Leptin is bound in human blood by a high affinity binding protein, which appears to be identical with the soluble leptin receptor (sOB-R). Using a ligand-mediated immunofunctional assay for the determination of serum sOB-R, we investigated its course during childhood, puberty, and adolescence in a large cohort of 581 healthy children and adolescents and a small group of 13 patients with anorexia nervosa. In the first years of life, sOB-R is detectable in remarkably high concentrations. Thereafter, a continuous decline of sOB-R levels was found. Consequently, correlation analyses demonstrated significant inverse relationships (P < 0.001) of sOB-R with age, IGF-I levels, pubertal stage, auxological and body composition parameters, as well as with leptin concentrations. Multiple regression analysis revealed that height, IGF-I, and age (only in girls) were independent predictors of sOB-R levels; these variables account for approximately 65% and 48% of the variation of sOB-R levels in boys and girls, respectively. The courses of age-dependent median values for the free leptin index (FLI, ratio between leptin and sOB-R levels) and for leptin levels were parallel in both genders. Correlation analyses demonstrated that in particular parameters of growth and sexual maturation are more closely related to the FLI than to leptin alone; this closer relationship is more pronounced among boys. Weight gains of patients with anorexia nervosa resulted in a significant increase in leptin and IGF-I levels (P < 0.01), whereas the median of sOB-R values decreased (P < 0.01). sOB-R and IGF-I levels were again significantly correlated (r = -0.55, P < 0.01). These findings suggest that high levels of sOB-R in emaciation may reflect an up-regulation of the sOB-R to suppress leptin action during energy deficiency. Furthermore, determinations of sOB-R and FLI are additional valuable tools to investigate the leptin axis during growth and sexual maturation.

Leptin acts not only as an anorexigenic hormone but also regulates cell-mediated immunity via leptin receptors (Ob-R) expressed on T and B lymphocytes. However, the impact of leptin on natural killer (NK) cells is currently elusive. We evaluated leptin effects on NK cells in relation to the body weight in rats using in vivo and in vitro approaches. Leptin was injected iv in male lean and diet-induced obese Lewis and F344 rats. NK cell numbers were analyzed in blood and spleen by fluorescence activated cell sorting and immunohistochemistry, and the activity of NK cells was measured by chromium release assay. Ob-R expression was investigated by confocal laser scanning and quantitative RT-PCR. To compare leptin-dependent intracellular signaling under basal and leptin- and tumor cell (MADB106)-stimulated conditions, intracellular target proteins of NK cells were evaluated by Western blotting. Number and distribution pattern of splenic NK cells were significantly different in lean and obese animals. Leptin administration resulted in a 4-fold higher stimulation of the NK activity in lean than obese animals. This was not due to a decreased expression of Ob-R because quantitative RT-PCR revealed significantly higher Ob-Rb mRNA levels in NK cells from obese rats. In contrast, postreceptor signaling is differentially abrogated in obese animals with significantly lower activation of postreceptor signaling components (Janus kinase-2p, protein kinase B pT308, AMPalphapT172) after an in vivo leptin challenge. In conclusion, the results for the first time assign leptin a central role as a modulator of NK cell number and activity only in lean but not obese subjects. The differential role of leptin has important implications for the influence of body weight in the response to systemic inflammations and in the immunological defense of cancer.

A B S T R A C T To investigate the role of glucagon in regulating hepatic glucose production in man, selective glucagon deficiency was produced in four normal men by infusing somatostatin (0.9 mg/h) and regular pork insulin (150-,uU/kg per min) for 2 h. Exogenous glucose was infused to maintain euglycemia. Arterial plasma glucagon levels fell by greater than 50%o whereas plasma insulin levels were maintained in the range of 10-14 .uU/ml. In response to these hormonal changes, net splanchnic glucose production (NSGP) fell by 75% and remained suppressed for the duration of the study.In contrast, when somatostatin alone was administered to normal men, resulting in combined insulin and glucagon deficiency (euglycemia again maintained), NSGP fell markedly but only transiently, reaching its nadir at 15 min. Thereafter, NSGP rose progressively, reaching the basal rate at 105 min.These data indicate that the induction of selective glucagon deficiency in man (with basal insulin levels maintained) is associated with a marked and sustained fall in hepatic glucose production.

We examined the effect of hyperglycemia per se on net splanchnic glucose balance. In 2 groups of normal postabsorptive men who had undergone hepatic vein catheterization, somatostatin was administered to block endogenous insulin and glucagon secretion. Exogenous glucose was infused in both groups to maintain euglycemia for 2 h in one group (n = 7) and to induce hyperglycemia of 220-240 mg/dl after 30 minutes of euglycemia in the second group (n = 4). In both groups the induction of insulinopenia and glucagonopenia with euglycemia maintained resulted in an initial 75% fall in net splanchnic glucose production (NSGP). In the group in which euglycemia was maintained NSGP returned to basal rates (157 +/- 31 mg/min) within 2 h. However, in the group in which hyperglycemia was induced, NSGP did not return to basal rates but remained suppressed (28 +/- 4 mg/min) for the duration of the study. These data in normal man indicate that hyperglycemia per se with insulin and glucagon acutely withdrawn can suppress splanchnic glucose production but does not induce net splanchnic glucose storage.

Increased supply of fatty acids to muscle and liver is causally involved in the insulin resistance syndrome. Using a tissue microdialysis technique in Wistar and Zucker fatty (ZF) rats, we determined tissue glycerol levels as a marker of lipolysis in gastrocnemius muscle (gMT), subcutaneous adipose (SAT), and visceral adipose tissue (VAT) as well as the reduction of plasma free fatty acids, glycerol, and triglycerides caused by the antilipolysis-specific adenosine-A1 receptor agonist (ARA). In Wistar and ZF rats, ARA significantly lowered dialysate glycerol levels in SAT, VAT, and gMT. Whereas in SAT and VAT the decrease in dialysate glycerol indicated adipocytic antilipolysis, this decrease in gMT was not caused by a direct effect of ARA on intramyocellular lipolysis, as demonstrated by the lack of inhibition of the protein kinase A activity ratio in gMT. In addition, no differences of the fed-starved-refed dynamics of intramyocellular triglyceride levels compared with untreated controls were measured by in vivo 1 Hspectroscopy, excluding any adenylate cyclase-independent antilipolysis in muscle. Treatment with ARA resulted in pronounced reductions of plasma free fatty acids, glycerol, and triglycerides. Furthermore, in ZF rats, ARA treatment caused an immediate improvement of peripheral insulin sensitivity measured by the euglycemic-hyperinsulinemic glucose clamp technique.