ENERGY HOMEOSTASIS IS regulated by discrete central

Transcription

1 /03/$15.00/0 Endocrinology 144(9): Printed in U.S.A. Copyright 2003 by The Endocrine Society doi: /en Fasting Activates the Nonhuman Primate Hypocretin (Orexin) System and Its Postsynaptic Targets SABRINA DIANO, BALAZS HORVATH, HENRYK F. URBANSKI, PETER SOTONYI, AND TAMAS L. HORVATH Departments of Obstetrics and Gynecology (S.D., T.L.H.) and Neurobiology (B.H., T.L.H.), Yale University School of Medicine, New Haven, Connecticut, 06520; Division of Neuroscience (H.F.U.), Oregon Regional Primate Research Center, Beaverton, Oregon, 97006; and Department of Anatomy and Histology (P.S.), Szent Istvan University, School of Veterinary Medicine, Budapest, Hungary, 1071 ENERGY HOMEOSTASIS IS regulated by discrete central nervous system circuits that, in predominantly hypothalamic areas, are sensitive to peripheral metabolic signals (1). A great part of our current understanding of human physiology and disorders of the hypothalamus is inferred from rodent models, whereas little is known about the primate hypothalamic circuits that control homeostasis. In particular, it is not clear whether the same peptidergic circuits known to be involved in the hypothalamic regulation of rodent homeostasis have similar functions in the primate brain. Hypocretin (HCRT, also called orexin), a peptide produced exclusively in the hypothalamus (2 4), has emerged as an important regulator of autonomic, endocrine, and metabolic functions of rodents (3, 5, 6). HCRT cells of rats and monkeys were found to express leptin receptor immunoreactivity and to provide synaptic input to distinct hypothalamic and brainstem circuits known to participate in the central regulation of endocrine, metabolic, and autonomic systems (4, 7). Although the determination of synaptic interactions between HCRT and other hypothalamic systems provided the map via which signals may travel within the hypothalamus, that does not demonstrate whether that signaling modality is involved in homeostatic regulation. For that determination, the cytochemical identification of c-fos as Abbreviations: ARC, Arcuate nucleus; DAB, diaminobenzidine; HCRT, hypocretin; LH-PF, lateral hypothalamus-perifornical; MPOA, medial preoptic area; NPY, neuropeptide Y; PB, phosphate buffer. In rodents, hypocretin (HCRT, also called orexin) influences a variety of endocrine, autonomic, and metabolic functions. The present study was undertaken to determine whether the HCRT-producing circuit is involved in the hypothalamic regulation of homeostasis in primates as well. We studied female monkeys (Cercopithecus aethiops) that were either fed or fasted for 24 h. Immunocytochemistry revealed HCRTproducing perikarya exclusively in the lateral hypothalamusperifornical region and dorsomedial hypothalamus of the monkey brain. HCRT axons and axon terminals were present in different parts of the hypothalamus and adjacent forebrain and thalamic nuclei. The 24-h fast resulted in an approximately 50% decline in circulating leptin levels and significantly elevated c-fos expression in the perifornical region; in the dorsomedial, ventromedial, and arcuate nuclei; and in the medial preoptic area. In the dorsomedial nucleus and perifornical region of fasted monkeys, three times more HCRTneurons expressed nuclear c-fos than those of the normally fed controls. Neurons in different parts of the hypothalamus and basal forebrain that expressed c-fos, but did not contain HCRT, were targets of HCRT-immunopositive boutons establishing asymmetric synapses. In the arcuate nucleus, subsets of these HCRT-targeted c-fos-expressing cells contained neuropeptide Y. The present study provides the first experimental evidence to implicate HCRT in the hypothalamic regulation of homeostasis in primates. The fact that these lateral hypothalamic cells have leptin receptors and can be activated by a metabolic challenge and that they innervate diverse brain regions indicates that the HCRT system may be a key integrator of environmental cues in their regulation of diverse brain activity. (Endocrinology 144: , 2003) an indicator of cell activity under different metabolic conditions may be used. After metabolic challenge, distinct populations of hypothalamic neurons begin to express c-fos in rats (8) as well as in monkeys (9). Using this paradigm, the current study aimed to determine whether, in the monkey hypothalamus, HCRT cells and their postsynaptic targets are activated by a metabolic challenge (fasting). Animals Materials and Methods Adult ( kg) female African green monkeys (Cercopithecus aethiops; n 6) were used. All monkeys were previously ovariectomized for use in an unrelated experiment; three monkeys were killed after fasting (food withdrawal for 24 h), whereas the other three were killed without fasting. The primate tissue was collected under protocols approved by the Yale Animal Care and Use Committee. After receiving a lethal dose of anesthetic, monkeys were killed by a transcardial perfusion of 500 ml heparinized saline (0.9%) followed by 2000 ml of fixative consisting of 4% paraformaldehyde, 15% saturated picric acid, and 0.08% glutaraldehyde in 0.1 m phosphate buffer (PB) at ph 7.4. The mediobasal hypothalamus was dissected out and postfixed for an additional 1.5 h in glutaraldehyde-free fixative. Tissue blocks were washed and stored in 0.1 m PB at 4 C. Light and electron microscopic double immunostaining Light microscopic double immunostaining for HCRT and c-fos was carried out according to our previously published protocol. Sections were incubated with one of the primary antisera (rabbit anti-hcrt, 1:2000) and processed with the avidin-biotin-peroxidase technique. The 3774

2 Diano et al. Brief Communications Endocrinology, September 2003, 144(9): immunoreaction was visualized with a modified version of the nickeldiaminobenzidine (DAB) reaction (15 mg DAB, 0.12 mg glucose oxidase, 12 mg ammonium chloride, 600 l 0.05 m nickel ammonium sulfate, and 600 l 10% -d-glucose in 30 ml PB), resulting in a dark-blue reaction product. After several rinses in PB, the sections were further incubated in sheep anti-c-fos (1:2000; Cambridge Research Biochemicals Inc., Wilmington, DE) for 24 h at 4 C and processed with the peroxidase-antiperoxidase technique. The tissue-bound peroxidase was visualized with a DAB reaction (15 mg DAB and 165 l 0.3% H 2 O 2 in 30 ml PB), resulting in a light-brown reaction product. After visualization of tissue antigens, some sections were processed for electron microscopy (1% OsO 4 in PB for 30 min., dehydrated through increasing ethanol concentrations using 1% uranyl acetate in the 70% ethanol for 30 min) and flat-embedded in araldite between liquid release-coated slides (Electron Microscopy Sciences, Fort Washington, PA). The specificity of the HCRT antisera has been thoroughly tested by us and others in previous studies (2, 4). Analysis of c-fos expressing cells in single- and doublelabeled (c-fos and HCRT) material To estimate the induction of c-fos expression in the primate hypothalamus after 24 h of fasting, matching sections from the hypothalami of control and fasted monkeys were coincubated in the same vials for either single immunostaining for c-fos or for double immunostaining of c-fos and HCRT (see above). To ensure that matching sections were coprocessed, sections were selected using the comparative assessment of the location of the following anatomical landmarks: optic chiasm or tract, shape of the third ventricle, anterior commissure, fornix, mammilo-thalamic tract, and median eminence. One group of sections (either the control or the experimental) was marked by placing a notch on one side of the tissue block before vibratome sectioning. The analyzer was blind to the code, which was revealed after the collection of the counts. When the immunostaining was completed (three sets of coincubations from six monkeys three control and three experimental), matching sections were placed on slides, coverslipped, and analyzed using light microscopy. The number of c-fos-immunolabeled cells, those that were labeled for HCRT, and those that were labeled for both antigens were noted in several nuclei of the hypothalamus (see below). Only those cells on the surface of the section were counted. Guided by the stereotaxic atlas of the African Green monkey (10), the following hypothalamic areas were analyzed: medial preoptic area (MPOA), parvicellular and magnocellular paraventricular nucleus, arcuate (infundibular) nucleus (ARC), dorsomedial hypothalamic nucleus, ventromedial nucleus, and the lateral hypothalamus perifornical region. After determining the mean values, the Student s t test was used for comparison. Significance was concluded at P Light microscopic triple immunostaining Because HCRT is not present in neuronal perikarya of the ARC, c-fos is only expressed in cell nuclei, and neuropeptide Y (NPY) is expressed in the cytoplasm of neurons, we triple-labeled hypothalamic sections using our triple-labeling protocol (4). Sections were first incubated for 24 h at room temperature with a mixture of the HCRT and c-fos antisera and processed with the avidin-biotin-peroxidase, and the tissue-bound peroxidase was visualized by a nickel-diaminobenzidine reaction, resulting in a dark-blue to black color. Subsequently, sections were further immunostained for NPY using sheep anti-npy (Auspep Pty Ltd., Perkwille, Australia) and the peroxidase antiperoxidase method, and the tissue-bound peroxidase was visualized by a DAB reaction to give a light-brown reaction product. Leptin measurements All of the blood samples were collected in the morning, and the serum was stored frozen. Samples were collected before and after animals were fasted. Leptin was measured using primate leptin RIA kit (Linco Research, Inc., St. Charles, MO). The minimum detectable concentration at 95% binding was 0.7 ng/ ml, and the interassay coefficient of variation was 6%; all of the serum samples were assayed together in a single RIA. A Student s t test with two-tail probabilities was used for statistical comparisons of leptin values between the two groups. Significance was concluded at P Results Circulating leptin levels in fasted and normally fed control African green monkeys The circulating leptin level values did not differ between the fasted and control groups at 0 h ( vs ng/ml; P 0.05), but they were significantly lower in the fasted animals after the 24-h fast ( vs ng/ml; P 0.05). HCRT in the hypothalamus. The overall pattern of HCRT immunolabeling in monkey corresponded to our earlier de- FIG. 1. Light micrographs of the monkey perifornical region (pf) after immunostaining for c- fos taken from control (A) and fasted animals (B). Bar scale, 100 m. f, Fornix. C, Bar graph showing the number of c-fos-immunoreactive nuclei in different parts of the hypothalamus in control (white bars) and fasted (black bars) animals.

3 3776 Endocrinology, September 2003, 144(9): Diano et al. Brief Communications scription (4, 7). HCRT-immunoreactive perikarya were present exclusively in the lateral hypothalamus, perifornical region, and, to a lesser extent, in the dorsomedial hypothalamus, whereas their projections were abundant in different hypothalamic nuclei. We detected no obvious differences between the HCRT immunoreactivity in the hypothalamus of control vs. fasted animals. c-fos. In accordance with a recent report (9), c-fos immunoreactivity was detected in different parts of the hypothalamus of both fasted and normally fed control monkeys. In control animals, a moderate number of c-fos-labeled nuclei were homogenously distributed in the MPOA; periventricular regions; paraventricular nucleus (both parvo- and magnocellular regions); anterior hypothalamus; suprachiasmatic, arcuate, dorsomedial, and ventromedial nuclei; lateral hypothalamus; and perifornical region. The number of c-fosexpressing cells in individual hypothalamic subnuclei of fasted monkeys was distinctly different from the control values (Fig. 1, A C). In fasted animals, a significantly higher number of c-fos-expressing cells was detected in the MPOA ( vs *; *, P 0.05), lateral hypothalamusperifornical (LH-PF) region ( vs *; *, P 0.05; Fig. 1, A and B), and arcuate ( vs *; *, P 0.05) and dorsomedial nuclei ( vs *; *, P 0.05). In addition, differences, albeit nonsignificant, were detected in the ventromedial-( vs , P 0.05) and in the parvicellular paraventricular ( vs , P 0.05) and magnocellular paraventricular hypothalamic nuclei ( vs , P 0.05). However, it should be noted that the low sample values (n 3) may mask significant differences that might have been detected with higher number of animals studied. c-fos expression in HCRT cells c-fos immunoreactivity was detected in cell nuclei of HCRT-immunoreactive cells of both control and 24-h-fasted animals (Fig. 2, A D). There was no difference in the mean number of HCRT-labeled cells in control and fasted animals ( vs , P 0.05; Fig. 2E). However, the percentage of HCRT cells expressing nuclear c-fos was robustly and significantly elevated in the fasted animals when compared with control values (P 0.05). In the control LH- PF, HCRT cells exhibited nuclear c-fos labeling, which represented 26.8% of the total HCRT cells counted. In contrast, in fasted animals, HCRT cells exhibited c-fos immunoreactivity, which represented 78.1% of the total HCRT cells counted (Fig. 2). The number of c-fos-expressing neurons that contained HCRT in control monkeys (135 35) represented approximately 75% of all the cells that were c-fos immunopositive (182 29) in the LH-PF region. This ratio remains similar in fasted animals, in which of the total c-fos-immunolabaled nuclei (562 42) were in HCRTimmunoreactive perikarya (Fig. 2F). Hypocretin axons synapse on c-fos-containing cells FIG. 2. Fasting-induced c-fos expression was associated exclusively with cell nuclei (A and C). In the perifornical region, many nuclear c-fos-labeled cells (red nuclei in B and D) were also immunopositive for HCRT. Bar scales (in A and C), 10 m (for A and B) and 100 m (for C and D). E, Bar graph showing c-fos-immunoreactive HCRT neurons (white bars) and total HCRT neurons (black bars) counted in control and fasted animals. F, Bar graph showing c-fos-immunoreactive HCRT neurons (white bars) and total c-fos-immunolabeled neurons (black bars) counted in control and fasted animals. When analyzed with light microscopy, it was apparent that both in the control and fasted animals, c-fos-immunolabeled cells were surrounded by numerous HCRT-immunopositive boutons (Fig. 3A). In the regions in which there were significant elevations in c-fos labeling after fasting, the MPOA, ARC, LH-PF and dorsomedial nucleus, we found no c-fos-immunolabeled cells either in the control or experimental group that were not contacted by numerous HCRTlabeled axon terminals. Synaptic membrane specializations between HCRT boutons and c-fos-expressing cells were asymmetrical contacts (Fig. 3, B and C). In the ARC, in which increased c-fos expression was detected after fasting and hypothalamic NPY-producing cells were located, HCRT axons were in close proximity to perikarya that contained both NPY and c-fos (Fig. 4). Discussion The present study provides the first suggestive evidence that the HCRT system of nonhuman primates is involved in

4 Diano et al. Brief Communications Endocrinology, September 2003, 144(9): FIG. 3. In all regions of the monkey hypothalamus in which c-fos induction was observed after fasting in cell nuclei, HCRT boutons were observed in close apposition to these activated cells (A), and electron microscopy showed (B and C) that these contacts were synapses (arrows in C point to membrane specializations). Bar scales, 10 m (for A) and 1 m (for B and C). FIG. 4. A subset of fasting-activated, HCRT-targeted cells were NPY cells in the ARC. Bar scale, 10 m. the hypothalamic signal transduction associated with metabolic alterations. The fact that short-term food withdrawal underlies robust activation of transcriptional mechanisms in the primate lateral hypothalamic HCRT system, together with the widespread distribution of HCRT projections in homeostatic centers and beyond, argues for a coordinating role for HCRT in autonomic, endocrine, and metabolic processes. This assertion is in line with evidence that is available for the role of HCRT in the modulation of food intake, blood pressure, and arousal (3, 5, 6, 11, 12). In a previous study (4), the monkey ARC was found to receive a robust HCRT innervation, with HCRT axons heavily innervating NPY cells (4). Whether this ARC NPY system may participate in the modulation of HCRT on food intake in primates, as suggested in the rat (13, 14), remains to be tested. The present observation of HCRT input to fasting-activated NPY cells in the African green monkey lends support to this possibility. Nevertheless, it is also conceivable that the activation of HCRT neurons by fasting occurs secondary to the activation of the ARC. However, ARC NPY cells are most likely protected by the blood-brain barrier, and the direct projection from the ARC to the lateral hypothalamus is not robust in the nonhuman primate (4). It may also be plausible that, because of the HCRT system is heavily involved with the regulation of sleep/wake cycles, it is disturbance of sleep during fasting that affects HCRT neuronal function. Alternatively, should metabolic signals affect HCRT cells during fasting, sleep disturbances may be the consequence HCRT neuronal activation. Leptin receptor immunoreactivity has been detected in the primate lateral hypothalamic HCRT cells (4), and leptin is suggested to cross the blood-brain barrier (15). Thus, the observed induction of c-fos gene expression in the HCRT cells may be a direct consequence of the diminishing leptin levels we observed after the short-term fasting. However, fasting paradigms, particularly when used on primates, do not allow for the distinction of the effect of particular hormones, because they underlie complex responses of a variety of endocrine mechanisms. For example, in the rodent, evidence is available to indicate that circulating insulin and/or glucose levels have a regulatory influence on hypothalamic HCRT expression (16). Future studies will determine the extent of contribution of each of the involved humoral and metabolic signals in the activation of primate HCRT systems. Interestingly, in the hypothalamic nuclei (including the ARC, LH-PF, dorsomedial nucleus, and MPOA), in which short-term fasting induced c-fos expression the most robustly, activated cells received multiple HCRT inputs. In rodents, HCRT exerts excitatory postsynaptic currents (17), and intracerebroventricular administration of HCRT induces c-fos expression in a manner not dissimilar to what we describe here in the nonhuman primate (18, 19). Because all the synaptic membrane specializations between HCRT axons and c-fos-expressing perikarya in the primate hypothalamus were asymmetrical, stimulatory synapses, the activation of hypothalamic circuitry by metabolic challenge may be mediated, in part, by the HCRT system. Acknowledgments We thank Anthony van den Pol for providing the HCRT antisera. Received March 3, Accepted June 24, Address all correspondence and requests for reprints to: Tamas L. Horvath, Department of Obstetrics and Gynecology, Yale Medical School, 333 Cedar Street, FMB 339, New Haven, Connecticut tamas.horvath@yale.edu. This work was supported by NIH Grant RR T.L.H. was an Albert Szent-Gyorgyi Fellow. References 1. Kalra, SP, Xu B, Dube MG, Pu S, Horvath TL, Kalra PS 1999 Interacting appetite regulating pathways in the hypothalamic regulation of body weight. Endocr Rev 20: de Lecea L, Kilduff TS, Peyron C, Gao XB, Foye PE, Danielson PE, Fukuhara C, Battenberg ELF, Gautvik VT, Bartlett II FS, Frankel WN, van den Pol AN, Bloom FE, Gautvik KM, Sutcliffe JG 1998 The hypocretins: two hypothalamic peptides with neuroexcitatory activity. Proc Natl Acad Sci USA 95: Sakurai T, Amemiya A, Ishii M, Matsuzaki I, Chemelli RM, Tanaka H, Williams SC, Richardson JA, Kozlowski GP, Wilson S, Arch JRS, Bucking-

5 3778 Endocrinology, September 2003, 144(9): Diano et al. Brief Communications ham RE, Haynes AC, Carr SA, Annan RS, McNulty DE, Liu W-S, Terrett JA, Elshourbagy NA, Bergsma DJ, Yanagisawa M 1998 Orexins and orexin receptors: a family of hypothalamic neuropeptides and G protein-coupled receptors that regulate feeding behavior. Cell 92: Horvath TL, Diano S, van den Pol AN 1999 Synaptic interaction between hypocretin (orexin) and neuropeptide Y cells in the rodent and primate hypothalamus: a novel circuit implicated in metabolic and endocrine regulations. J Neurosci 19: Chemelli RM, Willie JT, Sinton CM, Elmquist JK, Scammell T, Lee C, Richardson JA, Williams SC, Xiong Y, Kisanuki Y, Fitch TE, Nakazato M, Hammer RE, Saper CB, Yanagisawa M 1999 Narcolepsy in orexin knockout mice: molecular genetics of sleep regulation. Cell 98: Lin L, Faraco J, Li R, Kadotani H, Rogers W, Lin X, Qiu X, de Jong PJ, Nishino S, Mignot E 1999 The sleep disorder canine narcolepsy is caused by a mutation in the hypocretin (orexin) receptor 2 gene. Cell 98: Horvath TL, Peyron C, Diano S, Ivanov A, Aston-Jones G, Kilduff TS, van den Pol AN 1999 Hypocretin (orexin) activation and synaptic innervation of the locus coeruleus noradrenergic system. J Comp Neurol 415: Xu B, Li BH, Rowland NE, Kalra SP 1995 Neuropeptide Y injection into the fourth cerebroventricle stimulates c-fos expression in the paraventricular nucleus and other nuclei in the forebrain: effect of food consumption. Brain Res 698: Caston-Balderrama AL, Cameron JL, Hoffman GE 1998 Immunocytochemical localization of Fos in perfused nonhuman primate brain tissue: fixation and antisera selection. J Histochem Cytochem 46: Contreras CM, Mexicano G, Guzman-Flores C 1981 A stereotaxic brain atlas of the green monkey (Cecopithecus aethiops aethiops). Bol Estud Med Biol 31: Samson WK, Gosnell B, Chang JK, Resch ZT, Murphy TC 1999 Cardiovascular regulatory actions of the hypocretins in brain. Brain Res 83: Kukkonen JP, Holmqvist T, Ammoun S, Akerman KE 2002 Functions of the orexinergic/hypocretinergic system. Am J Physiol Cell Physiol 283:C1567 C Jain MR, Horvath TL, Kalra PS, Kalra SP 1999 Evidence that NPY Y1 receptors are involved in stimulation of feeding by orexins (hypocretins) in sated rats. Regul Pept 87: Lopez M, Seoane LM, Garcia Mdel C, Dieguez C, Senaris R 2002 Neuropeptide Y, but not agouti-related peptide or melanin-concentrating hormone, is a target peptide for orexin-a feeding actions in the rat hypothalamus. Neuroendocrinology 75: Bjorbaek C, Elmquist JK, Michl P, Ahima RS, Vanbueren A, Mccall AL, Flier JS 1998 Expression of leptin receptor isoform in rat brain microvessels. Endocrinology 139: Cai XJ, Widdowson PS, Harrold J, Wilson S, Buckingham RE, Arch JR, Tadayyon M, Clapham JC, Wilding J, Williams G 1999 Hypothalamic orexin expression: modulation by blood glucose and feeding. Diabetes 11: van den Pol AN, Gao XB, Obrietan K, Kilduff T, Belousov A 1998 Pre- and postsynaptic actions and modulation of neuroendocrine neurons by a new hypothalamic peptide, hypocretin/orexin. J Neurosci 18: Date Y, Ueta Y, Yamashita H, Yamaguchi H, Matsukura S, Kangawa K, Sakurai T, Yanagisawa M, Nakazato M 1999 Orexins, orexigenic hypothalamic peptides, interact with autonomic, neuroendocrine and neuroregulatory systems. Proc Natl Acad Sci USA 96: Mullett MA, Billington CJ, Levine AS, Kotz CM 2000 Hypocretin I in the lateral hypothalamus activates key feeding-regulatory brain sites. Neuroreport 11: