Serum leptin concentrations are correlated with the percentage of body fat, suggesting that most obese persons are insensitive to endogenous leptin production.

No abstract

We investigated the response of leptin to short-term fasting and refeeding in humans. A mild decline in subcutaneous adipocyte ob gene mRNA and a marked fall in serum leptin were observed after 36 and 60 h of fasting. The dynamics of the leptin decline and rise were further substantiated in a 6-day study consisting of a 36-h baseline period, followed by 36-h fast, and a subsequent refeeding with normal diet. Leptin began a steady decline from the baseline values after 12 h of fasting, reaching a nadir at 36 h. The subsequent restoration of normal food intake was associated with a prompt leptin rise and a return to baseline values 24 h later. When responses of leptin to fasting and refeeding were compared with that of glucose, insulin, fatty acids, and ketones, a reverse relationship between leptin and beta-OH-butyrate was found. Consequently, we tested whether the reciprocal responses represented a causal relationship between leptin and beta-OH-butyrate. Small amounts of infused glucose equal to the estimated contribution of gluconeogenesis, which was sufficient to prevent rise in ketogenesis, also prevented a fall in leptin. The infusion of beta-OH-butyrate to produce hyperketonemia of the same magnitude as after a 36-h fast had no effect on leptin. The study indicates that one of the adaptive physiological responses to fasting is a fall in serum leptin. Although the mediator that brings about this effect remains unknown, it appears to be neither insulin nor ketones.

This study was undertaken to investigate the changes in obesity (OB) gene expression and production of leptin in response to insulin in vitro and in vivo under euglycemic and hyperglycemic conditions in humans. Three protocols were used: 1) euglycemic clamp with insulin infusion rates at 40, 120, 300, and 1,200 mU / m / min carried out for up to 5 h performed in 16 normal lean individuals, 30 obese individuals, and 31 patients with NIDDM; 2) 64-to 72-h hyperglycemic (glucose 12.6 mmol/l) clamp performed on 5 lean individuals; 3) long-term (96-h) primary culture of isolated abdominal adipocytes in the presence and absence of 100 nmol/l insulin. Short-term hyperinsulinemia in the range of 80 to > 10,000 microU/ml had no effect on circulating levels of leptin. During the prolonged hyperglycemic clamp, a rise in leptin was observed during the last 24 h of the study (P < 0.001). In the presence of insulin in vitro, OB gene expression increased at 72 h (P < 0.01), followed by an increase in leptin released to the medium (P < 0.001). In summary, insulin does not stimulate leptin production acutely; however, a long-term effect of insulin on leptin production could be demonstrated both in vivo and in vitro. These data suggest that insulin regulates OB gene expression and leptin production indirectly, probably through its trophic effect on adipocytes.

Obese (ob) gene expression in abdominal subcutaneous adipocytes from lean and obese humans was examined. The full coding region of the ob gene was isolated from a human adipocyte cDNA library. Translation of the insert confirmed the reported amino acid sequence. There was no difference in the sequence of an reverse transcription PCR product of the coding region from five lean and five obese subjects. The nonsense mutation in the ob mouse which results in the conversion of arginine 105 to a stop codon was not present in human obesity. In all 10 human cDNAs, arginine 105 was encoded by CGG, consequently two nucleotide substitutions would be required to result in a stop codon. To compare the amount of ob gene expression in lean and obese individuals, radiolabed primer was used in the PCR reaction with (Iactin as a control. There was 72% more ob gene expression (P < 0.01) in eight obese subjects (body mass index, BMI = 42.8±2.7) compared to eight lean controls (BMI = 22.4±0.8). Regression analysis indicated a positive correlation between BMI and the amount of ob message (P < 0.005). There was no difference in the amount of ,B-actin expression in the two groups. These results provide evidence that ob gene expression is increased in human obesity; furthermore, the mutations present in the mouse ob gene were not detected in the human mRNA population. (J. Clin. Invest. 1995. 95:2986-2988

TL1A is a TNF-like cytokine that binds to the death-domain receptor (DR)3 and provides costimulatory signals to activated lymphocytes. Through this interaction, TL1A induces secretion of IFN-␥ and may, therefore, participate in the development of T helper-1-type effector responses. In this study, we investigated whether interactions between TL1A and DR3 are involved in the pathogenesis of chronic murine ileitis. We demonstrate that alternative splicing of DR3 mRNA takes place during the activation of lymphocytes, which results in up-regulation of the complete͞transmembrane (tm) form of DR3. Using two immunogenetically distinct animal models of Crohn's disease, we demonstrate that induction of intestinal inflammation is associated with significant up-regulation of TL1A and tm DR3 in the inflamed mucosa. In addition, within isolated lamina propria mononuclear cells from mice with inflammation, TL1A is primarily expressed on CD11c high dendritic cells. We also report that TL1A acts preferentially on memory CD4 ؉ ͞CD45RB lo murine lymphocytes by significantly inducing their proliferation, whereas it does not affect the proliferation of the naïve CD4 ؉ ͞ CD45RB hi T helper cell subpopulation. Finally, we demonstrate that TL1A synergizes with both the cytokine-dependent IL-12͞IL-18 pathway and with low-dose stimulation of the T cell receptor to significantly induce the secretion of IFN-␥ via an IL-18-independent pathway. Our results raise the possibility that interaction(s) between TL1A expressed on antigen-presenting cells and tm DR3 on lymphocytes may be of particular importance for the pathogenesis of chronic inflammatory conditions that depend on IFN-␥ secretion, including inflammatory bowel disease. Blockade of the TL1A͞DR3 pathway may, therefore, offer therapeutic opportunities in Crohn's disease.Crohn's disease ͉ cytokines ͉ mucosal inflammation T he differentiation of naïve CD4 ϩ lymphocytes into IFN-␥-secreting Th1 ''effector'' cells is a multistep process that involves several cell types, costimulatory molecules, transcription factors, and secreted cytokines (1). Antigen-presenting cell (APC)-derived IL-12 is essential for the induction of IFN-␥, an effect that is greatly enhanced by IL-18 (2). IL-12 up-regulates T-bet, a transcription factor that is critical for the stabilization of a T helper (Th)1-polarized phenotype (3). Recently, additional cytokines that play prominent roles during Th1 responses have been described, such as IL-27 and IL-23 (4). Engagement of the T cell receptor (TCR) provides further signals for the induction of IFN-␥, both in parallel to and independently of cytokinemediated pathways (1).Members of the TNF and TNF-receptor superfamilies of proteins (TNFSFPs and TNFRSFPs, respectively) are abundantly expressed in the immune system, and are critically involved in the differentiation, proliferation, and apoptosis of immune cells (5). Several members of these families induce secretion of IFN-␥ upon ligand͞receptor binding, thereby enhancing Th1-type responses (6-8). TL1A (TNFSPF15) is a r...

No abstract

We tested the hypothesis that liver protein kinase C (PKC) is increased in non-insulin-dependent diabetes mellitus (NIDDM). To this end we examined the distribution of PKC isozymes in liver biopsies from obese individuals with and without NIDDM and in lean controls. PKC isozymes a, 13, e and g were detected by immunoblotting in both the cytosol and membrane fractions. Isozymes y and 6 were not detected. There was a significant increase in immunodetectable PKC-a (twofold), -E (threefold), and -C (twofold) in the membrane fraction isolated from obese subjects with NIDDM compared with the lean controls. In obese subjects without NIDDM, the amount of membrane PKC isozymes was not different from the other two groups. We next sought an animal model where this observation could be studied further. The Zucker diabetic fatty rat offered such a model system. Immunodetectable membrane PKC-ci, -13, -e, and -t were significantly increased when compared with both the lean and obese controls. The increase in immunodetectable PKC protein correlated with a 40% elevation in the activity of PKC at the membrane. Normalization of circulating glucose in the rat model by either insulin or phlorizin treatment did not result in a reduction in membrane PKC isozyme protein or kinase activity. Further, phlorizin treatment did not improve insulin receptor autophosphorylation nor did the treatment lower liver diacylglycerol. We conclude that liver PKC is increased in NIDDM, a change that is not secondary to hyperglycemia. It is possible that PKCmediated phosphorylation of some component in the insulin signaling cascade contributes to the insulin resistance observed in NIDDM. (J. Clin. Invest. 1995Invest. .95:2938Invest. -2944