increased significantly (1.6-fold) in the epididymal WAT. Augmented expression of UCPs may provide defense against high-fat induced obesity and impairment of glucose metabolism.

Leptin is an adipocyte-derived blood-borne satiety factor that acts on its cognate leptin receptor (Ob-R) in the hypothalamus, thereby regulating food intake and energy expenditure. To explore whether mutations in the Ob-R gene cause obesity in humans, we have searched for mutations in the gene for Ob-Rb, a biologically active receptor isoform, in obese Japanese subjects. We have also examined associations between such mutants and obesity in the Japanese. Genomic DNAs were used as templates in polymerase chain reaction (PCR) with primers selected to amplify exons 2 to 20 of the human Ob-Rb gene. Direct sequence analysis of the PCR products revealed 7 nucleotide sequence variants (Lys109Arg, Gln223Arg, Ser343Ser, Ser492Thr, Lys656Asn, Ala976Asp, and Pro1019Pro) in the Ob-Rb coding region from 17 obese Japanese subjects with a family history of obesity (BMI 39.3 +/- 8.4 kg/m2). No missense and nonsense mutations were found such as those in Zucker fatty (fa/fa) rats and Koletsky (fa[k]/ fa[k]) rats. Nucleotide substitutions occurred at relatively high frequencies at codons 109, 223, 976, and 1019 (79, 91, 100, and 85%, respectively). Allele frequency of each variant determined by PCR-RFLP and PCR-single strand conformation polymorphism analyses showed no significant differences between 47 obese (BMI 35.1 +/- 6.5 kg/m2) and 68 non-obese (BMI 21.6 +/- 2.2 kg/m2) subjects. The present study represents the first report of sequence variants of the Ob-Rb gene in the Japanese and provides evidence against either obesity-causing mutations or association of sequence variants with obesity in obese Japanese subjects.

Insulin-stimulated glucose transport activity in rat adipocytes is inhibited by isoprenaline and enhanced by adenosine. Both of these effects occur without corresponding changes in the subcellular distribution of the GLUT4 glucose transporter isoform. In this paper, we have utilized the impermeant, exofacial bis-mannose glucose transporter-specific photolabel, 2-N-4-(1-azi-2,2,2-trifluoroethyl)benzoyl-1,3-bis-(D-mannos- 4-yloxy)-2-propylamine (ATB-BMPA) [Clark & Holman (1990) Biochem. J. 269, 615-622], to examine the cell surface accessibility of GLUT4 glucose transporters under these conditions. Compared with cells treated with insulin alone, adenosine in the presence of insulin increased the accessibility of GLUT4 to the extracellular photolabel by approximately 25%, consistent with its enhancement of insulin-stimulated glucose transport activity; the plasma membrane concentration of GLUT4 as assessed by Western blotting was unchanged. Conversely, isoprenaline, in the absence of adenosine, promoted a time-dependent (t1/2 approximately 2 min) decrease in the accessibility of insulin-stimulated cell surface GLUT4 of > 50%, which directly correlated with the observed inhibition of transport activity; the plasma membrane concentration of GLUT4 decreased by 0-15%. Photolabelling the corresponding plasma membranes revealed that these alterations in the ability of the photolabel to bind to GLUT4 are transient, as the levels of both photolabel incorporation and plasma membrane glucose transport activity were consistent with the observed GLUT4 concentration. These data suggest that insulin-stimulated GLUT4 glucose transporters can exist in two distinct states within the adipocyte plasma membrane, one which is functional and accessible to extracellular substrate, and one which is non-functional and unable to bind extracellular substrate. These effects are only observed in the intact adipocyte and are not retained in plasma membranes isolated from these cells when analysed for their ability to transport glucose or bind photolabel.