Galanin

Tamas Bartfai

Galanin is a neuropeptide which is not a member of any known family of neuropeptides, despite repeated efforts to discover related peptides. Its actions are mediated via Gi-protein-coupled receptors and ion channels, usually producing inhibition of secretion of a transmitter or hormone in the nervous and endocrine system. In many respects, these inhibitory actions of galanin remind us of those of gamma-aminobutyric acid (GABA) and of neuropeptide Y (NPY). Galanin coexists with GABA, noradrenaline, 5-hydroxytryptamine (5-HT), and NPY in several regions of the brain.

Its central nervous system (CNS) actions—and in particular its proposed role in Alzheimer's disease—have helped to fuel interest in galanin. The galanin literature is now numbering over 1200 papers. The late introduction of specific, high-affinity galanin antagonists in 1991 made it possible to study the role of endogenous galanin—which often appears as a strong tonic inhibitor of neuronal actions, a feature studied best by use of antagonists. Several excellent reviews (6, 57, 73) have appeared on different aspects of galanin actions, and these can provide more detail than this short overview. This chapter is devoted to the central actions and central neuroanatomy of galanin and galanin receptors. The literature on galanin actions on the pancreas, heart, and gut is abundant and rapidly increasing. We have, however, chosen to concentrate on the CNS actions of galanin which may underlie a role of galanin receptor ligands in therapy of Alzheimer's disease, depression, and eating disorders, respectively (see also GABA and Glycine, Coexisting Neurotransmitters in Central Noradrenergic Neurons, General Overview of Neuropeptides, Neuroendocrinology of Mood Disorders, Neuropeptide Alterations in Mood Disorders, Biological Markers in Alzheimer’s Disease, and Psychopharmacology of Anorexia Nervosa, Bulimia Nervosa, and Binge Eating).

The mRNA encoding the galanin is composed of a 5¢ portion encoding a signal sequence, followed by a Lys-Arg cleavage site, then the 29-amino-acid-long galanin peptide followed by Gly-Lys-Arg at the C-terminal containing the amide donor Gly and the cleavage site Lys-Arg (63, 72). Both the signal sequence and the galanin sequence show a very high degree of homology (>85%) between the rat, mouse, porcine, bovine, and human sequences (see Table 1). The galanin-encoding portion of the mRNA is followed by 180 bases encoding a 60-amino-acid-long peptide, named the galanin-message-associated peptide (GMAP). Several homology regions (some with homology up to 97%) of 15–20 amino acids within the GMAP in rat, mouse, bovine, porcine, and human sequences have been examined so far. The biological significance of GMAP as a neurohormone or precursor to neurohormones is not yet known.

Galanin mRNA levels rise in the pituitary during pregnancy, and they are also present in the placenta (72). During the estrous cycle of the rat, pituitary galanin mRNA levels vary 30-fold! Estrogen treatment induced changes (two- to threefold) in several hypothalamic nuclei. The pituitary galanin mRNA levels are controlled by thyroid hormones which appear to play a permissive role in galanin expression (40), an effect which may relate to galanin-mediated regulation of the secretion of thyrotropin (60). In human pituitary tumors a very high (67–100%) correlation between expression of galanin and ACTH was found (39). These tumors are also carrying galanin receptors in most cases. Galanin is found in the human adrenals, and pheochromocytoma show elevated levels of galanin-like immunoreactivity (galanin-LI) (8). In the rat PC12 pheochromocytoma cells, nerve growth factor (NGF) induces transcription of the galanin gene (40). The regulation of the galanin gene and of the tissue levels of galanin in the pituitary, as well as the regulation of circulating galanin levels, are under strong estrogen control in the female rat. The galanin mRNA is very strongly up-regulated by chronic estrogen treatment in an estrogen-induced pituitary tumor and in the female rat anterior pituitary (75). The interactions between the components of the hypothalamic–pituitary–adrenal axis involve and affect the expression of galanin, but nowhere as dramatically as in the anterior pituitary (60, 75).

Regulation of the galanin gene expression under conditions of axotomy of sensory neurons has also been observed (39). The increase in galanin-LI and galanin mRNA is dramatic (up to 100-fold). It appears that destruction of the axons and/or terminal fields of central galaninergic neurons also leads to a similar up-regulation of the galanin mRNA and galanin-LI in the cell bodies (17).

In the rat, the highest densities of galanin-LI are found in more ventral structures such as the amygdaloid complex, the hypothalamus, and the brainstem (12, 56, 66, 67).

In contrast, neocortical areas and cerebellum have considerably lower numbers of galanin-LI-positive fibers and cells (54). Galanin-immunoreactive cell bodies are observed in the telencephalon, including the bed nucleus of the stria terminalis, the nucleus of the diagonal band continuing into the medial septum, and the medial aspects of the central amygdaloid nucleus. In the diencephalon the only thalamic positive cell bodies are seen in the anterior dorsal and periventricular nuclei. The hypothalamus contains a large number of galanin-positive cell groups such as the medial and lateral preoptic nuclei, the arcuate nucleus, the periventricular nucleus, the dorsomedial nucleus, the lateral hypothalamus/medial forebrain bundle area, the tuberal, caudal, accessory supraoptic and paraventricular magnocellular nuclei, and the area lateral to the mamillary recess. In the mesencephalon/pons, large midline cell groups that are galanin-positive include the dorsal raphe nuclei. Many neurons in the locus coeruleus and dorsal raphe nucleus contain galanin-LI.

Galanin is present in neurons which express classical transmitters such as catecholamines and amino acids as well as other peptides (53, 55). Many such coexistence situations have been identified for galanin. In the ventral forebrain, galanin is present in a population of the large cholinergic neurons projecting to cortical areas (43, 63). These cells are present in the diagonal band nucleus extending into the medial septum but not in the basal nucleus of Meynert. In the rat they project especially to the hippocampal formation.

The hypothalamus is particularly rich in coexistence situations: In the arcuate nucleus, galanin coexists with tyrosine hydroxylase as well as with the GABA-ergic marker glutamic acid decarboxylase (GAD) (55), and with neurotensin-LI, growth-hormone-releasing-factor (GRF)-LI, and choline acetyltransferase-LI (25). The main signal substances in the latter neurons is presumably GRF, which is the stimulating factor for growth hormone release. A large proportion of the magnocellular and many parvocellular neurons in the paraventricular nucleus have galanin, which here coexists mainly with vasopressin-, dynorphin-, and cholecystokinin (CCK)-LI. The tuberomamillary nucleus has many galanin-positive neurons which also in part contain GAD-, histamine-, adenosineaminase-, and monoaminoxidase-LI, and these neurons also have the property of 5-hydroxytryptophan uptake (42). More recently the coexistence of galanin and luteinizing-hormone-releasing hormone (LHRH) has been demonstrated. In the lower brainstem a large proportion of the noradrenergic cell bodies of the locus coeruleus are strongly galanin-positive (37).

Coexistence is shown between galanin and 5-HT, both at the dorsal raphe nuclei and at the medullary raphe nuclei. Furthermore, nucleus raphe pallidus, obscurus, and magnus contain GAL-LI.

Several earlier studies addressed the distribution of galanin-LI in primate brain. Kordower et al. (41) have published a comprehensive distributional study on adult Cebus monkey, baboon, and human brains, demonstrating in the monkey that telencephalon galanin-positive cell bodies apparently have a wider spread than in the rat. Positive cell bodies are found in the anterior olfactory nucleus, the basal forebrain, the endopiriform nucleus, the hippocampus, and the bed nucleus of the stria terminalis. The caudate nucleus and putamen contained galanin-positive cell bodies, a location which so far has not been described conclusively in the rat. As in the rat, numerous hypothalamic nuclei were galanin-positive, including those in the medial preoptic area, the periventricular, suprachiasmatic, paraventricular, and arcuate areas, and the lateral hypothalamic area. No cell bodies were observed in the thalamus or mesencephalon, but small numbers of fibers could be seen in various nuclei, including the ventral tegmental area. In the lower brainstem the medial vestibular nucleus, nucleus prepositus, and the solitary tract and hypoglossal nuclei contained galanin-positive cell bodies. With regard to fibers, dense networks were observed in the spinal trigeminal nucleus, the solitary tract nucleus, the dorsal vagal motor nucleus, and the dorsal horn of the spinal cord. Of particular interest is the distribution of galanin-LI in the basal forebrain, in part because of its possible relation to Alzheimer's disease and because of apparent species differences. While galanin in the rat is restricted to the septal diagonal band complex, these cells have an apparently wider distribution in the monkey basal forebrain where also the nucleus basalis is included. So far it has not been possible to demonstrate galanin-LI or galanin mRNA in the human magnocellular basal forebrain, with the exception of some medium-sized neurons described by Chan-Palay (13, 14).

The sequence of galanin from six species is known, including the human galanin sequence (Table 1). The galanin peptides show a complete sequence homology in the N-terminal 15 amino acids, suggesting that this portion of the peptide is of importance for recognition by the galanin receptors and for exerting agonist activity. Furthermore, the 100% homology of the N-terminal half of the peptide suggests that galanin from any species will be recognized by galanin receptors in any other species (at least among these six known species), which is indeed the experimental observation. Despite some sequence variability in the C-terminal 16-29/30 portion of the peptide, the galanin molecules bind with the same high affinity (Kd = 0.8 nM) to receptors in any species tested so far (see ref. 45).

It is noteworthy that human galanin consists of 30 amino acids (24) and that the C-terminus is a free carboxylic acid, whereas all other known types of galanin are composed of 29 amino acids (69) and carry a C-terminal amide (i.e., the preprogalanin has 30 amino acids, but the 30th glycine is the amide donor).

The amino acid residues of importance for binding and for biological activity were studied most extensively in three preparations: rat hippocampus and hypothalamus and mouse pancreatic islets or in the Rin m5F rat insulinoma cells.

In the hippocampal test system, intracerebroventricularly (i.c.v.) injected galanin inhibits the hippocampal acetylcholine release measured by microdialysis technique (28). In the ventral hippocampus, rat galanin appears to be the most potent known endogenous inhibitor of the evoked acetylcholine release (experimentally induced by intraperitoneal injection of scopolamine 2 mg/kg). Galanin injected i.c.v. or into the paraventricular nucleus (PVN) of the hypothalamus causes a robust stimulation of food intake. In the mouse pancreatic islets, galanin potently inhibits the glucose-induced insulin release; and in Rin m5F cells, which possess an exceptionally high number of galanin receptors (2, 4, 31, 35) (~4000–20,000/cell), galanin also inhibits forskolin-stimulated cAMP accumulation. Thus in these systems the binding and biological effects of galanin, galanin analogues, and fragments were evaluated by two research groups (2, 30), with the following results:

The agonist action of galanin and of its N-terminal fragments is terminated by their degradation at the N-terminus by enzymes which cleave the first or the first two residues, with G-W yielding the inactive 3-29/30 and 3-16 fragments—as studied in the spinal cord and hypothalamus (44; Bedecs et al., in preparation). Preliminary data suggest that these diaminoacylpeptidases which cleave and inactivate galanin may belong to the class of metalloproteases because chelating agents afford some inhibition of the degradation of galanin by cerebrospinal fluid (CSF) and by spinal cord membranes.

The biological half-life of galanin upon incubation with rat CSF or with hypothalamic membranes is about 60 min, which is almost 10-fold longer than that of the equal-sized peptide vasoactive intestinal peptide. The long half-life may enable galanin to exert paracrine actions.

Galanin agonists with peptidase-resistant N-terminus are currently being synthesized and tested.

Among the galanin receptor agonists, one should mention two endogenously occurring, minor galanin isoforms isolated in small amounts (~1% of the amount of galanin-LI) from the porcine adrenal: galanin(7-29) and galanin(9-29) (8, 11). Whether or not these N-terminally elongated forms of galanin occur in human tissues is not known. In the rat spinal cord these N-terminally elongated forms of galanin bind with a 100-fold lower affinity than galanin and have commensurately weak agonist effect on the spinal flexor reflex (Wiesenfeldt et al., in preparation).

Recently a series of galanin receptor antagonists have been synthesized and tested in a variety of in vivo and in vitro models of galanin action (6).

The first galanin receptor antagonists are chimeric, bireceptor recognizing peptides composed of the N-terminal 1-12 fragment of galanin followed by proline13 and a sequence corresponding to the recognition sequence of another neuropeptide whose recognition and activity are dependent on its C-terminus (5). Thus M 15 or galantide is galanin(1-12)-pro-substance P(5-11) (5, 49), M 32 is galanin(1-12)-pro-neuropeptide Y(25-36), M 35 is galanin(1-12)-pro-bradykinin(2-9) (36, 79), and C 7 carries a substance P antagonist (i.e., spantide)—rather than an agonist—as its C-terminus: galanin(1-12)-pro-spantide. Finally M 40 and its analogues carry a C-terminal sequence with no known receptor: galanin(1-12)-(Pro)3(Ala-Leu)2Ala amide (19).

The key to the usefulness of these antagonists lies in their very high affinity (10-10–10-9 M) (6), making possible their use in vivo. The use of these galanin receptor antagonists, barely a year after their introduction, is widespread and will lead to definition of even more effects of endogenous galanin in the nervous and endocrine system. All of these antagonists have been tested in numerous models of galanin actions in the periphery and CNS (see Table 2). Their action in most systems is that of an antagonist. However, M 40 appears to distinguish between pancreatic and CNS galanin receptors (cf. receptor subtypes).

The major drawbacks of these antagonists are (a) their peptide nature which prevents their passage through blood–brain barrier, (b) their possible peptidase sensitivity, and (c) that when applied in very high concentration they may exhibit agonist-like effects due to their intact N-terminus. Because the galanin receptor antagonists hold considerable pharmacological potential, several large-scale screening programs are underway to identify nonpeptide galanin receptor antagonists.

The galanin receptor(s) has not yet been sequenced despite purification (15), solubilization, cross-linking (2), and cDNA cloning efforts. We know that it belongs to the class of G-protein-coupled receptors, and in particular to the Gi/Go-protein-coupled receptor subclass, because all actions of galanin at membrane (3), cellular (21), and in vivo levels in the hippocampus of freely moving rats (16) were abolished by pretreatment with pertussis toxin. Thus the receptor is assumed to be coupled via a pertussis toxin ADP-ribosylable G protein. There have been suggestions that the inhibitory action of galanin on the release of numerous signal substances would be exerted at a G protein directly involved in the release apparatus (65).

In the absence of sequence data on different galanin receptor proteins, definitions of galanin receptor subtypes have been proposed based on pharmacological studies with agonists and antagonists. The pancreatic and the CNS galanin receptors appear to accept the N-terminal 1-15/16 fragment as a full agonist (18, 27, 77), while it is suggested that smooth muscle galanin receptors have a structural requirement for both the N- and the C-termini of galanin (62). In addition, it has been suggested that the anterior pituitary of the rat expresses a specific galanin receptor which recognizes galanin(3-29) (76), whereas all other galanin receptors have an absolute requirement for the intact N-terminus of galanin.

Classification of galanin receptor subtypes proposed by the present authors is based on the observation that the antagonist M 40 blocks the galanin effects on feeding and also blocks acetylcholine release in the hippocampus but does not antagonize the galaninergic inhibition of the glucose-induced insulin release, suggesting that pancreatic galanin receptors may be different from CNS galanin receptors (Bartfai et al., in preparation).

Galanin has been reported to hyperpolarize neurons in a number of central and peripheral preparations. This hyperpolarization is brought about in some systems by opening of K+ channels. This has been demonstrated in pancreatic b cells (20, 21), as well as in noradrenergic cells in the locus coeruleus. In the Rin m5F rat insulinoma cells the galanin-mediated hyperpolarization appears to involve opening of an ATP-sensitive K+ channel (20).

Galanin is a potent inhibitor of the release of a number of neurotransmitters and hormones, including acetylcholine, dopamine, insulin, and gastrin. Its actions in some systems have been shown to involve a decrease in the cytosolic Ca2+ concentration (1, 58). This may be the consequence of the hyperpolarization brought about by opening of galanin-receptor-coupled K+-channels (see above), or it may be a result of the galanin-receptor-mediated closure of some Ca2+ channels—or of a combination of galanin effects on K+-channel opening and Ca2+-channel closure. In the hippocampus, pharmacological experiments using w-conotoxin suggested that closure of an N-type Ca2+ channel is part of the galanin action (61)). In the Rin m5F cells, pharmacological and electrophysiological (patch clamp) experiments suggest that the involvement of an L-type Ca2+ channel in the inhibitory galanin effect (38) on insulin release.

Galanin has been shown to inhibit the b-noradrenergic stimulation of adenylate cyclase in the rat cerebral cortex (59), and forskolin stimulated the accumulation of cAMP in the Rin m5F insulinoma cells (3). These effects could be abolished by pretreatment with pertussis toxin, indicating that the galanin receptor acts via a pertussis-toxin-sensitive G protein.

In tissue slices from the lumbar spinal cord the K+-depolarization-mediated rise in cGMP levels is partially inhibited by galanin in a dose-dependent manner (10). This observation is in line with the restriction of Ca2+ entry by galanin, because Ca2+ entry is a prerequisite for the full activity of the guanylate cyclase.

Galanin also diminishes the muscarinic-cholinergic-receptor-mediated stimulation of phospholipase C, resulting in less inositol triphosphate (IP3) production in the presence of galanin than by the muscarinic agonist alone (61). This interaction probably reflects a postsynaptic antagonism between acetylcholine and galanin released from the same septal afferent and acting probably on the same hippocampal cells, because it was previously shown that muscarinic stimulation of phospholipase C is a predominantly postsynaptic effect in the hippocampus.

The effects of galanin on nitric oxide synthesis have not yet been studied.

The phorbol-ester-stimulated (and therefore probably protein kinase-C-catalyzed) phosphorylation of two hippocampal proteins of 20 and 41 kD is inhibited in the presence of galanin (47). The galanin effect could be blocked by galanin receptor antagonists M 15 and M 35. The identity of the two proteins whose phosphorylation is inhibited is not yet known.

Galanin hyperpolarizes the noradrenergic neurons in the locus coeruleus and reduces the rate of spontaneous firing (5, 64). These effects are probably brought about by opening some non-ATP-sensitive K+ channels. The hyperpolarizing effects of galanin are long-lasting, dose-dependent, and fully reversible. Specific galanin receptor antagonists block this effect of galanin (5). The galaninergic control of firing rate is particularly interesting because locus coeruleus neurons contain large amounts of galanin in coexistence with noradrenaline. It is possible that dendritic release of galanin occurs within the structure and contributes to the overall regulation of noradrenergic activity in the CNS (see also Physiological and Anatomical Determinants of Locus Coeruleus Discharge: Behavior and Clinical Implications).

Galanin inhibits the slow muscarinic postsynaptic potential, especially in the hippocampus (23), by a presynaptic inhibition of the release of acetylcholine from the septal afferents, as demonstrated also by in vivo microdialysis studies (26). Galanin also inhibits the evoked release of the excitatory amino acids glutamate and asparate, while not affecting the release of GABA under the same conditions (78). By virtue of its pre- and postsynaptic effects on K+ and Ca2+ conductances, galanin is a hyperpolarizing agent in the hippocampus; galanin receptor antagonists accordingly lower the seizure threshold for picrotoxin-induced seizures, for example.

Galanin appears to affect the binding affinity of serotonin to serotonin receptors (5-HTIA) in the diencephalic and telencephalic areas. Galanin applied i.c.v. strongly reduces 5-HT metabolism in the hippocampus and frontoparietal cortex of the rat (29).

Galanin inhibits dopamine release from the median eminence and strongly stimulates the release of prolactin and growth hormone in humans and in rats (7, 9, 32, 33). In humans, intravenous injections of galanin cause substantial release of growth hormone, while reduction of the plasma levels of insulin, glucagon, somatostatin, and gastrin is also observed (7, 9, 32, 33). The galanin receptor antagonist M 15 has been shown to inhibit the amplitude of growth hormone release while not affecting the periodicity of the pulsatile release (30). The coexistence of dopamine, galanin, and GRF-LIs in hypothalamic nerve endings and neurons (53) may thus give rise to complex inhibitory and disinhibitory patterns in control of growth hormone and prolactin release.

Galanin, both applied i.c.v. and when injected into the PVN (18, 42) in a dose-dependent manner, stimulates feeding behavior—in particular the intake of fat. The increase induced by galanin is highly significant: 200% above baseline. Several galanin receptor antagonists have been shown to antagonize the stimulatory effects of galanin (48).

It should be noted that among the prominent peripheral actions of galanin is its inhibitory effect on the glucose-induced insulin release, making galanin a hyperglycemic substance (52). The present assumption is that galanin—which coexists with noradrenaline in the sympathetic nerves which innervate the pancreas—mediates the stress-induced inhibition of insulin release (22). Galanin appears to regulate both fat and glucose levels by its central and peripheral actions.

Galanin appears to act as a physiological antagonist of substance P in several types of pain, and intrathecal application of galanin in a dose-dependent manner reduces pain sensation (74). Intrathecal application of galanin receptor antagonists to rats with only superficial cuts caused a pronounced autotomy, demonstrating the important analgesic effects of endogenous galanin in peripheral nerve injury (70). These observations are in line with the exceptionally high increases in galanin and galanin mRNA in dorsal root ganglia upon axotomy (71) and suggest that one of the functions of galanin at nerve injury is to control pain.

Galanin (i.c.v.) impairs the performance of rats in the Morris swim maze (68) and in a one trial reward learning task (50). Galanin impairs working memory when injected into medial septal area (34). Furthermore, galanin antagonizes the acetylcholine-injection (i.c.v.)-produced improvements in cognitive performance of basal forebrain-lesioned rats (51). Galanin antagonist M 35, on the other hand, has been shown to improve performance of rats in the Morris swim maze (79). This improvement probably involves enhanced acquisition and, to a lesser extent, improved retention.

Galanin has numerous peripheral and central effects which are well worth exploiting. Accordingly, several groups in the industry as well as in the academic community work on specific, high-affinity, nonpeptide agonists and antagonists which will act as galanin receptors. Development of receptor-subtype-specific ligands will even further increase the usefulness of the galanin receptors as pharmacological targets.

It is argued by several groups that galanin receptor agonists may be useful in preventing anoxic damage because these ligands do not suppress GABA release or the release of the excitatory amino acids glutamate and aspartate (78). Thus the galanin receptor agonists would have an advantage over Ca2+-channel blockers, which, in general, suppress release. It is envisaged that under the conditions of open heart surgery (for example), galanin receptor agonists would be useful in preventing oxidative damage.

Galanin receptor agonists enhance the release of growth hormone without affecting the diurnal rhythm of this process and therefore appear as attractive agents to increase growth hormone secretion in humans (7, 9, 32, 33).

Galanin receptors seem to control prolactin release from pituitary adenomas and are considered as targets for endocrine manipulation of these tumors.

Galanin, although not a potent analgesic agent on its own, appears to prolong the morphine analgesia four- to eightfold. Application of CCKb antagonists together with morphine and galanin produced remarkable prolongation of the morphine effects and may therefore contribute to reduction of morphine doses needed to manage chronic pain (75).

Immunohistochemical data on autopsy samples from Alzheimer's-afflicted brains suggest that galaninergic fibers hyperinnervate the nucleus basalis cholinergic cells (13, 14). This may have a consequence similar to that found in the rat locus coeruleus—that is, that the cells are hyperpolarized by galanin and lower their firing rate (64). This would further deepen the hippocampal cholinergic deficit arising from the death of many cholinergic cells in this nucleus. Thus a galanin receptor antagonist may improve cholinergic function by enhancing the firing rate of surviving cholinergic neurons.

In the hippocampus, at the level of the cholinergic nerve endings it was shown by in vivo microdialysis that galanin suppresses the per-pulse release of acetylcholine in the ventral hippocampus of rats (28). Galanin receptor antagonists reverse this presynaptic inhibition (5). Thus galanin receptor antagonists may improve cholinergic function both at the level of the cell body by enhancing firing and at the cholinergic nerve terminal by enhancing the per-pulse release.

At the level of the hippocampal cholinoreceptive cell, galanin also opposes the slow depolarizing action of acetylcholine at muscarinic receptors because it reduces the muscarinic stimulation of IP3 production (61); thus, even at these cholinoreceptive cells, galanin receptor antagonists may contribute to the overall enhancement of cholinergic transmission.

Experiments with normal rats receiving galanin receptor antagonist M 35 have already shown an enhanced cognitive function (79). These studies are now being extended to old rats and to lesion models of Alzheimer's disease. Galanin receptor antagonists today appear as one of the more interesting new approaches to enhance cholinergic function (see Biological Markers in Alzheimer’s Disease).

Galanin is colocalized with both serotonin in the raphe nucleus and with noradrenaline in the locus coeruleus. In the latter structure, sufficient amount of data have been collected to suggest that galanin receptor antagonists enhance firing of noradrenergic neurons probably by reversing the tonic galanin-mediated inhibition/hyperpolarization (5). In addition, it is known that galanin also suppresses 5-HT metabolism (29). The strategic localization of both galanin and galanin receptors in the raphe nucleus and in the locus coeruleus, along with the neuronal-activity-enhancing effects of galanin receptor antagonists in the monoamine systems, argues for their potential as antidepressant agents (see Neuroendocrinology of Mood Disorders and Neuropeptide Alterations in Mood Disorders).

Because galanin antagonists suppress fat intake specifically (19, 48 but do not suppress all kinds of food intake, these compounds are interesting agents for development of drugs in the area of eating disorders.

This chapter summarizes the work carried out in collaboration with the following individuals: Dr. Thomas Hökfelt, Karolinska Institute; Dr. Sylvana Consolo, Mario Negri Institute; Dr. Jacqueline Crawley, NIMH, Dr. Zsuzsanna Wisenfeld-Hallin, Huddinge Hospital; and Dr. Ülo Langel, Stockholm University. The study was supported by the National Institute of Aging, Drug Discovery Programme, Swedish Medical Research Council, and Riksbankens Jubileumsfond.

published 2000