The discovery of leptin in 1994 has led to remarkable advances in obesity research. We now know that leptin is a cytokinelike hormone that is produced in adipose tissue and plays a pivotal role in regulation of energy balance and in a variety of additional processes via actions in the central nervous system. This symposium review covers current understandings of neuronal leptin receptor signaling and mechanisms of obesity-related leptin resistance in the central nervous system and provides recent insights into the regulation of peripheral glucose balance by central leptin action in rodents.
Statistics from Altmetric.com
Leptin, an adipocyte-derived hormone, acts principally on the central nervous system to activate its cognate receptor. The absence of leptin or of its receptor in Lepob/ob or Leprdb/db mice, respectively, results in morbid obesity, hyperphagia, neuroendocrine dysfunction, and severe hyperglycemia and insulin resistance.1-6
Leptin receptors are expressed in a number of specific brain regions,7,8and binding of leptin leads to regulation of a range of biological functions and processes, including energy intake and expenditure, body fat, neuroendocrine systems, autonomic function, and insulin-and-glucose balance.9-12Although the specific function of each brain nucleus in leptin action is yet largely unknown, data suggest that distinct biological actions of leptin are mediated by different brain nuclei but that overlapping or redundant functional sites also exist. The arcuate nucleus of the hypothalamus (ARC) is a key area for mediating leptin actions on energy homeostasis. Consistent with this, leptin receptor messenger RNA is densely expressed in the ARC of mice and rats,13,14and injection of leptin directly into the ARC is sufficient to acutely reduce food intake.15Moreover, restoration of leptin receptor expression in the ARC of leptin receptor-deficient Leprdb/db mice leads to long-term reduction of body weight and food intake,16and arcuate nucleus-specific Lepr gene therapy is sufficient to attenuate the obesity phenotype of leptin receptor-deficient Koletsky fak/fak rats.17
The ARC contains at least 2 subsets of leptin-responsive neurons, namely, the anorexigenic proopiomelanocortin (POMC) neurons and the orexigenic Agouti-related peptide (AgRP) neurons. Proopiomelanocortin neurons are depolarized by leptin, leading to release of α-melanocyte-stimulating hormone (α-MSH), a POMC-derived neuropeptide that mediates its anorexigenic effects through activation of melanocortin receptors.12,18,19Agouti-related peptide is a melanocortin receptor antagonist that potently stimulates feeding.20Consistent with this, AgRP neurons are inhibited by leptin, resulting in a reduction in AGRP neuropeptide release.21Mice lacking leptin receptors only in POMC or AgRP neurons are mildly obese, demonstrating that both groups of cells are required for maintenance of body weight homeostasis by leptin.22,23Furthermore, reexpression of leptin receptors exclusively in POMC neurons of the receptor-deficient Leprdb/db mice modestly reduces body weight and energy intake.24The modest body weight changes observed in these studies compared with the morbid obesity in Leprdb/db mice that lack all functional leptin receptors demonstrates that neurons apart from POMC and AgRP neurons are also important for body weight regulation by leptin.
In addition to the ARC, neurons that express leptin receptors are found in a number of other hypothalamic and extrahypothalamic brain nuclei,7,8and progress has been made in understanding their roles in leptin action. For example, leptin signaling of the ventromedial hypothalamic nucleus can, like the arcuate nucleus, mediate short-term energy intake suppression15and long-term weight loss.25Furthermore, the ventromedial hypothalamic nucleus may serve a role in regulation of the autonomic nervous system by leptin.26In addition, leptin signaling in thyrotropin-releasing hormone neurons within the paraventricular hypothalamic nucleus of rats may account, at least in part, for leptin's effects on the thyroid axis.27,28Furthermore, the ventral premammillary nucleus is likely critical for leptin's action on the neuroendocrine gonadal axis and reproduction.29Leptin receptors are also expressed in the ventral tegmental area (VTA) of the mid brain. Specifically, leptin directly targets dopamine neurons of the VTA, suggesting that leptin can affect critical brain reward circuitries.30Indeed, injection of leptin directly into the VTA reduces food intake and stimulates locomotor activity.31The nucleus tractus solitarius (NTS) located in the caudal hindbrain is a major projection zone for sensory nerve input from the gastrointestinal system and contains leptin-regulated neurons.32Interestingly, these latter neurons are also responsive to gastric distention in rats,24and intraparenchymal NTS administration of leptin acutely reduces food intake and body weight.33These latter data combined suggest that the effects of leptin on food intake in the hindbrain may result from modulation of gastrointestinal signal processing.34The effect of leptin on food intake thus seems to be mediated by leptin receptors in several nuclei within the hypothalamus, in part via reward neurons located in the mid brain and in part by neurons in the NTS of the caudal brainstem. Future studies are needed to determine how different neurons mediate the same behavioral effects by leptin or whether those neurons in fact serve specific and separate functions under different circumstances.
Leptin is structurally similar to cytokines consistent with its receptor belonging to the cytokine receptor class 1 superfamily.35Several isoforms of ObR exists, including a long signaling form (ObRb).36The murine ObRb receptor contains 3 conserved intracellular tyrosine residues, located at amino acid positions7985, 1077, and 1138. Tyrosine phosphorylation sites provide binding motifs for src homology 2 domain-containing proteins, such as signal transducers and activators of transcription (STATs). Leptin binding to its receptor activates Janus tyrosine kinase (JAK) enzymes that are constitutively associated with membrane-proximal regions of the receptor. Janus tyrosine kinase mediates leptin-dependent tyrosine phosphorylation of the leptin receptor itself.37Phosphorylated Tyr1138 recruits the latent cytoplasmic transcription factor STAT3, facilitating its phosphorylation by JAK followed by STAT3 dimerization and nuclear translocation.38Signal transducer and activator of transcription 3 plays a role in regulation of POMC and AgRP gene expressions by leptin.39,40In addition, we reported that leptin signaling via the STAT3 pathway rapidly induces hypothalamic expression of suppressor-of-cytokine-signaling-3 (SOCS-3), a potent inhibitor of leptin receptor signaling,41by binding to Tyr985 and inhibits JAK2 kinase activity, thereby acting in a negative feedback loop.42,43Another key negative leptin receptor regulator is protein-tyrosine-phosphotase-1B, which acts by directly inhibiting JAK2 kinase activity.44-46Tyr1077 has been reported to play a role in binding and activation of STAT5,47,48but the downstream signaling events of this pathway are yet unknown. Phosphorylated Tyr985 of ObRb also binds SHP2, a protein that participates in activation of extracellular signal-regulated kinase (mitogen-activated protein kinase) signaling and is important for c-fos transcriptional activation,43,49and likely other events. The specific intracellular mechanisms whereby ObRb regulates intracellular signaling via other effector proteins such as insulin-receptor-substrate 2,50-52phosphoinositol-3-kinase (PI3K),50,53-55mammalian target of rapamycin,56,57Fox01,40,58and adenosine monophosphate-activated protein kinase,59-62are currently less well understood. Further genetic and immunohistochemical studies are also required to determine whether these proteins are regulated only in first-order leptin-responsive neurons or in downstream neurocircuitries, or in both.
Genetic studies in mice demonstrate that signaling through Tyr1138 of the leptin receptor is required for normal regulation of energy balance. Specifically, mice with Tyr1138 mutated into a serine residue exhibit severe hyperphagia and obesity similar to that of Leprdb/db mice.63However, in contrast to Leprdb/db mice, Ser1138 mice are fertile, longer, and less hyperglycemic, altogether indicating that STAT3 signaling is critical for leptin's regulation of energy intake and whole body energy balance but that signals other than STAT3 are likely important for other leptin actions such as reproduction. In more recent studies also using homologous recombination in mice leading to mutation of Tyr985 into a leucine residue, it was reported that female, but not male, animals had modestly reduced body weight and energy intake and were protected from high-fat diet-induced obesity.64The mice also showed increased leptin sensitivity and preservation of reproductive function. Thus, these data suggest that Tyr985 of the leptin receptor may serve to convey inhibition of leptin signaling thereby attenuating the antiadiposity effects of leptin, especially in females. These data are consistent with those from in vitro signaling studies that strongly point to an inhibitory role of SOCS-3 acting via Tyr985. However, because Tyr985 recruits both SHP2 and SOCS3, either of these proteins could theoretically underlie the lean, leptin-sensitive phenotype. Alternatively, it may be speculated that because the lean phenotype is quiet modest and is only present in 1 sex, SOCS-3 and SHP2 may have opposite functions with regard to regulation of whole body energy balance by leptin. Further studies are needed to investigate this possibility and to determine the specific cellular functions of individual receptor tyrosine residues, including those of Tyr1077, within each anatomically and chemically separate population of leptin-responsive neurons.
In addition to these global manipulations of leptin receptor signaling, several intracellular proteins downstream of the leptin receptor have also been investigated with regard to their role in leptin action within specific neurons. For example, the STAT3 transcriptional factor has specifically been deleted from POMC and AgRP neurons in mice. In mice lacking STAT3 in POMC neurons, females exhibited reduced pomc gene expression and a modest increase in fat mass and total body weight.65The animals remained responsive to leptin-induced hypophagia and were not hypersensitive to development of increased weight given a high-fat diet. However, mutant mice failed to mount a normal compensatory refeeding response. These results suggest a role for STAT3 in transcriptional regulation of the pomc gene, consistent with those of previous in vitro studies of the pomc promoter,39and indicate that STAT3 expression in POMC neurons plays only a modest role in leptin's antiobesity actions. Removal of STAT3 or expression of a constitutive active form of STAT3 in AgRP neurons also demonstrates a requirement of STAT3 in those cells for normal energy balance.47,66For example, deletion of STAT3 from AgRP neurons leads to a slight weight gain of the mice.47These AgRP STAT3-deficient mice were also hyperleptinemic and exhibited high-fat diet-induced hyperinsulinemia. Agouti-related peptide messenger RNA levels were unaffected. Behaviorally, mice without STAT3 in AgRP neurons were mildly hyperphagic and hyporesponsive to leptin's intake inhibitory actions. Combined, STAT3 in AgRP and POMC neurons is therefore required for normal energy homeostasis, but it is suggested that STAT3 signaling in other leptin-responsive neurons also plays important roles in promoting leptin's antiobesity effect. As might be predicted from earlier studies,22,41deletion of SOCS-3 selectively from POMC neurons enhances leptin sensitivity, although weight gain with age was normal on a chow diet.67However, on a high-fat diet, the rate of weight gain was reduced. Interestingly, on the chow diet where the body weights were normal, baseline glucose levels were reduced. This altogether supports a role of POMC neurons in leptin's control of glucose balance5,24and suggests a role of SOCS-3 in those cells in the increased weight response to a high-energy diet.68,69In addition to those studies, the PI3K pathway has been examined in significant detail in POMC neurons. Mice with genetically disrupted PI3K signaling lack the normal response of leptin-induced POMC neuronal depolarization and increased firing frequency.70In addition, the suppression of feeding elicited by leptin was blunted. Interestingly, however, despite these alterations in POMC neuronal function, the mice had normal body weight. However, in apparent contrast to these studies, inactivation of PTEN, a phosphatidylinositol-3,4,5-trisphosphate phosphatase, specifically in POMC cells resulted in hyperphagia and a sexually dimorphic diet-sensitive obesity. Interestingly, and in contrast to the study by Hill et al., leptin failed to stimulate POMC electrical activity.71Similar however to the study by Hill et al., leptin was not able to acutely inhibit energy intake. Finally, younger mice with selective inactivation of 3-phosphoinositide-dependent protein kinase 1, an upstream activator of PI3K, in POMC-expressing cells display hyperphagia and increased body weight.72The reasons for the electrical discrepancies between these studies of PI3K signaling in POMC neurons are yet unclear. The observed metabolic differences between the studies may relate to variable impacts on the activity of the hypothalamic-pituitary-adrenal axis because the genetic strategies used also lead to gene alteration in pituitary corticotrophs owing to the activity of the POMC promoter driving CRE expression in those cells. Importantly, and in contrast to the studies of mice with leptin receptor mutations or cell-specific leptin receptor deletions or overexpression, the interpretation of metabolic data from mice with alterations of intracellular proteins (such as PI3K, PTEN, 3-phosphoinositide-dependent protein kinase 1, and adenosine monophosphate-activated protein kinase) suffers significantly from the fact that these proteins are regulated by many stimuli in addition to leptin. Furthermore, these enzymes affect a number of different signaling pathways that vary depending on the stimuli. It is therefore unclear whether the observed metabolic phenotypes are specifically due to alteration in leptin action or to changes in other signaling systems, or both.
Most mouse and rat strains develop obesity when given free access to a high-fat/high-carbohydrate-containing diet. Such diet-induced obese (DIO) rodents are principal models of common-type human obesity. Rodent and human obesity is characterized by hyperleptinemia and by leptin resistance that has yet to be understood.73,74Leptin likely enters the brain via the blood-brain barrier,41,75,76and decreased transport of leptin into the brain of DIO animals has been reported.77However, impaired blood-brain barrier leptin transport may be acquired during development of obesity,77,78suggesting that downstream intracellular signaling defects may be primary causes of leptin resistance. Interestingly, phospho-STAT3 immunohistochemistry on brain sections from leptin-treated DIO mice and rats has demonstrated regional differences in leptin sensitivity. Specifically, leptin signaling within the ARC is dramatically decreased whereas other hypothalamic and extrahypothalamic nuclei seem to remain relatively leptin sensitive.68The decreased leptin signaling of the arcuate nucleus includes POMC and AgRP neurons and is associated with altered release of these neuropeptides and with increased expression of SOCS-3.68,69This suggests that the ARC is selectively leptin resistant in DIO mice and may therefore play a direct role in the development of diet-induced obesity in rodents. Suppressor-of-cytokine-signaling-3 deficiency in the brain68and specifically in POMC neurons67enhances leptin-induced weight loss and protects mice from development of diet-induced obesity. Similarly, neuronal deletion of protein-tyrosine-phosphotase-1B increases leptin sensitivity and attenuates weight gain of high-fat-fed mice.46Altogether, these data suggest that defects in leptin action specifically within the arcuate play a critical role in the pathogenesis of leptin-resistant obesity and that drugs aimed at ameliorating arcuate leptin-resistance might prevent the development of diet-induced obesity. The mechanism by which the arcuate becomes resistant to leptin and the process leading to increased SOCS-3 expression of DIO mice, and the causal-relationship between these events and the appearance of obesity, are critical issues that have yet to be resolved.
In addition to its role in energy homeostasis, leptin can regulate peripheral glucose and insulin balance via the central nervous system. For example, leptin-deficient Lepob/ob mice exhibit profound diabetes that can be fully prevented after 3 weeks of low doses of leptin that do not affect body weight and food intake.5In addition, intracerebroventricular leptin can acutely stimulate glucose uptake in skeletal muscle59,79-81and inhibit hepatic glucose production.82,83Moreover, leptin dramatically improves insulin sensitivity in human lipodystrophy and in lipodystrophic mouse models, which are characterized by low serum leptin levels and by severe insulin resistance.84-86This altogether suggests that leptin has an independent specific capacity to regulate glucose balance, but the neurons mediating this action have remained elusive.
Lack of central melanocortin receptor action in mice results in marked obesity, hyperinsulinemia, and late-onset hyperglycemia,87and insulin resistance is detectable before the onset of obesity in these melanocortin-4-receptor-deficient mice.88In addition, ventricular infusion of α-MSH enhances short-term insulin-stimulated muscle glucose uptake and reduces hepatic glucose production, whereas a melanocortin receptor antagonist exerts opposite effects.89Furthermore, loss of glucose sensing by POMC neurons and subsequent glucose-dependent α-MSH release lead to impaired whole body glucose tolerance.90Moreover, genetic studies in diabetic mice suggest that the arcuate nucleus plays a major role in mediating effects of leptin on glucose balance16; however, the specific arcuate neurons responsible remain unspecified.
Given that arcuate POMC neurons express leptin receptors and that both leptin and the melanocortin system can influence glucose homeostasis, we hypothesized that specifically, POMC neurons mediate this leptin action and recently reported that CRE-mediated expression of ObRb only in POMC neurons in the morbidly obese and severely diabetic leptin receptor-deficient Leprdb/db mice remarkably leads to a complete normalization of blood glucose levels.24This occurred entirely independently of changes in energy intake and body weight. In addition, insulin sensitivity was enhanced, and hypothalamic α-MSH neuropeptide levels were greatly elevated in the transgenic mice. Based on these data, we conclude that leptin signaling in POMC neurons is sufficient to prevent diabetes in Leprdb/db mice and that this action might be mediated by the central melanocortin pathway. Future studies are needed to explain how deletion of leptin receptors in POMC of normal mice does not lead to significant impairments in glucose balance22but that reexpression of receptors in POMC neurons of diabetic Leprdb/db mice leads to this dramatic correction of blood glucose levels.24Regardless, POMC neurons and their downstream neurocircuitries hold promise for identifying novel pathways, which may eventually help develop antidiabetes drugs for humans experiencing severe insulin-resistant diabetes and morbid obesity.
Important questions and future areas of research include (a) determination of the role of individual brain nuclei and specific neuronal groups in each of leptin's actions; (b) identification of the mechanism underlying the redundancy of different brain regions each capable of mediating leptin's effects on intake inhibition; (c) studies aimed at explaining how different groups of neurons have additive effects on body weight regulation; (d) experiments directed toward increasing our understanding of how leptin receptor reexpression in POMC neurons of diabetic Leprdb/db mice appears to play a major role in glucose control, whereas deletion of the receptor from POMC neurons in normoglycemic lean mice does not lead to impairment of glucose homeostasis; (e) elucidation of the specific neurocircuitries downstream of POMC neurons and the peripheral processes responsible for the control of blood glucose by leptin in Leprdb/db mice; (f) identification of specific roles and relative importance of individual intracellular leptin receptor signaling pathways in neuronal functions, including regulation of electrical activity, neuromodulation, and gene expression; and (g) determination of the primary causes of diet-induced obesity and the role of leptin resistance in its development.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.