|Additional related information may be found at:|
|Neuropsychopharmacology: The Fifth Generation of Progress|
Neuropeptide Y and Related Peptides
Neuropeptide Y and Related Peptides
Claes Wahlestedt and Markus Heilig
THE NPY FAMILY OF PEPTIDES
NPY and Related Peptides
Avian pancreatic polypeptide (PP) was initially discovered as a by-product of insulin isolation (41), and was subsequently isolated from many other species. Although perhaps of little physiological significance, PP served as a predecessor of neuropeptide Y (NPY) and peptide YY (PYY). Using a technique to isolate peptides with C-terminal a-amide groups, Tatemoto and Mutt (62) reported that porcine brain and gut contained large amounts of a peptide resembling PP. The PP-like peptide isolated from gut was named peptide YY (PYY) because of its N- and C-terminal tyrosine (Y being the abbreviation for tyrosine in the single-letter amino acid code). At first PYY (61) was also thought to be the PP-like peptide in the brain but subsequent work showed the (predominant) brain peptide to differ from PYY, and because of its occurrence in brain, it was named neuropeptide Y (NPY) (63). Finally, a fourth nonmammalian 37-amino-acid, nonamidated member of the peptide family, pancreatic peptide Y (PY), was found in fish which showed 64% sequence identity to both NPY and PYY (5).
NPY, PYY, and PP are often said to belong to the "pancreatic polypeptide family." However, because of the fact that NPY has been much more conserved during evolution and also exhibits much greater biological activity when compared to PP, this peptide family should more appropriately be called the NPY family (43) (see also Colocalization in Dopamine Neurons, Coexisting Neurotransmitters in Central Noradrenergic Neurons, and General Overview of Neuropeptides).
NPY Is Highly Conserved Through Evolution
Porcine NPY and PYY have 69% sequence homology and activate most NPY-responsive receptors with similar potencies as described below. In contrast, PP shows low efficacy and potency compared to NPY and PYY.
Currently, over 50 sequences of peptides belonging to the NPY family are available (43). Sequence comparison data indicate that NPY is one of the most highly conserved neuroendocrine peptides known; for example, human NPY is 92% identical to torpedo NPY (43). Many investigators have suggested that such a remarkable conservation of NPY would imply an important functional role(s) for this peptide, perhaps one or several of those described below.
Certain Structural Features Are Common to NPY Family Members
A characteristic tertiary structure is observed in NPY (as well as in PYY and PP) which consists of an N-terminal polyproline helix (residues 1–8) and an amphiphilic a-helix (residues 15–30), connected with a b-turn, creating a hairpin-like loop (see ref. 21). This domain has been identified from the crystal structure of avian PP, and nuclear magnetic resonance (NMR) studies also agree well with this three-dimensional configuration. The helices are kept together by hydrophobic interactions. The amidated C-terminal end (residues 30–36) projects away from the hairpin loop (4).
NPY Is Widely Distributed in the Body
Central Nervous System
Neurons displaying NPY-like immunoreactivity are abundant in the central nervous system (CNS), most notably in the so-called limbic structures (14, 34). Immunohistochemical studies indicate that NPY is present within very similar neurons throughout the cerebral cortex and forebrain nuclei, but is contained in a variety of neurons in the hypothalamus, brainstem, and spinal cord. Coexistence with somatostatin and NADPH-diaphorase/nitric oxide synthase is common in cortex and striatum. Although NPY neurons in the latter brain regions receive few inputs, they make numerous contacts with dendrites, including GABAergic neurons. NPY is also extensively colocalized with GABA in the cortex, but not in the striatum (6).
Many studies (reviewed in ref. 34) have demonstrated the occurrence of NPY in a variety of brainstem monoaminergic neurons—that is, arguing coexistence of the peptide with (i) norepinephrine (the A1 group in the ventrolateral medulla, the A2 group of the dorsal medulla and the locus coeruleus), (ii) epinephrine (C1 and C2 groups and solitary nucleus), or (iii) serotonin (nucleus raphe pallidus). It is sometimes overlooked that NPY occurs also in nonmonoaminergic brainstem cranial nerve nuclei.
Unlike the case of NPY, PYY-containing neurons are few in brain and are primarily confined to the brainstem and the cervical spinal cord (10, 16). Finally, PP-positive neurons have not been found in the CNS.
Peripheral Nervous System
The existence of NPY in most sympathetic nerve fibers, particularly around blood vessels (i.e., perivascular fibers), has generated much interest (66). There are dense plexuses of NPY-like immunoreactivity (NPY-LI) found in vascular beds throughout the body. In addition, NPY also occurs in nonadrenergic perivascular, enteric, and cardiac nonsympathetic and parasympathetic nerves (60, 66).
In contrast, PYY occurs primarily in endocrine cells of the lower gastrointestinal tract. However, low amounts of PYY immunoreactivity are also found in certain sympathetic fibers. PP is predominantly located in endocrine cells of the pancreatic islets (60).
NPY Gene Expression Is Tightly Regulated
Human NPY cDNA was originally cloned from a pheochromocytoma (48), a tumor of the adrenal medulla known to often contain high levels of NPY. The corresponding mRNA contained a single open reading frame that predicted a fairly simple precursor for NPY, consisting of 97 amino acids. The predicted precursor (Fig. 1) contains a hydrophobic 28-amino-acid signal peptide that is required for entry into the lumen of the endoplasmic reticulum (ER) and thus, entry into the secretory compartment of the cell. The signal peptide is then cleaved, and the resulting prohormone is 69 amino acids in length. There is no peptide flanking the N-terminus of NPY, as in several other peptides. However, in the prohormone, mature NPY (36 amino acids) is flanked at its C-terminus by 33 amino acids, three of which are the glycine–lysine–arginine motif, necessary for NPY amidation (a feature that is critical for all known actions of NPY with the exception of mast cell degranulation; see ref. 55). The peptide formed by the remaining 30 amino acids of the precursor has been named CPON (C flanking peptide of NPY). Although CPON has been found to be highly conserved (not to the same extent as NPY, but at a level comparable to that of insulin; see ref. 43), its function remains unknown. As expected, CPON is found in every cell that produces NPY, and the two peptides are consequently co-released (4, 59).
The NPY gene is expressed in cells derived from the neural crest, and several factors are involved in its regulation. Consensus sequences for a number of DNA-binding proteins that could act as regulatory factors are contained within the NPY gene. These include five potential GC-rich SP-1 binding sites: two CCCCTC sites, a partial CAAT box, and one AP-1 binding site (3). Additional factors regulating NPY gene expression include activators of cyclic AMP and calcium- or phospholipid-dependent protein kinases (3, 53).
NPY Receptors Are Widely Distributed
Similar to "classical" neurotransmitters (e.g., its frequent coexistence partner, norepinephrine), NPY seems to have a wide range of effects on peripheral (blood vessels, heart, airways, gastrointestinal tract, kidney, pancreas, thyroid gland, platelet, mast cells, and sympathetic, parasympathetic and sensory nerves) and central (effects on pituitary hormone release, behavior, central autonomic control, and other neurotransmitter mechanisms) targets (reviewed in refs. 66 and 69). A number of these actions appear to be exerted by NPY per se, whereas others occur as a result of modulatory interactions with other agents—for example, norepinephrine and glutamate (66, 69). In any case, it is likely that NPY (as well as its related peptides) acts upon membrane receptors that are linked to G proteins (66, 69). When radioreceptor ligand studies employing radiolabeled NPY or PYY were conducted, binding sites in brain and peripheral organs—including vasculature, heart, kidney, spleen, and uvea—were detected (e.g., see refs. 66 and 74).
NPY Receptors Are Heterogeneous
Y1 and Y2 Receptors
Based on studies of sympathetic neuroeffector junctions, it was first proposed that there was heterogeneity among NPY/PYY receptors, specifically Y1 and Y2 receptor subclasses (73); this was later corroborated in other cell types and experimental systems (reviewed in refs. 66 and 69). Already by the mid-1980s, three types of NPY effects at sympathetic neuroeffector junctions were known: (i) a direct postjunctional response, (e.g., vasoconstriction manifested in certain vascular beds), (ii) a postjunctional potentiating effect on norepinephrine-evoked vasoconstriction, and (iii) a prejunctional suppression of stimulated norepinephrine release (73). [The two latter phenomena are probably reciprocal; thus, norepinephrine may affect NPY mechanisms similarly (69).] In principle, long C-terminal amidated fragments of NPY and PYY are essentially inactive in the assays for postjunctional activity (direct as well as potentiating effect), while retaining their efficacy prejunctionally (inhibition of transmitter release). On the basis of the selective prejunctional effect of C-terminal NPY (or PYY) fragments, it was thus proposed that NPY/PYY receptor subtypes may exist (73). The nomenclature Y1 and Y2 was introduced to denote the receptor that required the entire NPY (or PYY) molecule for activation (Y1), and the other receptor subtype which was selectively stimulated by the long C-terminal NPY (or PYY) fragments (Y2) (68). In the neuromuscular preparations, the postjunctional receptors were thus (predominantly) of the Y1-subtype.
It has become apparent in recent years that some actions of NPY cannot be mimicked by PYY. This has been the major argument supporting the existence of an exclusively NPY-responsive Y3-type receptor. This receptor is likely to be present in, for example, adrenal medulla, heart, and brainstem (21).
Features of the Y1 Receptor
Presence on Vascular Smooth Muscle Cells
Initial bioactivity studies suggested that the Y1 receptor is postjunctional at the vascular sympathetic neuroeffector junction (73) and is the sole mediator of pressor responses to NPY (66). The recent cloning of the Y1 receptor cDNA has provided strong evidence (see ref. 42) that fully supports the concept that vascular smooth muscle cells express the Y1 receptor.
Restricted Brain Distribution of the Y1 Receptor
Autoradiography studies using [Pro34]NPY, a Y1 receptor agonist, on frozen sections of rat brain have shown that Y1 receptors are discrete (being much less abundant than Y2 receptors) and mainly localized in distinct layers of the cerebral cortex, anterior olfactory nucleus, amygdala, and a few thalamic and hypothalamic nuclei (1, 15, 66). Furthermore, in situ hybridization with rat Y1 receptor cDNA (see below) has localized expression of this receptor protein in thalamus, cerebral cortex, dentate gyrus of hippocampus, and arcuate nucleus of hypothalamus (17, 66).
In the CNS the Y1 receptor has been associated with a number of biological actions—for instance, with anxiolysis and stimulation of luteinizing-hormone-releasing hormone (LHRH) release (see refs. 66 and 69). However, it is possible that the pronounced effect of NPY on feeding (see below) is mediated through a receptor that is similar but not identical to Y1.
Requirement of Both N- and C-Termini for Y1 Receptor Activation
NPY and its related peptides are capable of retaining their distinct tertiary structure in solution, in contrast to most other small peptide messengers (64). There has been considerable interest in structure–function studies of NPY-related peptides in attempts to develop receptor-specific agonists and antagonists. These experiments were extensively examined using various NPY fragments and analogues both in receptor assays and different biological preparations. In general, PYY and NPY are equipotent and equally effective in all Y1 (and Y2) receptor assays studied (69). A characteristic of the Y1 receptor is that truncation of the first N-terminal residue Tyr1 (NPY 2-36) results in a marked loss of biological activity or affinity (21). The hairpin loop of NPY is thought to present the N- and C-termini closely together for recognition at the Y1 receptor (21).
Molecular Biology of the Y1 Receptor
The Y1 receptor was first cloned as an "orphan" (unidentified) receptor candidate from a rat forebrain cDNA library by a technique that relied on its homology with already cloned members of the G-protein-coupled superfamily of receptors (17). The receptor was subsequently identified as Y1, and the human homologue was isolated (35, 42).
Features of the Y2 Receptor
In the periphery, Y2 receptors are generally considered to be localized at prejunctional sites of autonomic fibers, where they suppress the release of transmitters (66).
Brain Distribution of Y2 Receptors
A differential localization of Y1 and Y2 receptors has been suggested by autoradiographic data from rat brain using radiolabeled PYY, displaced with Y1 or Y2 selective ligands (1, 15). It appears that the Y2 receptor is the quantitatively predominant NPY/PYY receptor type in the rat brain (1, 15, 74).
There is a dense population of Y2 receptors in the hippocampus (1, 15). Electrophysiological studies have suggested the existence of Y2 receptors on excitatory (glutamatergic) inputs to rat hippocampal CA1 neurons (13). NPY has been shown to affect memory processing in a complex manner, possibly reflecting a Y2-receptor-mediated action in the rostral part of the hippocampus (18). The presence of Y2 receptors, and associated inhibition of Ca2+ influx, has also been found in rat dorsal root ganglion neurons (9)).
The C-terminus of NPY is sufficient to activate the Y2 receptor. Unlike the case of the Y1 receptor, NPY 2-36 is about equipotent with NPY or PYY, and N-terminally truncated NPY fragments from NPY 2-36 to 22-36 are rather potent.
Biochemistry/Molecular Biology of Y2 Receptors
The gene encoding the Y2 receptor has not yet been cloned. At present, several groups are pursuing its biochemical isolation using affinity labeling techniques to probe the structure of Y2 receptors from several species and tissues, notably hippocampus and kidney (see ref. 21).
Features of the Y3 Receptor
PYY Has Low Affinity to the Y3 Receptor
The pharmacological order of potency of NPY and its related peptides differs markedly from those of Y1/Y2 receptors in several binding and bioactivity studies on various tissues and cultured cells, which provides support for the existence of a Y3 receptor (66). The main characteristic of these Y3 receptors is that they recognize NPY, while PYY is several orders of magnitude less potent.
Functional Y3 Receptors in the Brain
Injections of NPY and related peptides into the nucleus of the solitary tract (NTS) were recently used to characterize the receptors mediating the evoked cardiovascular effects (22). The pharmacological profile of this response, which is associated with inhibition of glutamate, suggests that the Y3 receptor may be involved (22). These Y3 receptors may also be located in the hippocampus, because NPY, [Leu31,Pro34]NPY, and NPY 13-36, but not PYY or PP, were found to potentiate the excitatory response to the glutamate-receptor agonist N-methyl-D-aspartate (NMDA) in CA3 neurons of rat hippocampus (49). However, a recent slice-patch study showed that NPY produced no changes in NMDA conductances in CA3 pyramidal neurons (47).
Do Additional Receptor Subtypes Exist?
Several lines of evidence have suggested further heterogeneity among the NPY receptors. First, and perhaps most important, it may be that food intake is mediated by a hypothalamic receptor that is similar, but not identical, to the Y1 receptor. Thus, central administration of NPY potently stimulates feeding behavior in rats, and this effect can be mimicked by the Y1 agonist [Leu31,Pro34]NPY, whereas the Y2 agonist NPY 13-36 is much less active. However, uncharacteristic of the Y1 receptor, NPY 2-36 is at least as potent as NPY itself (for references, see ref. 57). Second, although controversial, it has been proposed that NPY and PYY may bind to brain sigma and phencyclidine binding sites (52). Third, a drosophila NPY receptor was recently cloned (45), but it is not yet clear if mammalian homologues exist. Finally, it is unlikely that NPY-induced mast cell degranulation is mediated by any of the defined NPY receptor types, but rather may be the result of direct G-protein activation (55).
NPY Receptors Signal Through G Proteins
Inhibition of cAMP Accumulation
There is much evidence indicating that most, if not all NPY/PYY receptors are coupled to inhibition of adenylate cyclase and, consequently, decreased levels of cAMP (66).
Elevated Intracellular Ca2+
NPY has been observed to raise intracellular Ca2+ levels in many cell types (66). It is controversial whether NPY affects Ca2+ by stimulating phosphatidyl inositide (PI) turnover.
Second Messengers and NPY Receptor Subtypes
The study of second messenger systems may not be useful in distinguishing receptor subtypes because the Y1, Y2, and Y3 receptors appear to be capable of activating the same intracellular pathways in many systems, causing the reduction of cAMP accumulation and elevation of intracellular Ca2+ concentrations (66). Currently, it is not known if one or more G proteins (probably Gi and/or Go) mediate the coupling to cAMP and Ca2+. This dual coupling is also a feature of the cloned Y1 receptor (35, 42).
NPY IN BRAIN FUNCTION AND DYSFUNCTION
Some Hypothalamic Actions Suggest a Role of NPY in Psychiatric Illness
Among numerous actions of central NPY, the peptide exerts a profound influence on some hypothalamic systems which are thought to be dysregulated in depressive syndrome—for example, circadian rhythms, the hypothalamic–pituitary–adrenal (HPA) axis, and food intake.
NPY and Circadian Rhythms
NPY is present in a pathway from the thalamic ventrolateral geniculate to the hypothalamic suprachiasmatic nucleus (SCN), the major entraining pacemaker of the mammalian brain (50). When NPY (or avian PP) was injected into the SCN of hamsters kept under constant light, the free-running rhythm of the animals was shifted in a phase-dependent manner. The length of the activity phase was not affected. The effects resembled those of dark pulses and seemed to be highly localized, because they were not reproduced by intracerebroventricular (i.c.v.) injection of NPY (2).
NPY and the HPA Axis
NPY administered into the hypothalamic paraventricular nucleus (PVN) produced an increase in plasma adrenocorticotropic hormone (ACTH) and corticosterone (72). After i.c.v. administration, low NPY doses suppressed corticosterone secretion, whereas higher doses increased both ACTH and corticosterone levels (25). NPY fibers innervate corticotropin-releasing factor (CRF) cells of the PVN (46), and NPY administration increases hypothalamic CRF levels (23). The release of CRF seems to be stimulated by NPY (65). Mediation of this action through a2-adrenoceptors has been suggested (24). Postsynaptic action of CRF seems also to be potentiated by NPY. In dog the ACTH response to NPY was partially blocked by a CRF antagonist, while a subthreshold dose of NPY potentiated the effects of exogenous CRF (36).
NPY and Feeding
When injected centrally with NPY (or PYY > PP), mammals eat excessively. This effect of NPY has been characterized in numerous elegant studies, recently reviewed in ref. 57. NPY's appetitive action is hypothalamically mediated, although the precise site within the hypothalamus is a matter of some debate. Several groups found the magnitude of NPY-induced feeding to be higher than that induced by any pharmacological agent previously tested, and also extremely long-lasting (e.g., see refs. 12, 44, and 57). NPY-induced stimulation of feeding has been reproduced in a number of species (e.g., see ref. 57). Among the three basic macronutrients (fat, protein, and carbohydrate), the intake of carbohydrates was preferentially stimulated (57). Drinking was affected in a much less consistent manner, which, in addition, varied with species (e.g., see refs. 51 and 57). No tolerance was seen towards the orexigenic effect of NPY, and when administration of the peptide was repeated over 10 days, a marked increase in the rate of weight gain was observed (57!popup(ch52ref57)). Following starvation, the concentration of NPY in the hypothalamic PVN increased with time, and it returned rapidly to control levels following food ingestion (54). This indicates that pharmacological effects of NPY on food intake may be representative of inherent physiological mechanisms. Such a conclusion is also directly supported by push–pull studies of hypothalamic NPY release and its inverse correlation with satiety (37). Because C-terminal fragments of NPY capable of activating Y2 receptors do not induce food intake, it has initially been assumed that the orexigenic action of NPY is mediated by Y1 receptors. Converging data from structure–activity studies (57, 66) and from work with in vivo antisense blockade of Y1-receptor synthesis (see below) suggest, however, that an NPY receptor different from the cloned Y1 species may be involved.
The role of NPY in regulation of appetite may be of clinical relevance in itself. Elevated CSF content of NPY has been reported both in underweight amenorrheic anorectics and in the same amenorrheic patients restudied within 6 weeks after weight restoration. An inverse relationship between CSF NPY and caloric intake in healthy volunteer women was also found. Thus, the increase in CSF levels of NPY may be a secondary, compensatory response to decreased food intake. Also, because NPY regulates a number of endocrine parameters including the secretion of luteinizing hormone (LH), folliclestimulating hormone (FSH), and ACTH, several symptoms of anorexia could be secondary to such an increase in NPY secretion (39, 40).
NPY and Depressive Illness
What direct evidence is there then, to link NPY with the pathogenesis of any psychiatric illness? The levels of NPY-LI were decreased in the CSF of patients with major depression, compared to sex- and age-matched controls (77). The chromatography profiles of immunoreactive material differed between patients and controls (77). This seemed to indicate the processing of the NPY precursor to be altered in depression. Recently, marked reductions of tissue levels of NPY-LI were reported in suicide victims. Particularly dramatic reduction was seen in subjects with a verified diagnosis of major depression (76). The difficulties of purely descriptive radioimmunoassay (RIA)-based studies are however, illustrated by the fact that using an assay based on a different antibody, Berrettini et al. (8) did not find a decrease in CSF NPY-LI in depressed patients.
Increased levels of NPY-LI were found in rat brain tissue after chronic treatment with tricyclic antidepressants (TADs). The most consistent increases were seen in frontal cortical areas (32). Using assays with differing epitope-specificity as well as mRNA quantitation, variable results have been obtained by others, perhaps indicating an altered processing of NPY rather than increased synthesis as a result of TAD treatment (7, 56). Another treatment known to be effective for depressive illness, electroconvulsive shocks (ECS), has yielded more consistent results. In several studies by different groups, ECS produced increased tissue levels of NPY-LI in frontal cortical and in hippocampal areas (58, 67). Increased tissue levels can be due to increased synthesis or decreased utilization. Therefore, studies aimed at quantitating the actual level of transcription are needed. Interestingly, lithium was found to enhance NPY (but not proenkephalin) gene expression (75).
Olfactory bulbectomy has been suggested as an animal model of depression. Bulbectomized rats kill mice ("muricide behavior"). This behavior is inhibited by chronic treatment with antidepressant drugs, but also by injecting norepinephrine (NE) into the central nucleus of the amygdala. This "antidepressant" action of NE is markedly potentiated by NPY (38) (see also Neuropeptide Alterations in Mood Disorders).
Cocaine withdrawal produces depressive symptoms, among which anhedonia (i.e., an inability to experience pleasure) is prominent. The anhedonia of cocaine withdrawal is closely related to that of "endogenous" depression and is effectively treated with conventional antidepressant therapy. Chronic cocaine administration to rats decreased levels of both NPY-LI and NPY mRNA in brain areas important for mechanisms of reward, such as the nucleus accumbens. The decrease persisted after cocaine administration was discontinued, possibly in parallel to the decrease in NPY-LI seen in the CSF and brain tissue of depressed patients. Transsynaptic mechanisms seem to be involved in cocaine's suppression of NPY synthesis, because lesions of medial prefrontal cortex prevented this action of the drug (70) (see also Animal Models of Drug Addiction and Cocaine).
NPY and Anxiety
A lack of correlation between severity of depression and NPY-LI levels argued against a direct, causal relationship between the two. A strong negative correlation was however, seen between anxiety scores and NPY-LI levels in the CSF of depressed patients (28). Anxiety is a core component of the depressive syndrome, and the two psychopathological states may share a common biological basis because both respond to chronic TAD treatment. Shortly following its isolation, it was reported that i.c.v. administered NPY produced a synchronization of the electroencephalogram (19). As a behavioral correlate, a decrease of locomotor activity was seen both in a familiar environment (the home cage of the animal) and in a novel environment (open field) (27). These effects were dose-dependent and fully reversible. In other experiments, i.c.v. administration of NPY to a large extent prevented gastric ulceration induced by a strong stressor, water restraint (26). Thus, both electrophysiological and behavioral observations were compatible with a sedative/anxiolytic action of the peptide. In agreement with such a hypothesis, i.c.v. administration of nanomolar doses of NPY produced a marked anxiolytic-like action in several pharmacologically and ethologically validated animal models of anxiety, including Vogel's punished drinking test, the elevated plus-maze (31), and Geller-Seifter's operant punished responding test (30).
As mentioned above, i.c.v. administration of NPY is known to dramatically increase food intake. Control experiments indicated that the anxiolytic-like action of NPY was specific, and not related to the orexigenic properties of the peptide. In addition, the anxiolytic-like action of NPY was also present in the elevated plus-maze, and this exploratory anxiety model does not involve consummatory behaviors (31). It remained, however, a concern whether the apparent anxiolytic-like action of NPY was in some way related to NPY's effects on appetite. To address this issue, studies were performed to separate these two actions of NPY anatomically. The amygdala is a key structure in central integration of emotionally relevant information, and it seems to encode the stressful effect of aversive inputs. Local microinjections of NPY into the central nucleus of the amygdala reproduced the anxiolytic-like effect of i.c.v. injections with an ED50 of less than 50 pmol/side, but did not affect food intake (29). Thus, the actions of NPY on appetite and anxiety can be both anatomically and functionally separated, and seem to be independent of each other.
Mediation of NPY's action on anxiety by the central nucleus of the amygdala is also interesting in the context of NPY's ability to protect against stress-induced gastric ulcers (see above). The amygdala integrates autonomic responses associated with emotion, and the central nucleus in particular may be the point of output to areas controlling visceral responses to such information. The central nucleus receives a dense NPY-ergic innervation (11) which innervates neurons projecting to the dorsal vagal complex (20). Interactions between NPY and NE may be of importance for NPY's stress protective action. NPY-positive terminals in the central nucleus originate largely from cell bodies in the NTS, in a majority of which NPY is colocalized with NE (50).
With possible drug design in mind, a crucial question was which type of NPY receptors mediate the actions of the peptide on anxiety. After i.c.v. administration, intact NPY was "anxiolytic", whereas a C-terminal fragment with relative preference for the Y2-receptor, NPY 13-36, was not (31, 33). This suggested the effects of NPY on anxiety to be mediated by Y1 receptors. Subsequently, local injections of NPY receptor agonists into the amygdala showed that NPY itself and the highly selective Y1-receptor agonist [Leu31,Pro34]NPY were roughly equipotent with respect to their anxiolytic-like action. The fragment NPY 13-36 produced only marginal effects (29). The accumulated evidence thus suggests the anxiolytic-like action of NPY to be mediated by Y1 receptors in the amygdala.
The strongest evidence for a Y1 mediation of NPY's anxiolytic-like action is supplied by experiments performed to address a related question. While an anxiolytic-like action of exogenously administered NPY seems to be well established and highly specific, it does not constitute conclusive evidence for a similar action of the endogenous transmitter. Similar to the situation with most peptide mediators, the lack of specific receptor antagonists has made it difficult to establish such a role for endogenous NPY. We have attempted to circumvent this difficulty by a novel approach. Short antisense oligonucleotides complementary to a specific mRNA can be taken up through receptor-mediated endocytosis, and they inhibit the translation of a specific message into functional protein. We administered an 18-mer antisense oligonucleotide targeted at the Y1-receptor message using repeated i.c.v. injections. This led to a 60% decrease in the Bmax of Y1-type binding, without affecting Y2 receptors. The decrease in Y1 receptors was accompanied by marked signs of anxiety in one of our standard animal models, the elevated plus-maze, in which NPY itself is markedly "anxiolytic" (71). This provides evidence that endogenous NPY acts in an anxiolytic-like manner by activating Y1 receptors, and that disturbed NPY transmission leads to symptoms of anxiety.
It can be noted that no effects were seen on food intake after the antisense blockade of Y1 receptor synthesis. While not conclusive, this observation suggests that hypothalamic NPY receptors involved in food intake regulation may differ from the cloned Y1 receptor (see above). On the basis of differing pharmacological profiles, this has independently been suggested by others (reviewed in ref. 66; also see Pharmacological Challenges in Anxiety Disorders).
NPY-Containing Neurons Are Relatively Unaffected by Neurodegenerative Disorders
While it deserves mention that NPY-containing neurons show a potential to survive even advanced cases of neurodegenerative disorders (i.e., Huntington's, Alzheimer's, and Parkinson's disease), it is difficult to conceive that NPY plays any unique role in these disorders and, consequently, that potential NPY-ergic drugs may significantly benefit such patients.
CONCLUDING REMARKS: POTENTIAL USE OF NPY-RELATED DRUGS IN NEUROPSYCHIATRIC DISORDERS
No NPY antagonists, in the classical pharmacological meaning, with reasonable potency and/or receptor selectivity have yet gained full acceptance. Moreover, no nonpeptide agonist has been described (66).
First, a centrally acting Y1 agonist will most certainly have anxiolytic properties. Conversely, a Y1 receptor antisense oligodeoxynucleotide, administered directly into the brain of rats, was shown to be markedly anxiogenic, implying that endogenous NPY mechanisms act to tonically relieve anxiety (71). A nonpeptide (and nonbenzodiazepine) anxiolytic drug that mimics NPY at Y1 receptors might find a place in the treatment of affective disorders, including major depression as well as the clinically similar condition that often follows psychostimulant (e.g., cocaine) withdrawal. The latter syndromes have been associated with reduced brain NPY synthesis and, quite often, with severe anxiety.
Second, many investigators work towards the development of a centrally acting NPY antagonist effectively suppressing feeding elicited by (endogenous) NPY. Interestingly, several lines of evidence have recently pointed to the possibility that an "atypical" (not Y1, Y2, or Y3) hypothalamic receptor is involved in this robust feeding response. Hence, in theory it might be possible to find an antagonist capable of suppressing food intake, with specific affinity to the "atypical" hypothalamic NPY receptor, and thereby to avoid possible side effects associated with Y1, Y2, or Y3 receptors.