The melanocortin-4 receptor (MC4-R) is a G protein-coupled, seven-transmembrane receptor expressed in the brain. Inactivation of this receptor by gene targeting results in mice that develop a maturity onset obesity syndrome associated with hyperphagia, hyperinsulinemia, and hyperglycemia. This syndrome recapitulates several of the characteristic features of the agouti obesity syndrome, which results from ectopic expression of agouti protein, a pigmentation factor normally expressed in the skin. Our data identify a novel signaling pathway in the mouse for body weight regulation and support a model in which the primary mechanism by which agouti induces obesity is chronic antagonism of the MC4-R.

The recent positional cloning of the mouse ob gene and its human homology has provided the basis to investigate the potential role of the ob gene product in body weight regulation. A biologically active form of recombinant mouse OB protein was overexpressed and purified to near homogeneity from a bacterial expression system. Peripheral and central administration of microgram doses of OB protein reduced food intake and body weight of ob/ob and diet-induced obese mice but not in db/db obese mice. The behavioral effects after brain administration suggest that OB protein can act directly on neuronal networks that control feeding and energy balance.

The hypothesis that leptin (OB protein) acts in the hypothalamus to reduce food intake and body weight is based primarily on evidence from leptin-deficient, ob/ob mice. To investigate whether leptin exerts similar effects in normal animals, we administered leptin intracerebroventricularly (icv) to Long-Evans rats. Leptin administration (3.5 g icv) at the onset of nocturnal feeding reduced food intake by 50% at 1 h and by 42% at 4 h, as compared with vehicletreated controls (both P Ͻ 0.05). To investigate the basis for this effect, we used in situ hybridization (ISH) to determine whether leptin alters expression of hypothalamic neuropeptides involved in energy homeostasis. Two injections of leptin (3.5 g icv) during a 40 h fast significantly decreased levels of mRNA for neuropeptide Y (NPY, which stimulates food intake) in the arcuate nucleus ( Ϫ 24%) and increased levels of mRNA for corticotrophin releasing hormone (CRH, an inhibitor of food intake) in the paraventricular nucleus (by 38%) (both P Ͻ 0.05 vs. vehicle-treated controls). To investigate the anatomic basis for these effects, we measured leptin receptor gene expression in rat brain by ISH using a probe complementary to mRNA for all leptin receptor splice variants. Leptin receptor mRNA was densely concentrated in the arcuate nucleus, with lower levels present in the ventromedial and dorsomedial hypothalamic nuclei and other brain areas involved in energy balance. These findings suggest that leptin action in rat hypothalamus involves altered expression of key neuropeptide genes, and implicate leptin in the hypothalamic response to fasting. ( J. Clin. Invest. 1996. 98:1101-1106.)

Melanocortins are peptides, cleaved from the pro-opiomelanocortin (POMC) precursor, that act in the brain to reduce food intake and are potential mediators of leptin action. In the forebrain, melanocortins are derived from POMC-containing neurons of the hypothalamic arcuate nucleus. To test the hypothesis that these POMC neurons are regulated by leptin, we used in situ hybridization to determine whether reduced leptin signaling (as occurs in fasting), genetic leptin deficiency (in obese ob/ob mice), or genetic leptin resistance (in obese db/db mice) lower expression of POMC mRNA. We further hypothesized that leptin administration would raise hypothalamic POMC mRNA levels in leptin-deficient animals, but not in mice with defective leptin receptors. In wild-type mice (n = 12), fasting for 48 h lowered POMC mRNA levels in the rostral arcuate nucleus by 53%, relative to values in fed controls (n = 8; P < 0.001). Similarly, arcuate nucleus POMC mRNA levels were reduced by 46 and 70% in genetically obese ob/ob (n = 6) and db/db mice (n = 6), respectively, as compared with wild-type mice (n = 5) (P < 0.01 for both comparisons). Five daily intraperitoneal injections of recombinant murine leptin (150 microg) raised levels of POMC mRNA in the rostral arcuate nucleus of ob/ob mice (n = 8) by 73% over saline-treated ob/ob control values (n = 8; P < 0.01), but was without effect in db/db mice (n = 6). In normal rats, two injections of a low dose of leptin (3.5 microg) into the third cerebral ventricle (n = 15) during a 40-h period of fasting also increased POMC mRNA levels in the rostral arcuate nucleus to values 39% greater than those in vehicle-treated controls (n = 14; P = 0.02). We conclude that reduced central nervous system leptin signaling owing to fasting or to genetic defects in leptin or its receptor lower POMC mRNA levels in the rostral arcuate nucleus. The finding that leptin reverses this effect in ob/ob, but not db/db, mice suggests that leptin stimulates arcuate nucleus POMC gene expression via a pathway involving leptin receptors. These findings support the hypothesis that leptin signaling in the brain involves activation of the hypothalamic melanocortin system.

Mutations reducing the functional activity of leptin, the leptin receptor, alpha-melanocyte stimulating hormones (alpha-MSH) and the melanocortin-4 receptor (Mc4r) all lead to obesity in mammals. Moreover, mutant mice that ectopically express either agouti (Ay/a mice) or agouti-related protein (Agrp), antagonists of melanocortin signalling, become obese. These data suggest that alpha-MSH signalling transduced by Mc4r tonically inhibits feeding; however, it is not known to what extent this pathway mediates leptin signalling. We show here that Mc4r-deficient (Mc4r-/-) mice do not respond to the anorectic actions of MTII, an MSH-like agonist, suggesting that alpha-MSH inhibits feeding primarily by activating Mc4r. Obese Mc4r-/-mice do not respond significantly to the inhibitory effects of leptin on feeding, whereas non-obese Mc4r-/- mice do. These data demonstrate that melanocortin signalling transduced by Mc4r is not an exclusive target of leptin action and that factors resulting from obesity contribute to leptin resistance. Leptin resistance of obese Mc4r-/- mice does not prevent their response to the anorectic actions of ciliary neurotrophic factor (CNTF), corticotropin releasing factor (CRF), or urocortin; or the orexigenic actions of neuropeptide Y (NPY) or peptide YY (PYY), indicating that these neuromodulators act independently or downstream of Mc4r signalling.