Clifford B. Saper

Narcolepsy in orexin Knockout Mice: Molecular Genetics of Sleep Regulation

Neurons containing the neuropeptide orexin (hypocretin) are located exclusively in the lateral hypothalamus and send axons to numerous regions throughout the central nervous system, including the major nuclei implicated in sleep regulation. Here, we report that, by behavioral and electroencephalographic criteria, orexin knockout mice exhibit a phenotype strikingly similar to human narcolepsy patients, as well as canarc-1 mutant dogs, the only known monogenic model of narcolepsy. Moreover, modafinil, an anti-narcoleptic drug with ill-defined mechanisms of action, activates orexin-containing neurons. We propose that orexin regulates sleep/wakefulness states, and that orexin knockout mice are a model of human narcolepsy, a disorder characterized primarily by rapid eye movement (REM) sleep dysregulation.

Hypothalamic regulation of sleep and circadian rhythms

A series of findings over the past decade has begun to identify the brain circuitry and neurotransmitters that regulate our daily cycles of sleep and wakefulness. The latter depends on a network of cell groups that activate the thalamus and the cerebral cortex. A key switch in the hypothalamus shuts off this arousal system during sleep. Other hypothalamic neurons stabilize the switch, and their absence results in inappropriate switching of behavioural states, such as occurs in narcolepsy. These findings explain how various drugs affect sleep and wakefulness, and provide the basis for a wide range of environmental influences to shape wake–sleep cycles into the optimal pattern for survival.

The sleep switch: hypothalamic control of sleep and wakefulness

More than 70 years ago, von Economo predicted a wake-promoting area in the posterior hypothalamus and a sleep-promoting region in the preoptic area. Recent studies have dramatically confirmed these predictions. The ventrolateral preoptic nucleus contains GABAergic and galaninergic neurons that are active during sleep and are necessary for normal sleep. The posterior lateral hypothalamus contains orexin/hypocretin neurons that are crucial for maintaining normal wakefulness. A model is proposed in which wake- and sleep-promoting neurons inhibit each other, which results in stable wakefulness and sleep. Disruption of wake- or sleep-promoting pathways results in behavioral state instability.

DIFFERENTIAL EXPRESSION OF OREXIN RECEPTORS 1 AND 2 IN THE RAT BRAIN

Orexins (hypocretins) are neuropeptides synthesized in the central nervous system exclusively by neurons of the lateral hypothalamus. Orexin‐containing neurons have widespread projections and have been implicated in complex physiological functions including feeding behavior, sleep states, neuroendocrine function, and autonomic control. Two orexin receptors (OX1R and OX2R) have been identified, with distinct expression patterns throughout the brain, but a systematic examination of orexin receptor expression in the brain has not appeared. We used in situ hybridization histochemistry to examine the patterns of expression of mRNA for both orexin receptors throughout the brain. OX1R mRNA was observed in many brain regions including the prefrontal and infralimbic cortex, hippocampus, paraventricular thalamic nucleus, ventromedial hypothalamic nucleus, dorsal raphe nucleus, and locus coeruleus. OX2R mRNA was prominent in a complementary distribution including the cerebral cortex, septal nuclei, hippocampus, medial thalamic groups, raphe nuclei, and many hypothalamic nuclei including the tuberomammillary nucleus, dorsomedial nucleus, paraventricular nucleus, and ventral premammillary nucleus. The differential distribution of orexin receptors is consistent with the proposed multifaceted roles of orexin in regulating homeostasis and may explain the unique role of the OX2R receptor in regulating sleep state stability. J. Comp. Neurol. 435:6–25, 2001.

From lesions to leptin: hypothalamic control of food intake and body weight.

By 1940, Hetherington and Ranson had laid to rest most doubts regarding the importance of the hypothalamus in regulating body weight. However, their electrolytic lesions were inherently crude and started debate about which specific hypothalamic cell groups are critical in the control of feeding and body weight. Despite intense activity, during the 45 years that followed the discovery of the ventromedial nucleus syndrome, we learned relatively little about the actual pathways in the hypothalamus that mediate feeding. Because of the similarity of the leptin deficiency syndrome to the ventromedial nucleus syndrome, the discovery of leptin and its receptors revitalized the field. In fact, the original observations of Hetherington and Ranson, implicating the “ventromedial nucleus, but also including the adjacent arcuate nucleus and parts of the dorsomedial and ventral premammillary nuclei” in the regulation of feeding, have proven to be both prescient and surprisingly accurate in identifying the hypothalamic cell groups with the highest levels of long-form leptin receptors.In the 4 years since the discovery of leptin, there has been remarkable progress in studying the effects of starvation, leptin deprivation, and leptin administration on the expression of genes for transcription factors (c-fos), signaling molecules (SOCS-3), and neurotransmitters (neuropeptide Y, agouti-related protein, α-melanocyte stimulating hormone, CART, MCH, and ORX) involved in feeding. The availability of these molecular tools, coupled with tract tracing, has resulted in striking progress in dissecting an extensive network of hypothalamic circuitry that regulates feeding (Figure 8Figure 8). Despite the fact that a number of pieces of this puzzle are still missing, the outline of the hypothalamic system for regulation of feeding is now more clear. In retrospect, the ventromedial nucleus itself is only a part of this network. It sits, however, at the epicenter of a web of pathways, running between the arcuate nucleus and the paraventricular nucleus and lateral hypothalamus, and this epicenter defines the mechanisms of neuronal regulation of feeding.‡To whom correspondence should be addressed (e-mail: csaper@caregroup.harvard.edu).Figure 8A Schematic Drawing Summarizing Some of the Major Pathways and Neurotransmitters that Have Been Implicated in the Regulation of FeedingPathways that are activated by leptin (and therefore presumably have an anorexic influence) are illustrated in red, whereas those that are inhibited by leptin (and are presumed to have a phagic influence) are in green. Circadian influences are illustrated by dashed blue pathways. Leptin is proposed to exert its effects on the hypothalamus by entering through the median eminence. Note that a lesion centered on the ventromedial nucleus (VMH) and arcuate nucleus (ARC), such as that shown in Figure 1Figure 1, would eliminate leptin influence, resulting in hyperphagia and obesity. A lesion in the lateral hypothalamus (as in Figure 1Figure 1) that destroys the MCH and ORX cells, which promote feeding, would result in aphagia and inanition. Other abbreviations are the same as in Figure 1, Figure 4, Figure 6, Figure 7.View Large Image | View Hi-Res Image | Download PowerPoint Slide

Efferent connections of the parabrachial nucleus in the rat

The efferent connections of the parabrachial nucleus have been analyzed in the rat using the anterograde autoradiographic method. Fibers originating from the lateral parabrachial nucleus (PBl) ascend in the periventricular system, the dorsal tegmental bundle and the central tegmental tract. The PBl projects to the dorsal raphe nucleus, the superior central raphe nucleus, and the Edinger-Westphal nucleus. It also innervates the intralaminar (centromedian, centrolateral, paracentral, parafascicular), the midline (paraventricular, reuniens), and the ventromedial basal (VMb) thalamic nuclei as well as much of the hypothalamus, including the dorsomedial, the ventromedial, the arcuate and the paraventricular nuclei, the lateral hypothalamic and the lateral preoptic areas. The PBl sends fibers via the ansa peduncularis into the amygdala, innervating the anterior, the central, the medial, the basomedial, and the posterior basolateral nuclei. In addition, it projects to the lateral part of the bed nucleus of the stria terminalis. Descending PBl fibers travel mainly through the ventrolateral medulla, passing through the region of the A1 and A5 catecholamine cell groups, the ventrolateral reticular formation and the region that contains parasympathetic preganglionic neurons. A small component travels in Probsts bundle to the ventral part of the nucleus of the solitary tract. Only a few PBl axons continue caudally into the lateral funiculus of the spinal cord, but these could not be followed beyond the first few cervical segments. The projections of the medial parabrachial nucleus (PBm) are similar to those of PBl, but two major differences have been noted. One difference is that the PBm provides a direct input to 4 regions of cerebral cortex: (1) the granular insular cortex; (2) the deep layers of the frontal cortex; (3) the septo-olfactory area; and (4) the infralimbic cortex. The other difference is that unlike the PBl, the PBm appears to provide almost no input to the medial hypothalamic nuclei (dorsomedial, ventromedial, arcuate nuclei) nor to the medial amygdaloid nucleus. The PBm projects heavily to the nucleus ambiguus and there was no evidence for an input to the nucleus of the solitary tract. The projections of the Kölliker-Fuse nucleus (KF) are distinct from those of either PBm or PBl. The KF projects via the central tegmental tract to the lateral hypothalamic area, the lateral preoptic area, and the central nucleus of the amygdala. The contralateral projection to the zona incerta, the lateral hypothalamic area, and the lateral preoptic areas is more prominent than the ipsilateral projections. Descending KF fibers travel mainly through the ventrolateral medullary reticular formation passing through regions which give rise to parasympathetic preganglionic fibers of the VIIth, IXth and Xth cranial nerves and the A1 and A5 catecholamine cell groups. In one experiment, fibers could be followed to the intermediolateral cell column of the upper thoracic spinal cord.

Distributions of leptin receptor mRNA isoforms in the rat brain

Leptin, secreted by white adipocytes, has profound feeding, metabolic, and neuroendocrine effects. Leptin acts on the brain, but the specific anatomic sites and pathways responsible for mediating these effects are still unclear. We have systematically examined distributions of mRNA of leptin receptor isoforms in the rat brain by using a probe specific for the long form and a probe recognizing all known forms of the leptin receptor. The mRNA for the long form of the receptor (OB‐Rb) localized to selected nuclear groups in the rat brain. Within the hypothalamus, dense hybridization was observed in the arcuate, dorsomedial, ventromedial, and ventral premamillary nuclei. Within the dorsomedial nucleus, particularly intense hybridization was observed in the caudal regions of the nucleus ventral to the compact formation. Receptors were preferentially localized to the dorsomedial division of the ventromedial nucleus. Hybridization accumulated throughout the arcuate nucleus, extending from the retrochiasmatic region to the posterior periventricular region. Moderate hybridization was observed in the periventricular hypothalamic nucleus, lateral hypothalamic area, medial mammillary nucleus, posterior hypothalamic nucleus, nucleus of the lateral olfactory tract, and within substantia nigra pars compacta. Several thalamic nuclei were also found to contain dense hybridization. These groups included the mediodorsal, ventral anterior, ventral medial, submedial, ventral posterior, and lateral dorsal thalamic nuclei. Hybridization was also observed in the medial and lateral geniculate nuclei. Intense hybridization was observed in the Purkinje and granular cell layers of the cerebellum. A probe recognizing all known forms of the leptin receptor hybridized to all of these sites within the brain. In addition, intense hybridization was observed in the choroid plexus, meninges, and also surrounding blood vessels. These findings indicate that circulating leptin may act through hypothalamic nuclear groups involved in regulating feeding, body weight, and neuroendocrine function. The localization of leptin receptor mRNA in extrahypothalamic sites in the thalamus and cerebellum suggests that leptin may act on specific sensory and motor systems. Leptin receptors localized in nonneuronal cells in the meninges, choroid plexus, and blood vessels may be involved in transport of leptin into the brain and in the clearance of leptin from the cerebrospinal fluid. J. Comp. Neurol. 395:535–547, 1998.

Subnuclear organization of the efferent connections of the parabrachial nucleus in the rat

In summary, we have demonstrated the subnuclear organization of PB, and correlated this with the origins of its efferent connections. In general, PBm projects primarily to the insular, infralimbic and lateral frontal cortex, and to associated areas in the thalamus, hypothalamus and amygdala. PBl chiefly innervates the autonomic nuclei of the hypothalamus and related portions of the amygdala and the bed nucleus of the stria terminalis. KF is the main source of descending projections from PB to the region of the nucleus of the solitary tract, the ventrolateral medulla and the intermediolateral cell column in the thoracic spinal cord. Further subnuclear organization of the origins of these projections within the major PB subdivisions has been described in detail. While PB afferents tend to terminate in specific subnuclei, one cannot reliably predict from the functional properties of the major inputs to a subnucleus what information will be carried in its efferents. Further anatomical and physiological studies of the input-output relationships of single PB neurons will be necessary to help resolve this enigma. However, recent immunohistochemical observations suggest that the subnuclear organization of PB afferent and efferent connections may reflect, at least in part, their biochemical specificity.

The Need to Feed: Homeostatic and Hedonic Control of Eating

Feeding provides substrate for energy metabolism, which is vital to the survival of every living animal and therefore is subject to intense regulation by brain homeostatic and hedonic systems. Over the last decade, our understanding of the circuits and molecules involved in this process has changed dramatically, in large part due to the availability of animal models with genetic lesions. In this review, we examine the role played in homeostatic regulation of feeding by systemic mediators such as leptin and ghrelin, which act on brain systems utilizing neuropeptide Y, agouti-related peptide, melanocortins, orexins, and melanin concentrating hormone, among other mediators. We also examine the mechanisms for taste and reward systems that provide food with its intrinsically reinforcing properties and explore the links between the homeostatic and hedonic systems that ensure intake of adequate nutrition.

Activation of Ventrolateral Preoptic Neurons During Sleep

The rostral hypothalamus and adjacent basal forebrain participate in the generation of sleep, but the neuronal circuitry involved in this process remains poorly characterized. Immunocytochemistry was used to identify the FOS protein, an immediate-early gene product, in a group of ventrolateral preoptic neurons that is specifically activated during sleep. The retrograde tracer cholera toxin B, in combination with FOS immunocytochemistry, was used to show that sleep-activated ventrolateral preoptic neurons innervate the tuberomammillary nucleus, a posterior hypothalamic cell group thought to participate in the modulation of arousal. This monosynaptic pathway in the hypothalamus may play a key role in determining sleep-wake states.