Posted by: Indonesian Children | April 14, 2010

Journal Watch : Orexins in the Brain-Gut Axis

Endocrine Reviews 23(1):1–15 Copyright © 2002 by The Endocrine Society

Orexins in the Brain-Gut Axis

ANNETTE L. KIRCHGESSNER  Department of Physiology and Pharmacology, State University of New York Downstate Medical Center, Brooklyn, New York 11203-2098

Orexins (hypocretins) are a novel pair of neuropeptides im­plicated in the regulation of energy balances and arousal. Previous reports have indicated that orexins are produced only in the lateral hypothalamic area, although orexin-containing nerve fibers were observed throughout the neu-roaxis. Recent evidence shows that orexins and functional orexin receptors are found in the periphery. Vagal and spinal primary afferent neurons, enteric neurons, and endocrine cells in both the gut and pancreas display orexin- and orexin receptor-like immunoreactivity. Orexins excite secretomotor neurons in the guinea pig gut and modulate gastric and intestinal motility and secretion. In addition, orexins modulate hormone release from pancreatic endocrine cells. Moreover, fasting up-regulates the phosphorylated form of cAMP re­sponse elementbinding proteininorexin-immunoreactive en­teric neurons, indicating a functional response to food status in these cells. The purpose of this article is to summarize evidence for the existence of a brain-gut network of orexin-containing cells that appears to play a role in the acute reg­ulation of energy homeostasis. (Endocrine Reviews 23: 1–15, 2002)

Abbreviations: [Ca2+]i, Intracellular calcium concentration; CCK, cholecystokinin; CNS, central nervous system; DMV, dorsal motor nu­cleus of the vagus; EC, enterochromaffin cell; ENS, enteric nervous system; 5-HT, 5-hydroxytryptamine; LHA, lateral hypothalamic area; MCH, melanin concentrating hormone; NPY, neuropeptide Y; NTS, nucleus of the solitary tract; PVN, paraventricular nucleus; VIP, vaso-active intestinal peptide; VMH, ventromedial hypothalamic nucleus.

THE OREXINS (1), also called hypocretins (2), were first described just 4 yr ago; yet more than 280 papers on the peptides have already been published. Interest in the orexins began with the observation that these novel neuropeptides are produced by a small group of neurons in the perifornical and lateral hypothalamic area [LHA (1–4)], a region classi­cally implicated in the control of mammalian feeding be­havior   (5–7).   Intracerebroventricular   administration   (1,8–10), or direct injection of orexins into the LHA (11), has been shown to increase food intake in rodents in a dose-dependent manner. Conversely, an orexin receptor antibody inhibits food intake in fasted rats, indicating that endogenous orexins are necessary for feeding (12). Furthermore, orexin mRNA expression is up-regulated by fasting (1, 13), sug­gesting that these neurons become activated under condi­tions of hunger.

Orexin neurons project within the hypothalamus and throughout the central nervous system (CNS) to nuclei known to be important in the control of feeding (3, 4, 14, 15). In addition, abundant orexin nerve fibers and orexin recep­tors are found in nuclei concerned with maintenance of wakefulness (14, 15), and several reports implicate a dys­function of the orexin system in human and canine narco­lepsy (16–18). Recently, genetic ablation of orexin neurons in mice was shown to result in narcolepsy, hypophagia, and obesity (19). This finding confirms the importance of hypo-thalamic orexin-containing neurons in the regulation of sleep/wake states and further suggests a role in energy metabolism.

Although the hypothalamus has received considerable at­tention regarding energy homeostasis, the gut also partici­pates in the regulation of food intake. The presence of food in the bowel, through activation of chemo- and mechano-sensitive endings, stimulates the release of several regulatory peptides that control gut motility and secretion (20, 21). Sev­eral of these peptides [for example, cholecystokinin (CCK)] also act as feedback “satiety” signals responsible for termi­nation of a meal (21–23). Satiety signals are inhibited during fasting, and replaced by “hunger”-related signals, also from the gut, that may be enhanced by low plasma concentrations of glucose (21).

The enteric nervous system (ENS), which is composed of neurons that reside within the wall of the gastrointestinal tract and contains as many neurons as the entire spinal cord (24), directly senses, integrates, and regulates the machinery of the gut involved in energy metabolism (24, 25). It is the only region of the peripheral nervous system that is intrin­sically capable of mediating gut-related reflex activity (25). These reflexes, which may be secretory or peristaltic, are made possible by the presence within the bowel of micro-circuits that contain the necessary primary afferent neurons and interneurons, as well as the excitatory and inhibitory motor neurons that innervate gastrointestinal smooth muscle and glands.

Recently, neurons and endocrine cells in the gut were reported to display orexin-like immunoreactivity (26). In ad­dition, orexins were shown to modulate the electrical prop­erties and synaptic inputs of secretomotor neurons and stim­ulate colonic motility (26). We also provided evidence that orexins were increased during periods of fasting, indicating a functional response to food status in these cells (26). With these observations, orexins were added to the growing list of peptides that coexist and act in both the CNS and ENS, giving further support to the concept of a peptidergic brain-gut axis involved in the regulation of feeding and energy homeostasis.

The discovery that orexins are produced in the gut is not surprising. Many different classes of neurotransmitter have been found in the ENS, including glutamate (27), the major excitatory neurotransmitter of the brain (28). The gut also contains, in neurons and/or epithelial endocrine cells, most of the orexigenic and anorectic neuropeptides found in the hypothalamus, including neuropeptide Y [NPY (29, 30)], ghrelin (31), galanin (32, 33), and cocaine- and amphetamine-regulated transcript (34, 35). Moreover, like the hypothala-mus, the ENS contains neurons that are sensitive to glucose (36) and are modulated by the adipocyte-derived hormone leptin (36). Although some of these factors may lack phys­iological relevance in the regulation of food intake, it is likely that many will prove to act as peripheral mediators of energy homeostasis.

During the past 4 yr, several reviews on the orexins and the central regulation of feeding and arousal have been pub­lished (14, 15, 37–39). The purpose of the present review is to give an overview of the orexins in the brain-gut axis and clarify the role of these neuropeptides in enteric and pan­creatic function.

II. Discovery of the Orexins

The orexins were discovered during a search for endog­enous ligands that activate orphan G protein-coupled recep­tors (1). Using more than 50 cell lines, each expressing a distinct orphan G protein-coupled receptor cDNA, Sakurai et al. (1) tested the ability of HPLC fractions of rat brain extracts, to increase intracellular calcium concentrations ([Ca2+]i). The investigators discovered several HPLC fractions that elicited a robust increase in [Ca2+]i in a human embryonic kidney (HEK293) cell line expressing a receptor originally termed HFGAN72. This receptor was initially identified as an ex­pressed sequence tag from human brain. When the major peak of bioactivity in the fractions was purified to homoge­neity and sequenced, a novel peptide of 33 amino acids in length, with an N-terminal pyroglutamyl residue and an amidated C terminus, was discovered. The peptide, termed orexin-A, also contained two intrachain disulfide bonds, and sequencing of similar extracts from bovine brain revealed exact interspecies homology.

In addition to orexin-A, the HPLC fractions contained two minor peaks of activity, designated B and B’. Peak B con­sisted of a 28-amino acid peptide, termed orexin-B, which also possessed an amidated C terminus and was 46% iden­tical in sequence to orexin-A. Peak Bconsisted of an N-terminally truncated orexin-B, which was termed orexin-B (3-28). Further analysis revealed that both orexin-A and orexin-B were derived from the same 130-amino acid pre­cursor, rat prepro-orexin, by proteolytic processing. Human and mouse prepro-orexin sequences were determined and found to be 83% and 95% identical to their rat counterparts, respectively. Radiation hybrid mapping showed that the hu­man prepro-orexin gene maps to a locus at chromosome 17q21. Thus, the prepro-orexin gene has been proposed to be a candidate gene for a group of neurodegenerative disorders called “chromosome 17-linked dementia” (40).

Independently, de Lecea et al. (2), using directional tag PCR subtraction, identified a hypothalamic-specific mRNA encoding a precursor protein that they called prepro-hypo-cretin and predicted that processing of this prepro-peptide would yield two peptides, one of 39 and another of 29 amino acids. Because the cell bodies that expressed this gene were thought to be located exclusively in the hypothalamus, and because of a weak homology to the gut peptide secretin, the peptides were named hypocretin-1 and hypocretin-2. Sub­sequent comparisons revealed that prepro-orexin and pre-pro-hypocretin were the same gene, and that hypocretin-1 and hypocretin-2 had sequences in common with orexin-A and -B, respectively (15,37). Thus, orexin and hypocretin are the same molecule. Nevertheless, in this article, the term orexins will be used to denote the orexin/hypocretin pep­tides, since orexin immunoreactivity and orexin mRNA expression have been found outside the hypothalamus (1, 26, 41).

III. Characterization of the Orexin Receptors

In their remarkable paper, Sakurai et al. (1) not only re­ported the structure of the orexins, but also identified the amino acid sequences of the receptors for the two peptides. The original HFGAN72 receptor, subsequently called OX1R, was shown to bind orexin-A with high affinity and bind orexin-B with 100- to 1,000-fold lower affinity. However, a related receptor, OX2R, identified by searching database en­tries with the OX1R sequence, was demonstrated to have equally high affinities for both peptides. Thus, OX2R was concluded to be a nonselective receptor for both orexin-A and -B peptides, while OX1R was concluded to be moder­ately selective for orexin-A. The binding of both ligands to either receptor was associated with changes in intracellular calcium concentrations. Evidence from receptor-expressing cells suggests that OX1R is coupled exclusively to the Gq subclass of G proteins, whereas OX2R may couple to Gi/o and/or Gq (1, 15, 42).

IV. The Orexin System in the Brain

Immunohistochemical and in situ hybridization studies have shown that in the CNS, orexin-producing cells are re­stricted to a few nuclei in the hypothalamus, including the perifornical nucleus, the LHA, and the dorsomedial hypo-thalamic nucleus (Fig. 1 and Refs. 1–4, 43, and 44). Orexin neurons are organized bilaterally and symmetrically and have been observed in all species investigated so far, includ­ing bovine, guinea-pig (A. L. Kirchgessner, unpublished ob­servations), hamster, human, monkey, mouse, rat, and frog (Fig. 1and Refs. 1–4 and 43–48). Approximately 1,100 orexin-containing cell bodies have been estimated to be present in the rat brain, using an antibody to prepro-orexin (14). Despite their highly restricted origin, orexin nerve fibers ramify widely throughout the CNS, with particularly abundant pro­jections found in the olfactory bulb, cerebral cortex, thala-mus, hypothalamus, brainstem, and all levels of the spinal cord (3, 4, 13–15, 45–49). The widespread projections of the orexin neurons throughout the neuroaxis suggest that acti­vation of orexin circuits probably modulates a variety of systems (14, 15), including those involved in the regulation of food intake. The fact that orexins can increase the release of either excitatory or inhibitory neurotransmitters, by acting directly on axon terminals (2, 42), indicates that the peptides could ultimately increase or decrease the activity of inner­vated brain circuits.

A. Hypothalamus

Within the hypothalamus, orexin neurons project to the arcuate nucleus (3, 4) and specifically innervate NPY-containing cell bodies (46). Reciprocal connections from NPY neurons to orexin neurons in the LHA have also been iden­tified (43, 45, 46). In addition, orexin-containing nerve fibers terminate in close apposition to NPY-immunoreactive nerve terminals in the paraventricular nucleus [PVN (45, 46)]. NPY

is a potent orexigenic peptide that is released in the PVN and surrounding sites to stimulate feeding (50, 51). Since arcuate NPY neurons are excited by orexins (15, 42), probably through the activation of an OX1R Gq-coupled pathway (15), this suggests that orexin-stimulated feeding might occur through NPY pathways (4, 52). Both NPY Y1 and Y5 receptor antagonists reduce orexin-stimulated feeding (53, 54). Thus, activation of NPY-containing feeding pathways is at least partially responsible for the effects of orexin on food intake.

B. Dorsal vagal complex

In the hindbrain, orexin-immunoreactive fibers are found in the dorsal vagal complex, comprising the nucleus of the solitary tract (NTS) and dorsal motor nucleus of the vagus (DMV), and the area postrema (3, 15). The NTS relays vagally transmitted afferent signals from the gut that are related to feeding (21, 23). The NTS also contains glucosensitive neu­rons that respond with altered electrical activity to changes in blood glucose and the presence of food in the gut (55, 56). The DMV consists of motor neurons that control gut motility and the secretory responses that are especially important for digestion. The DMV is also responsible for the initiation of cephalic-phase responses that prepare the gut for the arrival and subsequent digestion of nutrients (57). The area pos-trema, which is interconnected with the dorsal vagal com­plex, is an important circumventricular organ through which circulating systemic factors, such as gut peptides and glu­cose, can gain access to the brain (58).

Although the typesof neurons in the dorsal vagal complex that are innervated by orexin-containing nerve terminals have not been identified, it is likely that orexins alter the activity of vagal motor neurons and/or modulate the re­sponse of NTS neurons to gastrointestinal stimuli. It has been demonstrated that stimulation of the LHA excites neurons in the DMV (59) and increases the activity of vagal efferents

(60). Thus, it is reasonable to expect that modulation of vagal activity by orexins could influence cephalic phase reflexes and/or affect gastrointestinal motility and secretion. Fur­thermore, since neurons in the NTS project back to the LHA (61), orexin neurons may be regulated by satiety signals relayed through the NTS.

C. Distribution of orexin receptors

In parallel to the diffuse orexin-containing projections from the LHA, in situ hybridization studies with orexin re­ceptor riboprobes demonstrate that orexin receptors are ex­pressed in a pattern consistent with orexin nerve fibers (62– 65). However, the expression patterns for OX1R and OX2R are strikingly different. Within the hypothalamus, OX1R mRNA is most abundant in the dorsomedial portion of the ventromedial hypothalamic nucleus (VMH). Dense expres­sion of OX1R is also found in the anterior hypothalamic area just dorsal to the suprachiasmatic nucleus. In contrast, OX2R mRNA is expressed in many hypothalamic nuclei including the tuberomammillary nucleus, the LHA, the arcuate nu­cleus, and PVN (63, 65). The tuberomammillary nucleus is the only source of histamine in the CNS. Since histamine is crucial for the maintenance of wakefulness (66), OX2R in this region has been postulated to play a role in the regulation of sleep/wake states (65).

The expression of orexin receptors in the VMH, LHA, arcuate nucleus, and PVN is consistent with a role of hypo-thalamic orexin systems in regulating food intake. The VMH is strongly implicated in the regulation of food intake since its destruction causes obesity (67). The PVN is the site of action of many orexigenic agents including NPY (50, 51) and galanin (68). The PVN is also involved in the regulationof gut functions via its projection to the dorsal vagal complex (69). For example, stimulation of the PVN evokes an increase in gastric acid secretion (70) and a transient increase in motility (71).

Surprisingly, little orexin receptor mRNA has been de­tected in the NTS and DMV (65), although these regions appear to receive a moderately dense orexin innervation (3, 15). Locations with dense orexin immunoreactivity but little receptor mRNA may reflect presynaptic innervation on axon terminals (42), whose cell bodies are located at some dis­tances. Recent findings support a presynaptic action of orexin-Ainthe DMV (72). However,apostsynaptic action on DMV neurons via OX1R receptors has also been demon­strated and appears to mediate the stimulatory effects of central orexin-A on gastric motility (72).

Other brain areas that display relatively dense expression of OX1R include the CA1 and CA2 regions of the hippocam­pus, raphe nuclei, and the locus coeruleus (62–65). The locus coeruleus and dorsal/median raphe nuclei are major centers for the noradrenergic and serotonergic neurons, respectively. High levels of OX1R expression in these nuclei suggest a regulatory role of orexins on the monoaminergic systems. On the other hand, OX2R mRNA is also present in basal fore-brain structures (amygdala and bed nucleus of the stria ter-minalis), linked to such functions as memory storage and attention, and the nucleus accumbens (62–65). The nucleus accumbens is the major recipient of the mesolimbic dopa-

minergic projection and serves a key role in brain reward mechanisms, which may mediate the positive reinforcing effect of food (73).

V. Orexins and the Regulation of Feeding Behavior

A. Orexins increase food intake

Sakurai et al. (1) initially examined the effects of the orexins on feeding behavior, because the mRNA for the precursor of these peptides was abundantly expressed in the LHA, a region classically implicated in the regulation of both food intake and metabolism (5-7). In fact, Anand and Brobeck (74) called the LHA a “feeding center,” since lesions of the LHA caused substantial reductions in food intake and body weight leading to starvation if the animals were not force-fed (7). The fact that electrical stimulation of the LHA produced vigorous feeding, leading to an increase in body weight (7), suggested that stimulation activated an orexigenic pathway originating within or in the vicinity of the LHA.

Sakurai et al. (1) demonstrated that intracerebroventricular administration of orexin-A or orexin-B in rats increased food intake in a dose-dependent manner, with orexin-A signifi­cantly more effective than orexin-B, possibly due to activa­tion of both OX1R and OX2R subtypes (1). Based on these findings, the orexins were named after the Greek word orexis, which means appetite (1). Subsequently, orexin-A has been reported to increase food intake in several species (8-10, 15, 75), including goldfish (76), and after microinjections in several hypothalamic nuclei, including the PVN, dorsome-dial hypothalamic nucleus, LHA, and perifornical area (11, 14, 15). In addition, Yamada et al. (12) reported a profound inhibition of natural feeding in fasted rats by central injection of an antiorexin-A antibody. Furthermore, a selective OX1R receptor antagonist, SB-334867-A, has been shown to inhibit spontaneous nighttime feeding over several days, as well as orexin-A-induced feeding, and (over the first 4 h) feeding stimulated by an overnight fast (77, 78). Thus, endogenous orexins and stimulation of the OX1R receptor appear to be necessary for normal feeding.

In contrast to orexin-A, the results obtained with orexin-B have been more variable; therefore, orexin-B has been con­cluded to have little, if any, effect on feeding (8, 10, 11, 79). The lack of apparent feeding disturbances in OX2R mutant dogs further supports this conclusion (80); however, the moderately dense expression of OX2R in the arcuate, PVN, and LHA (65) suggests that conclusions regarding the spe­cific receptor(s) involved in the feeding effects of orexins cannot be made without further study.

A comparative evaluation of the potency of orexins, ad­ministered intracerebroventricularly, with other hypotha­lamic orexigenic peptides has demonstrated that orexin-A is significantly less potent in stimulating food intake than NPY (8, 51). In addition, unlike NPY, chronic administration of orexin-A does not induce obesity in normal rats (81, 82). However, its duration of action is longer than that of NPY (1), and the magnitude of the effect of orexins is similar to that of other hypothalamic appetite-stimulating peptides, such as melanin concentrating hormone (MCH) and galanin (8). In­terestingly, MCH and galanin also do not cause obesity in 

normal rats (83, 84). However, mice with targeted deletions of the MCH gene are hypophagic (85), as are orexin knockout mice (19). Thus, orexins appear to be involved in the short-term regulation of feeding, rather than the long-term main­tenance of body weight.

Orexin-A appears to increase food intake by delaying be­havioral satiety, i.e., the normal transition from eating through grooming to resting (75). In addition, results with chronic intracerebroventricular infusion of orexin-A over several days, suggest that the peptide also disrupts the nor­mal circadian feeding pattern in rats by increasing daytime and decreasing nighttime food intake (52, 77, 79, 81). Met­abolic effects of orexins have also been shown to be depen­dent on circadian phase (86). Interestingly, disruptions of normal circadian feeding patterns are well described effects of LHA lesions (7). In addition, a role for the LHA in the regulation of sleep-wakefulness has been established (6, 87). This suggests that increased arousal or prolonged wakeful-ness may contribute to orexin-A-stimulated food intake in continuously infused rats.

Orexin neurons innervate and activate brain areas that promote wakefulness, such as the aminergic locus coeruleus, which diffusely activates the cortex, and the tuberomammil-lary nucleus (14, 15, 88). Application of orexin-A increases the firing rates of aminergic neurons, in vitro (88) and sup­presses rapid eye movement sleepin adose-dependent man­ner (89). Furthermore, orexin knockout mice exhibit a phe-notype strikingly similar to human narcolepsy (90), as do canines with a mutation of the OX2R gene (80). People with narcolepsy have chronic, sometimes severe, daytime sleep­iness that is often accompanied by episodic intrusions of rapid eye movement sleep and sudden episodes of muscular paralysis or weakness known as cataplexy. These findings leave no doubt that orexin neurons play essential roles in the control of feeding and energy balance but also regulate wake-fulness. This is of significance because, during periods of nutritional depletion, alertness may help to ensure survival. Clearly, in such circumstances, searching for food would be preferable to sleeping.

B. Orexin neurons respond to metabolic signals

Orexin neurons respond to several metabolic signals that reflect the state of energy resources. Hypothalamic prepro-orexin mRNA levels are increased significantly after 48 h of fasting and by acute (6 h) insulin-induced hypoglycemia (1, 13, 91), suggesting activation of these neurons under condi­tions of hunger. However, no changes in expression occurred when rats with acute or chronic insulin-induced hypogly-cemia were allowed to eat (91, 92) or when they were given glucose to maintain euglycemia (93). In addition, no in­creases in hypothalamic prepro-orexin mRNA levels were seen in rats with increased appetite due to insulin-deficient diabetes, or access to palatable foods (91, 94). Thus, orexin neurons are not stimulated under all conditions of hunger. Based on these findings, Cai et al. (91) concluded that low plasma glucose levels and/or absence of food from the gut stimulates orexin neurons and postulated that orexins are involved in short-term feeding behavior. Furthermore, they suggested that orexin neurons might belong to a subset of

LHA neurons that are stimulated by falls in serum glucose and inhibited by vagally transmitted satiety signals that are relayed through the NTS.

It is well known that a decline of blood glucose level can signal the initiation of food intake (95). Circulating glucose concentrations show a dip before the onset of most meals in human subjects and rodents. When the glucose dip is pre­vented, the next meal is delayed (21). Mayer’s (96) glucostatic theory of feeding postulates that eating occurs to maintain glucose availability. The LHA contains “glucosensitive” neu­rons that are activated by hypoglycemia and suppressed by elevated blood glucose, suggesting a role in short-term nu­trient sensing. Glucosensitive neurons account for approxi­mately 25% of LHA neurons (95,97); therefore, at least some of the orexin-containing neurons may be glucosensitive. Re­cent studies support this suggestion. Approximately 30% of orexin-immunoreactive neurons were shown to display Fos-like immunoreactivity, a marker of neuronal activation, dur­ing insulin-induced hypoglycemia (92, 98). In addition, Shi-raishi et al. (99) demonstrated that orexin-A caused an increase in spike discharge in about 67% of glucosensitive LHA neurons that they tested. Thus, glucosensitive cells express excitatory orexin receptors. Interestingly, orexin-A inhibited the activity of glucoresponsive neurons found in the VMH. Glucoresponsive neurons are excited by glucose, and stimulation of these cells has been postulated to con­tribute to the cessation of eating. The opposite effects of orexins on the activity of glucosensitive and glucoresponsive hypothalamic neurons are consistent with the antagonistic roles of the LHA and VMH in feeding regulation (7).

The dorsal vagal complex is another important component of orexin feeding circuits. Therefore, it is not surprising that neuronal activation of orexin neurons was accompanied by the appearance of Fos immunoreactivity in neurons in the NTS and adjacent DMV (92). The NTS relays information from vagal afferents, including glucoreceptors in the gut and liver. In addition, the NTS, like the LHA, contains glucosen­sitive cells that are stimulated by hypoglycemia (5, 55, 56). The similar pattern of Fos activity in the NTS and LHA during insulin-induced hypoglycemia led the authors to con­clude that the signals that triggered orexin neurons might be relayed via the NTS, which has a major projection to the LHA (61). Thus, the NTS may be an important regulator of orexin neurons and their responses to changes in glucose availabil­ity and prandial signals.

Orexin-containing neurons may also be sensitive to leptin. Leptin is a protein product of the ob (obese) gene (100) that is secreted by adipocytes in proportion to fat stores. Exog-enously administered leptin reduces body weight and can inhibit the increased feeding stimulated by several orexi-genic peptides (51, 101). The arcuate nucleus is a major site of leptin-responsive neurons and is considered an important “satiety center” on the basis of lesioning studies (51). Leptin-mediated inhibition of arcuate NPY neurons, and excitation of neurons that coexpress the anorectic peptides, cocaine-and amphetamine-regulated transcript and POMC, are be­lieved to underlie the suppression of appetite by leptin (15, 102).

Leptin receptor immunoreactivity and signal transducer and activator of transcription 3, a transcription factor ac

vated by leptin, are found in orexin-containing cells (103). Beck and Richy (104) showed that chronic administration of leptin reduces orexin-A levels in the LHA. In addition, Lopez et al. (105) found that leptin could inhibit the increase in orexin gene expression induced by fasting. They also found increased OX1R mRNA expression with fasting that was suppressed by leptin treatment. Interestingly, no change in OX2R mRNA expression was detected in either fasted or leptin-treated conditions. Thus, it seems likely that orexin cells are modulated by leptin, probably via OX1R, and changes in orexin expression are involved in the response to fasting.

C. Orexins and cephalic phase reflexes

The anatomical and experimental data described above clearly imply that central orexins play a role in the regulation of feeding behavior. Although several lines of evidence sug­gest that the hypothalamus is the primary brain site targeted by the orexins, the projection from the NTS to the LHA also appears to be an important regulator of orexin neurons and their response to changes in glucose availability and prandial signals (92). As such, the NTS may be involved in triggering hunger and eating in response to hypoglycemia and perhaps in terminating feeding episodes.

Orexin fibers and receptors are also found in the DMV (3, 15, 65). This raises the possibility that the orexins may control vagal outflow to the gastrointestinal tract and modulate ac­tivities such as gastric acid secretion and/or motility. Orexin-A dose-dependently increases gastric acid secretion when given centrally and with an intact vagus nerve (106). Others have found that orexin-A potently increases gastric motility when applied to the DMV (72). These findings sug­gest a role for orexins in the brain-gut axis.

In 1895, Pavlov and Schumowa-Simanowskaja (107) dem­onstrated that sensory stimulation induced by sham feeding evokes gastric acid secretion. It is now well known that gastric acid secretion occurs as part of cephalic phase re­flexes, which consist of simultaneous activation of gastric acid and pancreatic enzyme secretion, antroduodenal mo-tility, release of pancreatic polypeptide and gastrin, and gall­bladder contraction. Cephalic phase responses are provoked by the thought, sight, smell, taste, and chewing of food that has an appetizing effect in the subjects studied (108). These responses are important because they prime the secretory capability of the gut, increasing the efficiency of the subse­quent gastric and intestinal phases of secretion that occur in response to a meal. The potent stimulatory effect of orexin-A on gastric acid secretion and motility and its dependence on vagal cholinergic outflow to the stomach suggests that orex-ins may mediate cephalic phase reflexes (109). Other pep-tides that have been shown to evoke cephalic phase re­sponses include TRH (109), NPY (110), and ghrelin (111). Interestingly, like orexin, both NPY (82) and ghrelin (31, 112) stimulate feeding, and both peptides are found in the gut. Thus, orexigenic gut peptides evoke cephalic phase responses.

normal rats (83, 84). However, mice with targeted deletions of the MCH gene are hypophagic (85), as are orexin knockout mice (19). Thus, orexins appear to be involved in the short-term regulation of feeding, rather than the long-term main­tenance of body weight.

Orexin-A appears to increase food intake by delaying be­havioral satiety, i.e., the normal transition from eating through grooming to resting (75). In addition, results with chronic intracerebroventricular infusion of orexin-A over several days, suggest that the peptide also disrupts the nor­mal circadian feeding pattern in rats by increasing daytime and decreasing nighttime food intake (52, 77, 79, 81). Met­abolic effects of orexins have also been shown to be depen­dent on circadian phase (86). Interestingly, disruptions of normal circadian feeding patterns are well described effects of LHA lesions (7). In addition, a role for the LHA in the regulation of sleep-wakefulness has been established (6, 87). This suggests that increased arousal or prolonged wakeful-ness may contribute to orexin-A-stimulated food intake in continuously infused rats.

Orexin neurons innervate and activate brain areas that promote wakefulness, such as the aminergic locus coeruleus, which diffusely activates the cortex, and the tuberomammil-lary nucleus (14, 15, 88). Application of orexin-A increases the firing rates of aminergic neurons, in vitro (88) and sup­presses rapid eye movement sleepin adose-dependent man­ner (89). Furthermore, orexin knockout mice exhibit a phe-notype strikingly similar to human narcolepsy (90), as do canines with a mutation of the OX2R gene (80). People with narcolepsy have chronic, sometimes severe, daytime sleep­iness that is often accompanied by episodic intrusions of rapid eye movement sleep and sudden episodes of muscular paralysis or weakness known as cataplexy. These findings leave no doubt that orexin neurons play essential roles in the control of feeding and energy balance but also regulate wake-fulness. This is of significance because, during periods of nutritional depletion, alertness may help to ensure survival. Clearly, in such circumstances, searching for food would be preferable to sleeping.

B. Orexin neurons respond to metabolic signals

Orexin neurons respond to several metabolic signals that reflect the state of energy resources. Hypothalamic prepro-orexin mRNA levels are increased significantly after 48 h of fasting and by acute (6 h) insulin-induced hypoglycemia (1, 13, 91), suggesting activation of these neurons under condi­tions of hunger. However, no changes in expression occurred when rats with acute or chronic insulin-induced hypogly-cemia were allowed to eat (91, 92) or when they were given glucose to maintain euglycemia (93). In addition, no in­creases in hypothalamic prepro-orexin mRNA levels were seen in rats with increased appetite due to insulin-deficient diabetes, or access to palatable foods (91, 94). Thus, orexin neurons are not stimulated under all conditions of hunger. Based on these findings, Cai et al. (91) concluded that low plasma glucose levels and/or absence of food from the gut stimulates orexin neurons and postulated that orexins are involved in short-term feeding behavior. Furthermore, they suggested that orexin neurons might belong to a subset of


LHA neurons that are stimulated by falls in serum glucose and inhibited by vagally transmitted satiety signals that are relayed through the NTS.

It is well known that a decline of blood glucose level can signal the initiation of food intake (95). Circulating glucose concentrations show a dip before the onset of most meals in human subjects and rodents. When the glucose dip is pre­vented, the next meal is delayed (21). Mayer’s (96) glucostatic theory of feeding postulates that eating occurs to maintain glucose availability. The LHA contains “glucosensitive” neu­rons that are activated by hypoglycemia and suppressed by elevated blood glucose, suggesting a role in short-term nu­trient sensing. Glucosensitive neurons account for approxi­mately 25% of LHA neurons (95,97); therefore, at least some of the orexin-containing neurons may be glucosensitive. Re­cent studies support this suggestion. Approximately 30% of orexin-immunoreactive neurons were shown to display Fos-like immunoreactivity, a marker of neuronal activation, dur­ing insulin-induced hypoglycemia (92, 98). In addition, Shi-raishi et al. (99) demonstrated that orexin-A caused an increase in spike discharge in about 67% of glucosensitive LHA neurons that they tested. Thus, glucosensitive cells express excitatory orexin receptors. Interestingly, orexin-A inhibited the activity of glucoresponsive neurons found in the VMH. Glucoresponsive neurons are excited by glucose, and stimulation of these cells has been postulated to con­tribute to the cessation of eating. The opposite effects of orexins on the activity of glucosensitive and glucoresponsive hypothalamic neurons are consistent with the antagonistic roles of the LHA and VMH in feeding regulation (7).

The dorsal vagal complex is another important component of orexin feeding circuits. Therefore, it is not surprising that neuronal activation of orexin neurons was accompanied by the appearance of Fos immunoreactivity in neurons in the NTS and adjacent DMV (92). The NTS relays information from vagal afferents, including glucoreceptors in the gut and liver. In addition, the NTS, like the LHA, contains glucosen­sitive cells that are stimulated by hypoglycemia (5, 55, 56). The similar pattern of Fos activity in the NTS and LHA during insulin-induced hypoglycemia led the authors to con­clude that the signals that triggered orexin neurons might be relayed via the NTS, which has a major projection to the LHA (61). Thus, the NTS may be an important regulator of orexin neurons and their responses to changes in glucose availabil­ity and prandial signals.

Orexin-containing neurons may also be sensitive to leptin. Leptin is a protein product of the ob (obese) gene (100) that is secreted by adipocytes in proportion to fat stores. Exog-enously administered leptin reduces body weight and can inhibit the increased feeding stimulated by several orexi-genic peptides (51, 101). The arcuate nucleus is a major site of leptin-responsive neurons and is considered an important “satiety center” on the basis of lesioning studies (51). Leptin-mediated inhibition of arcuate NPY neurons, and excitation of neurons that coexpress the anorectic peptides, cocaine-and amphetamine-regulated transcript and POMC, are be­lieved to underlie the suppression of appetite by leptin (15, 102).

 


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

Categories

%d bloggers like this: