| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ARTICLE |
-Opioid Receptors Prevents the Acute Orexigenic Action of Agouti-Related Protein
Department of Animal Physiology (S.B.), University of Groningen, 9750 AA, Haren, Groningen, The Netherlands; and Department of Psychiatry (D.J.C., S.C.W., R.J.S.), University of Cincinnati, Cincinnati, Ohio 45267
Address all correspondence and requests for reprints to: Randy J. Seeley, Ph.D., Department of Psychiatry, University of Cincinnati, Cincinnati, Ohio 45267-0559. E-mail: randy.seeley{at}uc.edu.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
- nor the µ-specific antagonist was there any effect to block the effects of AgRP on food intake. However, administration of both the - and µ-receptor antagonists does significantly reduce the effect of AgRP. The current results implicate opioid receptors as critical downstream mediators of the potent effects of AgRP to increase food intake but indicate that either µ- or -receptor activation is sufficient for AgRP’s effect. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-, ß-, and 
-MSH and ACTH. -MSH is an agonist at both the MC3 and 4 receptors, and central administration of -MSH or synthetic analogs, such as MTII, reduce food intake and body weight (3, 4). A unique feature of the MC system is that, in addition to the well-documented role for endogenous agonists, there is also an important role for endogenous antagonists for the various MC receptors. The first antagonist described in the MC system was agouti signaling protein (ASP). ASP acts in the periphery to alter MC1 receptor signaling to regulate peripheral skin and hair color (5). Overexpression of ASP results in a yellow coat color via the MC1 receptor (6). Overexpression of ASP also results in increased food intake and obesity via interaction with MC3 and 4 receptors in the CNS (4).
Given that ASP is not normally found in the CNS, a role for ASP in regulating MC receptor signaling in the brain is unlikely. However, a homolog of ASP termed agouti-related protein (AgRP) is made in the arcuate nucleus of the hypothalamus in a population of neurons adjacent, but not overlapping, with neurons that produce -MSH (7, 8, 9). AgRP acts as an antagonist at the MC3 and MC4 receptors (7, 10). Hence, AgRP opposes the actions of -MSH. Consistent with this, AgRP and the -MSH precursor, proopiomelanocortin, are regulated in opposite directions by energy balance and leptin. Food deprivation decreases proopiomelanocortin gene expression, whereas it increases AgRP gene expression in the arcuate nucleus (11, 12, 13). Overexpression of AgRP results in an obese phenotype similar to that of mice that overexpress ASP (10).
When a biologically active fragment of AgRP is administered into the third ventricle (i3vt) of rats, the increase of food intake lasts for up to 6 d (14, 15). Although the short-term effects of AgRP seem to depend on antagonism at the identified MC receptors, the long-term effects do not (14). Moreover, like neuropeptide Y (NPY), the orexigenic effect of AgRP can be blocked by the nonspecific opioid receptor antagonist naloxone (16, 17). Thus, the potent effects of AgRP to increase food intake depend critically on activation of opioid receptors. Opioid receptors come in three identified subtypes: µ, , and 
(18); and of these, both µ and have been strongly implicated in the control of food intake (19). Thus, the aim of these experiments was to assess the role of these specific opioid receptor subtypes in mediating the effects of AgRP to increase food intake.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Surgery
A guide sleeve [Plastics One Guide (22 G, catalog no. C313G)] and obdurator (18 G, catalog no. C313 DC), aimed at the third cerebral ventricle, were implanted. Coordinates were on the midline, 2.2 mm posterior to bregma and 7.5 mm ventral to dura, with bregma and 
at the same vertical coordinates (Paxinos and Watson, 1986). After a minimum of 10 d of recovery, accuracy of placement was verified by injection (i3vt) of 10 ng angiotensin in saline through an injector inserted into the guide sleeve. Only rats that drank 5 ml or more in the hour after injection were used in the experiments.
Chemicals
AgRP(83–132) was purchased from Phoenix Pharmaceuticals, Inc. (Belmont, CA). Nor-Binaltorphimine (NorBNI) hydrochloride and ß-funaltrexamine (ß-FNA) hydrochloride were purchased from Sigma (St. Louis, MO). NorBNI has been widely used as a selective antagonist for -opioid receptors (20, 21), and ß-FNA has been widely used as a selective µ-opioid receptor antagonist (18, 22). The selective antagonist action of ß-FNA has a unique time course, becoming more selective for this action over an apparent 20 h after administration (23). Thus, for all of the experiments, ß-FNA is actually administered 20 h before food intake measurements. AgRP was dissolved in physiological saline, which served as the control solution. NorBNI and ß-FNA were dissolved in 100% methanol that served as their control solution. All solutions were administered i3vt in a 2-µl vol.
Feeding schedule
The animals had access to a high-fat (HF), modified AIN-93M-purified, pelleted rodent diet (Dyets, Inc., Bethlehem, PA; see Table 1
for diet composition) for 2 h each day, starting at lights-off. For the remaining 22 h, the rats had access to pelleted laboratory chow (Harlan-Teklad, Indianapolis, IN). This paradigm was used because it produces very reliable 2-h intakes and because both AgRP and the ligands of the opioid receptors potently influence intake of a preferred HF diet more than other diets (24). Consequently, if opioids are important in mediating the effects of AgRP, it is more likely to be observed during HF feeding than alternative paradigms. Water was available ad libitum. Body weights and baseline food intake were measured throughout the experiment.
|
-opioid receptor antagonist
Experiment 2: dose response determination for the µ-opioid receptor antagonist
Rats from experiment 1 were used in experiment 2 after they had 7 d with no experimental manipulations and their baseline intakes and body weight had returned to normal (which occurred within 48 h). Analogous to experiment 1, we determined the i3vt dose-effect curve for the µ- selective antagonist ß-FNA by administering 0, 0.1, 0.5, 1, and 5 nmol doses, 20 h before access to the HF diet, and measuring subsequent HF and chow consumption. Data were analyzed by a one-way ANOVA, followed by Dunnet’s t tests, to compare each dose with the vehicle control.
Experiment 3: effect of NorBNI on AgRP(83–132)-induced food intake
On each experimental day, rats received two i3vt injections. The first occurred 1 h before the second and contained NorBNI at the highest subthreshold dose determined in experiment 1 or vehicle. The second injection contained either saline or AgRP (1 nmol). The second injection occurred 30 min before lights off, and therefore, 30 min before access to the HF diet. Each rat (n = 22) was tested twice, 7 d apart, in two of the four conditions: 1) methanol + saline; 2) NorBNI (5 nmol) + saline; 3) methanol + AgRP (1 nmol); or 4) NorBNI (5 nmol) + AgRP (1 nmol). In this 2 x 2 design, AgRP vs. saline was a within-subjects factor, whereas NorBNI vs. methanol was a between-subjects factor with condition order counterbalanced across subjects. Food intake was measured every 30 min during access to the HF diet. After 2 h, the HF diet was replaced with chow, and chow intake was recorded over the remaining 22 h, and data were analyzed using a mixed-model ANOVA followed by Tukey’s post hoc tests.
Experiment 4: effect of ß-FNA on AgRP(83–132)-induced food intake
Experiment 4 used a separate group of naive animals (n = 18) in a design that exactly parallels experiment 3 except that the selective opioid antagonist ß-FNA, at the maximally subthreshold dose determined from experiment 2 (0.1 nmol), or methanol was given 20 h before AgRP or saline administration. Again, food intake was measured every 30 min during access to the HF diet. After 2 h, the HF diet was replaced with chow, and chow intake was recorded over the remaining 22 h, and data were analyzed using a mixed-model ANOVA followed by Tukey’s post hoc tests.
Experiment 5: effect of combined NorBNI and ß-FNA on AgRP(83–132)-induced food intake
The goal was to determine whether combined blockade of both the µ- and -opioid receptors attenuates AgRP-induced food intake. Rats from experiment 3 and experiment 4 were used in this experiment after at least 7 d rest and food intake and body weight had returned to baseline levels. On the test day, animals received three separate injections. Twenty hours before receiving either AgRP or saline, rats received an injection of either methanol or ß-FNA (0.1 nmol). Methanol or NorBNI (5 nmol) was then injected 1 h before the final injection of saline or AgRP (1 nmol). At lights off, animals were given access to the HF diet for 2 h. Food intake was measured every 30 min. After 2 h, the HF diet was replaced with chow, and chow intake was measured over the remaining 22 h. Each animal was run in only one condition (n = 6 or 7/group) Data were analyzed by a two-way ANOVA followed by Tukey’s post hoc tests.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
. The one-way ANOVA on 2 h cumulative intake was significant [F(3, 18) = 3.211 P < 0.05]. Injection of 10 nmol NorBNI significantly reduced intake of the HF diet during the first 2 h of the dark phase, compared with vehicle (P < 0.05). The total 2-h intake is representative of the intake at earlier times (data not shown). The 5-nmol dose had no reliable effect on 2-h intake and was therefore the highest subthreshold dose for use in subsequent experiments (see Fig. 1). No dose of NorBNI reduced chow intake over the subsequent 22 h (data not shown). In all of our experiments, methanol was used as the vehicle. Under some conditions, methanol has been shown to be toxic to neurons. To determine whether methanol could independently influence food intake in our rats, we compared the food intake on the day before the beginning of the injections to the methanol-alone condition. On the methanol-alone day, rats ate 13.8 ± 0.96 g; whereas, on the noninjection day, rats consumed 14.2 ± 1.1 g (P > 0.05, paired-sample t test), indicating that, in this paradigm, methanol alone did not have a significant effect on intake.
|
. The one-way ANOVA on 2-h cumulative intake was significant [F(3, 18) = 3.334 P < 0.05], and post hoc analyses revealed that injection of 1 nmol ß-FNA significantly suppressed 2-h intake of the HF diet, compared with vehicle (P < 0.05,). There was a nonsignificant trend (P = 0.06) for the 0.5-nmol dose to inhibit food intake (see Fig. 2). Consequently, we chose to use 0.1 nmol as the highest, clearly subthreshold dose for further experiments. Chow intake over the following 22 h was not reduced by any dose of ß-FNA.
|
] and the subsequent chow intake (data not shown). There was no effect of NorBNI to influence food intake [F(1, 16) = 1.3, P > 0.05], nor was there a significant interaction between AgRP and NorBNI [F(1, 16) = 0.72, P > 0.05, see Fig. 3]. Thus, unlike subthreshold doses of the nonspecific opioid antagonist naloxone, NorBNI did not reduce AgRP’s orexigenic effect at 2 h. It should be noted that the baseline food intakes (both on uninjected days and on the vehicle-injected days) are lower in experiments 3, 4, and 5 than in experiments 1 and 2. Experiments 3, 4, and 5 were all run during the summer months, when both the ambient temperature and ambient humidity are higher than in February, when experiments 1 and 2 were run. Despite the vivarium being both temperature and humidity controlled, we have observed in other experiments that the animal rooms do have higher levels of humidity during the summer and that this reduces consumption of the HF diet. Importantly, the dose of NorBNI that we determined to be subthreshold in experiment 1 remained subthreshold in experiment 3 despite the lower baseline food intakes.
|
, F(1, 14) = 10.29, P < 0.01] and the subsequent chow intake (data not shown). ß-FNA did not alter food intake on its own [F(1, 14) = 2.5 > 0.1], nor did it alter this increase in 2-h food intake caused by AgRP [see Fig. 4, interaction F(1, 14) = 0.04, P < 0.8], nor did it act to alter the AgRP effect at any other time point (data not shown). Thus, unlike subthreshold doses of the nonspecific antagonist naloxone, ß-FNA did not reduce AgRP’s orexigenic effect.
|
] and chow intake (data not shown), compared with the baseline condition. The combined administration of NorBNI and ß-FNA significantly attenuated the effect of AgRP to increase intake of the HF diet [interaction, F(1, 17) = 7.47, P = 0.01, see Fig. 5]. Importantly, this occurred despite the fact that the combined administration of the two opioid receptor antagonists at these doses had no effect on food intake in the absence of AgRP [F(1, 17) = 3.67, P > 0.05]. The effect of AgRP to increase 22-h chow intake was not altered by the combined opioid receptor antagonists (data not shown).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-selective antagonist alone did not reduce the ability of AgRP to increase intake. However, when µ- and -antagonists were combined at doses that had no effect on food intake when administered without AgRP, they significantly attenuated the ability of AgRP to increase intake of the HF diet. Like naloxone, this effect is associated only with the acute effects of AgRP, and combined µ- and - receptor antagonism could not block the long-term effects of AgRP.
These data are therefore consistent with the hypothesis that the acute effect of AgRP to increase food intake depends on activation of opioid receptors. However, these data also indicate that either µ- or -opioid receptor activation is sufficient to mediate the effects of AgRP, such that only when both receptor subtypes are blocked does AgRP no longer activate critical aspects of CNS circuitry necessary to increase food intake acutely. This result is potentially analogous to NPY, where suprathreshold doses of either - and µ- (but not -) receptor antagonists reduce the orexigenic effect of NPY (25). However, in those experiments, it was not determined whether combined - and µ-receptor blockade would further inhibit NPY-induced feeding.
NPY and AgRP are largely colocalized within the arcuate nucleus of the hypothalamus (8, 9, 26), and mRNA for both is elevated after fasting and also in ob/ob and db/db mice (27). Moreover, the acute effects of both NPY and AgRP are associated with recruitment of many of the same brain regions (28). Consequently, the current data extend the similarity between NPY and AgRP to include recruiting the same opioid receptor subtypes.
Several lines of evidence implicate the opioid system in the control of food intake (29). Microinjection experiments have identified numerous sites, at every level of the neuroaxis, in which opioid agonist and antagonists affect food intake (for review, see Ref. 30). In particular, whereas opioid agonists and antagonists increase and decrease food intake, respectively, they also influence food choice. Both - and µ- selective antagonists have been found to decrease intake of an HF diet to a greater extent than they decrease intake of a high-carbohydrate diet (31). Controversy exists, however, as to whether the effect of opioids is to alter the animal’s choice of macronutrients or simply to drive the organism to consume more of an already preferred food (32). Regardless of the interpretation, this is consistent with an important role for these opioid receptor subtypes in mediating the effects of AgRP. Similar to what occurs after manipulation of opioid receptor signaling, administration of AgRP selectively increases intake of a preferred HF diet (17). Additionally, stimulation of µ-receptors in the nucleus accumbens preferentially enhances HF feeding (33). Such strong parallels between AgRP and opioid receptor ligands on food intake and food selection strengthen the hypothesis that these two systems interact in an important way to influence food intake and food choice.
The opioid system encompasses a number of distributed receptor populations as well as a number of endogenous ligands for those receptors. In these experiments, the opioid receptor antagonists were delivered into the third ventricle and, hence, close to the hypothalamus. However, given the lipophyllic nature of these compounds and the rostral- to-caudal flow of cerebrospinal fluid in the ventricular system, it is likely that these antagonists could interact with opioid receptors in a variety of neuroanatomical sites (34). Although these experiments, therefore, cannot determine the exact location of the critical µ- and -opioid receptors, some clues can be derived from the areas that show increases in fos-like-immunoreactivity after third-ventricular administration of AgRP. Two hours after AgRP administration, fos is increased in the accumbens shell, lateral septum, and paraventricular and dorsomedial nuclei of the hypothalamus. After 24 h, fos is increased in the accumbens shell, lateral septum, central nucleus of the amygdala, lateral hypothalamus, and nucleus of the solitary tract (35). All of these regions show evidence for µ- or -receptor binding and/or expression, but only the nucleus of the solitary tract and the lateral hypothalamus show high levels of binding for both receptors (33, 36, 37). Future research will need to address whether the activation of either µ- or -receptor subtypes sufficient to produce AgRP’s effects is in the same neuroanatomical sites or occurs as a result of activation in disparate sites.
A separate (but related) question is which endogenous opioid ligand or ligands that are blocked by the specific receptor antagonists have increased release after AgRP administration. The search for specific functions of opioid ligands has been hampered by the fact that they all interact with multiple subtypes of opioid receptors (20, 38) and the recent demonstration that opioid receptor subtypes can make heterodimers (39). Nevertheless, both ß-endorphin and dynorphin have high affinity for both µ- and -opioid receptors (40, 41, 42), and ß-endorphin is made in the arcuate nucleus of the hypothalamus (43), in cells that have been hypothesized to express the MC3 receptor (44, 45) and in the nucleus of the solitary tract in a region of high MC4 receptor expression. Thus, it is possible that AgRP given into the third ventricle could exert direct effects on one or both of these ß-endorphin populations. Food intake stimulated by ß- endorphin can be significantly and dose-dependently attenuated by pretreatment with either the µ- or the -selective antagonists (46, 47). As reported by Silva et al. (47), because the antagonists only worked at high equimolar doses, relative to ß-FNA, it is conceivable that they could be exerting their effects through multiple (rather than specific) opioid receptors. Because none of the antagonist doses used in the Silva study completely eliminated the ß-endorphin-induced feeding, they proposed that, analogous to our findings with AgRP, blockade of multiple opioid receptors might be necessary to produce this effect. Dynorphin is made both in the paraventricular nucleus and lateral area of the hypothalamus, and both regions express high levels of the MC4 receptor (48). Both our group and Berthoud and colleagues (35) have shown that the AgRP-induced fos in the lateral hypothalamus is colocalized with orexin-A immunoreactivity. Because dynorphin is heavily colocalized with orexin-A in the lateral hypothalamus (49), this provides direct evidence that AgRP can activate dynorphin-containing neurons. Use of either mice with targeted deletion of ß-endorphin (50, 51) or dynorphin could clarify the role of these two opioid peptides in the orexigenic actions of AgRP.
The exact role of the opioid system in the control of food intake remains uncertain. Given its large number of different receptors and ligands, however, it is likely that its role is not one-dimensional and, rather, encompasses a number of aspects of ingestive behavior (19). The present data point to a specific role for µ- and -opioid receptors in mediating effects of the hypothalamic MC system. This intimate relationship with the MC system points to the long-term regulation of energy balance as one of the important roles played by the opioid system. Future research will need to address the functional interactions of the opioid system with hypothalamic peptides and also explore the potential involvement of the opioid system in the etiology and treatment of disorders of energy-balance regulation, such as obesity.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Abbreviations: AgRP, Agouti-related protein; ASP, agouti signaling protein; CNS, central nervous system; ß-FNA, ß-funaltrexamine; HF, high-fat; i3vt, administered into the third ventricle; MC, melanocortin; NorBNI, nor-Binaltorphine; NPY, neuropeptide Y.
Received February 27, 2002.
Accepted for publication July 22, 2002.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-melanocyte stimulating hormone in vivo. Endocrinology 139:4428–4431-opioid antagonist in analgesic and receptor binding assays. J Pharmacol Exp Ther 246:255–258 opioid antagonists upon sucrose intake in sham-fed rats. Brain Res 685:205–210[CrossRef][Medline]
, and opioid receptor mRNA expression in the rat CNS: an in situ hybridization study. J Comp Neurol 350:412–438[CrossRef][Medline]
opioid receptors: a role in opiate synergy. J Neurosci 20:RC110
This article has been cited by other articles:
 |
|
 |
|
  F. Y. Obese, B. K. Whitlock, B. P. Steele, F. C. Buonomo, and J. L. Sartin Long-term feed intake regulation in sheep is mediated by opioid receptors J Anim Sci, January 1, 2007; 85(1): 111 - 117. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 V. S. Densmore, N. M. Morton, J. J. Mullins, and J. R. Seckl 11{beta}-Hydroxysteroid Dehydrogenase Type 1 Induction in the Arcuate Nucleus by High-Fat Feeding: A Novel Constraint to Hyperphagia? Endocrinology, September 1, 2006; 147(9): 4486 - 4495. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Tabarin, Y. Diz-Chaves, M. d. C. Carmona, B. Catargi, E. P. Zorrilla, A. J. Roberts, D. V. Coscina, S. Rousset, A. Redonnet, G. C. Parker, et al. Resistance to Diet-Induced Obesity in {micro}-Opioid Receptor-Deficient Mice: Evidence for a "Thrifty Gene" Diabetes, December 1, 2005; 54(12): 3510 - 3516. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. R. Vulliemoz, E. Xiao, L. Xia-Zhang, S. L. Wardlaw, and M. Ferin Central Infusion of Agouti-Related Peptide Suppresses Pulsatile Luteinizing Hormone Release in the Ovariectomized Rhesus Monkey Endocrinology, February 1, 2005; 146(2): 784 - 789. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
