Katsunori Nonogaki

Up-Regulation of Peroxisome Proliferator-Activated Receptors (PPAR-α) and PPAR-γ Messenger Ribonucleic Acid Expression in the Liver in Murine Obesity: Troglitazone Induces Expression of PPAR-γ-Responsive Adipose Tissue-Specific Genes in the Liver of Obese Diabetic Mice1

Peroxisome proliferator-activated receptors (PPARs) are transcription factors that play an important role in the regulation of genes involved in lipid utilization and storage, lipoprotein metabolism, adipocyte differentiation, and insulin action. The three isoforms of the PPAR family, i.e. alpha, delta, and gamma, have distinct tissue distribution patterns. PPAR-alpha is predominantly present in the liver, and PPAR-gamma in adipose tissue, whereas PPAR-delta is ubiquitously expressed. A recent study reported increased PPAR-gamma messenger RNA (mRNA) expression in the liver in ob/ob mice; however, it is not known whether increased PPAR-gamma expression in the liver has any functional consequences. The expression of PPAR-alpha and -delta in the liver in obesity has not been determined. We have now examined the mRNA levels of PPAR-alpha, -delta, and -gamma in three murine models of obesity, namely, ob/ob (leptin-deficient), db/db (leptin-receptor deficient), and serotonin 5-HT2c receptor (5-HT2cR) mutant mice. 5-HT2cR mutant mice develop a late-onset obesity that is associated with higher plasma leptin levels. Our results show that PPAR-alpha mRNA levels in the liver are increased by 2- to 3-fold in all three obese models, whereas hepatic PPAR-gamma mRNA levels are increased by 7- to 9-fold in ob/ob and db/db mice and by 2-fold in obese 5-HT2cR mutant mice. PPAR-delta mRNA expression is not altered in ob/ob or db/db mice. To determine whether increased PPAR-gamma expression in the liver has any functional consequences, we examined the effect of troglitazone treatment on the hepatic mRNA levels of several PPAR-gamma-responsive adipose tissue-specific genes that have either no detectable or very low basal expression in the liver. The treatment of lean control mice with troglitazone significantly increased the expression of adipocyte fatty acid-binding protein (aP2) and fatty acid translocase (FAT/CD36) in the liver. This troglitazone-induced increase in the expression of aP2 and FAT/CD36 was markedly enhanced in the liver in ob/ob mice. Troglitazone also induced a pronounced increase in the expression of uncoupling protein-2 in the liver in ob/ob mice. In contrast to the liver, troglitazone did not increase the expression of aP2, FAT/CD36, and uncoupling protein-2 in adipose tissue in lean or ob/ob mice. Taken together, our results suggest that the effects of PPAR-gamma activators on lipid metabolism and energy homeostasis in obesity and type 2 diabetes may be partly mediated through their effects on PPAR-gamma in the liver.

Serotonin systems upregulate the expression of hypothalamic NUCB2 via 5-HT2C receptors and induce anorexia via a leptin-independent pathway in mice.

NEFA/nucleobindin2 (NUCB2), a novel satiety molecule, is associated with leptin-independent melanocortin signaling in the central nervous system. Here, we show that systemic administration of m-chlorophenylpiperazine (mCPP), a serotonin 5-HT1B/2C receptor agonist, significantly increased the expression of hypothalamic NUCB2 in wild-type mice. The increases in hypothalamic NUCB2 expression induced by mCPP were attenuated in 5-HT2C receptor mutant mice. Systemic administration of mCPP suppressed food intake in db/db mice with leptin receptor mutation as well as lean control mice. On the other hand, the expression of hypothalamic NUCB2 and proopiomelanocortin (POMC) was significantly decreased in hyperphagic and non-obese 5-HT2C receptor mutants compared with age-matched wild-type mice. Interestingly, despite increased expression of hypothalamic POMC, hypothalamic NUCB2 expression was decreased in 5-HT2C receptor mutant mice with heterozygous mutation of beta-endorphin gene. These findings suggest that 5-HT systems upregulate the expression of hypothalamic NUCB2 via 5-HT2C receptors, and induce anorexia via a leptin-independent pathway in mice.

Ghrelin and feedback systems.

Ghrelin is produced primarily in the stomach in response to hunger, and circulates in the blood. Plasma ghrelin levels increase during fasting and decrease after ingesting glucose and lipid, but not protein. The efferent vagus nerve contributes to the fasting-induced increase in ghrelin secretion. Ghrelin secreted by the stomach stimulates the afferent vagus nerve and promotes food intake. Ghrelin also stimulates pituitary gland secretion of growth hormone (GH) via the afferent vagus nerve. GH inhibits stomach ghrelin secretion. These findings indicate that the vagal circuit between the central nervous system and stomach has a crucial role in regulating plasma ghrelin levels. Moreover, body mass index modulates plasma ghrelin levels. In a lean state and anorexia nervosa, plasma ghrelin levels are increased, whereas in obesity, except in Prader-Willi syndrome, plasma ghrelin levels are decreased and the feeding- and sleeping-induced decline in plasma ghrelin levels is disrupted. There are two forms of ghrelin: active n-octanoyl-modified ghrelin and des-acyl ghrelin. Fasting increases both ghrelin types compared with the fed state. Hyperphagia and obesity are likely to decrease plasma des-acyl ghrelin, but not n-octanoyl-modified ghrelin levels. Hypothalamic serum and glucocorticoid-inducible kinase-1 and serotonin 5-HT2C/1B receptor gene expression levels are likely to be proportional to plasma des-acyl ghrelin levels during fasting, whereas they are likely to be inversely proportional to plasma des-acyl ghrelin levels in an increased energy storage state such as obesity. Thus, a dysfunction of the ghrelin feedback systems might contribute to the pathophysiology of obesity and eating disorders.

Liraglutide suppresses obesity and hyperglycemia associated with increases in hepatic fibroblast growth factor 21 production in KKAy mice.

Social isolation contributes to the development of obesity and insulin-independent diabetes in KKAy mice. Here we show that systemic administration of liraglutide, a long-acting human glucagon-like peptide-1 (GLP-1) analog, significantly decreased food intake, body weight, and blood glucose levels at 24 h after its administration while having no significant effects on plasma insulin and glucagon levels in individually housed KKAy mice. In addition, the systemic administration of liraglutide significantly increased plasma fibroblast growth factor (Fgf) 21 levels (1.8-fold increase) associated with increases in the expression of hepatic Fgf21 (1.9-fold increase) and Pparγ (1.8-fold increase), while having no effects on the expression of hepatic Pparα and Fgf21 in white adipose tissue. Moreover, systemic administration of liraglutide over 3 days significantly suppressed food intake, body weight gain, and hyperglycemia in KKAy mice. On the other hand, despite remarkably increased plasma active GLP-1 levels (4.2-fold increase), the ingestion of alogliptin, a selective dipeptidyl peptidase-4 inhibitor, over 3 days had no effects on food intake, body weight, blood glucose levels, and plasma Fgf21 levels in KKAy mice. These findings suggest that systemic administration of liraglutide induces hepatic Fgf21 production and suppresses the social isolation-induced obesity and diabetes independently of insulin, glucagon, and active GLP-1 in KKAy mice.

Serotonin 5-HT2C receptor-independent expression of hypothalamic NOR1, a novel modulator of food intake and energy balance, in mice

NOR1, Nur77 and Nurr1 are orphan nuclear receptors and members of the NR4A subfamily. Here, we report that the expression of hypothalamic NOR1 was remarkably decreased in mildly obese beta-endorphin-deficient mice and obese db/db mice with the leptin receptor mutation, compared with age-matched wild-type mice, whereas there were no genotypic differences in the expression of hypothalamic Nur77 or Nurr1 in these animals. The injection of NOR1 siRNA oligonucleotide into the third cerebral ventricle significantly suppressed food intake and body weight in mice. On the other hand, the decreases in hypothalamic NOR1 expression were not found in non-obese 5-HT2C receptor-deficient mice. Moreover, systemic administration of m-chlorophenylpiperazine (mCPP), a 5-HT2C/1B receptor agonist, had no effect on hypothalamic NOR1 expression, while suppressing food intake in beta-endorphin-deficient mice. These findings suggest that 5-HT2C receptor-independent proopiomelanocortin-derived peptides regulate the expression of hypothalamic NOR1, which is a novel modulator of feeding behavior and energy balance.

Fluvoxamine exerts anorexic effect in 5-HT2C receptor mutant mice with heterozygous mutation of β-endorphin gene

Serotonin (5-hydroxytryptamine; 5-HT) 2C receptors and the downstream melanocortin pathway are suggested to mediate the anorexic effects of m-chlorophenylpiperazine (mCPP) and fenfluramine. We previously reported that fluvoxamine, a selective serotonin reuptake inhibitor, together with pharmacological inactivation of 5-HT2C receptors exert feeding suppression through activation of 5-HT1B receptors in mice. Here, we report that fluvoxamine exerted anorexic effects in 5-HT2C receptor mutant mice with heterozygous mutation of beta-endorphin gene (2CREnd mice), whereas fluvoxamine had no effect on food intake in age-matched wild-type mice and 5-HT2C receptor mutant mice, which are associated with decreases in hypothalamic proopiomelanocortin (POMC) expression. mCPP suppressed food intake in 5-HT2C receptor mutant mice, 2CREnd mice and age-matched wild-type mice. These results suggest that fluvoxamine-induced feeding suppression requires a perturbation of 5-HT2C receptor and beta-endorphin signalling plus functional hypothalamic POMC activity, whereas mCPP-induced feeding suppression does not always require functional 5-HT2C receptor, beta-endorphin, and POMC activity in mice.

The contribution of serotonin 5-HT2C and melanocortin-4 receptors to the satiety signaling of glucagon-like peptide 1 and liragultide, a glucagon-like peptide 1 receptor agonist, in mice

Glucagon-like peptide 1 (GLP-1), an insulinotropic gastrointestinal peptide produced mainly from intestinal endocrine L-cells, and liraglutide, a GLP-1 receptor (GLP-1R) agonist, induce satiety. The serotonin 5-HT2C receptor (5-HT2CR) and melanoroctin-4 receptor (MC4R) are involved in the regulation of food intake. Here we show that systemic administration of GLP-1 (50 and 200μg/kg)-induced anorexia was blunted in mice with a 5HT2CR null mutation, and was attenuated in mice with a heterozygous MC4R mutation. On the other hand, systemic administration of liraglutide (50 and 100μg/kg) suppressed food intake in mice lacking 5-HT2CR, mice with a heterozygous mutation of MC4R and wild-type mice matched for age. Moreover, once-daily consecutive intraperitoneal administration of liraglutide (100μg/kg) over 3days significantly suppressed daily food intake and body weight in mice with a heterozygous mutation of MC4R as well as wild-type mice. These findings suggest that GLP-1 and liraglutide induce anorexia via different central pathways.

Hypothalamic orexin and pro-opiomelanocortin activities are essential for the anorexic effects of m-chlorophenylpiperazine in mice

Hypothalamic pro-opiomelanocortin (POMC) activity is reportedly essential for satiety signalling downstream of serotonin (5-HT). Here we show that food-restricted wild-type mice, which exhibited decreased hypothalamic POMC expression and increased hypothalamic orexin expression, were responsive to m-chlorophenylpiperazine (m-CPP), a 5-HT(2C/1B) receptor agonist, leading to anorexia, whereas food-restricted A(y) mice with decreased hypothalamic POMC and orexin expression, were not. Injection of POMC small interfering RNA (siRNA) oligonucleotide+orexin siRNA oligonucleotide into the third cerebral ventricle was unresponsive to mCPP-induced anorexia, whereas a single injection of POMC or orexin siRNA oligonucleotides elicited a response. The injection of POMC siRNA oligonucleotides suppressed the anorexic effects of sibutramine, a serotonin and noradrenaline re-uptake inhibitor. The injection of orexin siRNA oligonucleotides suppressed the hyperphagia induced by the injection of POMC siRNA oligonucleotides. These findings suggest that functional hypothalamic POMC and orexin activity has a critical role in satiety signalling of mCPP in mice.

Novel modulators for body weight changes induced by fasting and re-feeding in mice

Catch-up weight gain after malnutrition is a risk factor for metabolic syndrome. Here we show that social isolation enhanced fasting-induced weight loss and suppressed weight gain induced by re-feeding for 6 days following a 24-h fast in prepubertal wild-type mice. These effects of social isolation on weight gain were not associated with significant changes in daily average food consumption. Under the same housing condition, genetic deletion of beta-endorphin reduced the fasting-induced weight loss and enhanced the re-feeding-induced weight gain in prepubertal mice. These effects of social isolation or genetic deletion of beta-endorphin on these weight changes were attenuated and reversed in postpubertal mice. Moreover, genetic deletion of beta-endorphin attenuated these effects of social isolation on the catch-up weight gain in prepubertal mice and reversed them in postpubertal mice. Thus, social isolation, endogenous beta-endorphin, and age can be novel modulators for body weight changes induced by fasting and re-feeding in mice.

Serotonin conflict in sleep-feeding.

Short sleep duration has been suggested to be a risk factor for weight gain and adiposity. Serotonin (5-HT) substantially contributes to the regulation of sleep and feeding behavior. Although 5-HT predominately promotes waking and satiety, the effects of 5-HT depend on 5-HT receptor function. The 5-HT1A, 5-HT1B, 5-HT2A, 5-HT2C, 5-HT6, and 5-HT7 receptors reportedly contribute to sleep-waking regulation, whereas the 5-HT1B and 5-HT2C receptors contribute to the regulation of satiety. The 5-HT1B and 2C receptors may therefore be involved in the regulation of sleep-feeding. In genetic studies, 5-HT1B receptor mutant mice display greater amounts of rapid eye movement sleep (REMS) than wild-type mice, while displaying no effects on waking or slow wave sleep (SWS). On the other hand, 5-HT2C receptor mutant mice exhibit increased wakefulness and decreased SWS, without any effect on REMS. Moreover, the 5-HT2C receptor mutants display leptin-independent hyperphagia, leading to a middle-aged onset of obesity, whereas 5-HT1B receptor mutants do not display any effect on food intake. Thus, the genetic deletion of 5-HT2C receptors results in sleep loss-associated hyperphagia, leading to the late onset of obesity. This is a quite different pattern of sleep-feeding behavior than is observed in disturbed leptin signaling, which displays an increase in sleep-associated hyperphagia. In pharmacologic studies, 5-HT1B and 5-HT2C receptors upregulate wakefulness and downregulate SWS, REMS, and food intake. These findings suggest that 5-HT1B/2C receptor stimulation induces sleep loss-associated anorexia. Thus, the central 5-HT regulation of sleep-feeding can be dissociated. Functional hypothalamic proopiomelanocortin and orexin activities may contribute to the dissociated 5-HT regulation.