Peripheral administration of urocortin suppresses operant responding for food reward

Peptides 22 (2001) 583–587

Peripheral administration of urocortin suppresses operant responding for food reward Jefferson W. Kinneya,*, Brandi Scruggsb, David D. Averyb a

Section on Behavioral Neuropharmacology, NIMH, Bethesda, MD 20892-1375, USA Department of Psychology, Colorado State University, Ft. Collins, CO 80523-1876, USA

b

Received 25 April 2000; accepted 9 November 2000

Abstract The effects of peripheral systemic administration of urocortin on operant responding to obtain food were investigated using three separate concentrations. The drug was administered intraparitoneally at a concentration of 10 ␮g/ml/Kg, 5 ␮g/ml/Kg, and 0 ␮g/ml/Kg suspended in saline at a volume of 1 ml/Kg to Sprague-Dawley rats fifteen minutes prior to being exposed to an operant bar press task. Eleven subjects were used, each receiving a single injection of each concentration on separate days with the order of treatment counterbalanced. The results indicated that the administration of urocortin in a dose dependent manner reduced responding of food deprived subjects for a food reward in a thirty minute session. These data indicated that the peripheral administration of urocortin reduced the motivation of food deprived subjects to respond. © 2001 Elsevier Science Inc. All rights reserved. Keywords: Urocortin; Corticotropin-releasing factor; Satiety; Operant responding

1. Introduction The investigation of ingestive behavior, specifically satiety has been a focus of considerable research in the past thirty years. A number of signals have been implicated to be involved in the endogenous regulation of feeding and satiety. In addition to the investigation of the chemical signals that regulate feeding behavior (for a review see 9,14), a large body of work has been focused on the brain regions that mediate such behavior. Some of the research to date has investigated the hypothalamus, specifically the VMH and LH, and the pituitary gland [4,19]. Some of the chemical signals that have been implicated include: bombesin [15, 10], cholecystokinin [22], serotonin [7] morphine [5,16], phenalpropanolomine [18], and the corticotropin releasing factor [12]. A great number of studies have relied on psychopharmacological investigations to determine the contribution these substances make. Such investigations have consisted of subjects receiving an injection (either centrally or peripherally) followed by some behavioral measure of

* Corresponding author. Tel.: ⫹301-496-4838; fax: ⫹301-480-1164. E-mail address: [email protected]

food or water intake. For a review of such procedures please see [6]. The injection of an endogenous ligand capable of producing satiation has become more of an interest to research as the incidence of obesity continues to increase. Examples of previous research that have focused on inducing satiety in food deprived rats through injection of an exogenous chemical include studies examining the effects of cholecystokinin (CCK) and bombesin (BBS) on motivation to perform the bar press task in an operant chamber for a food reward. Results from these studies have shown that peripheral injections of CCK and BBS curb operant responding for a food reward [3]. The majority of the work investigating CCK and BBS affecting satiety is based on the hypothesis that these endogenous ligands are released in response to food intake and represent a peptide that directly mediates satiety upon release. An alternative mechanism that has been investigated in the induction of satiety in a hungry animal has focused on the stress response. Numerous environmental stimuli have the capacity to elicit behaviors associated with increased sympathetic arousal. Behaviors that are typical following activation of the hypothalamus-pituitary-adrenal axis (HPAstress axis) include increased grooming, locomotion, decreased exploration of a novel environment, and decreased

0196-9781/01/$ – see front matter © 2001 Elsevier Science Inc. All rights reserved. PII: S 0 1 9 6 - 9 7 8 1 ( 0 1 ) 0 0 3 6 7 - 9

584

J.W. Kinney et al. / Peptides 22 (2001) 583–587

food intake [17]. The investigation of the reduction of food intake following sympathetic arousal has attempted to identify the chemical messages responsible for decreased food intake. One of the important messages linked to the decrease in ingestive behaviors due to a stressor being introduced is a peptide known as the corticotropin releasing factor (CRF). CRF is a hypothalamic neuropeptide thought to be responsible for the release of ACTH from the pituitary [1]. Previous research has shown that the behavioral consequences of injecting CRF intra-cerebroventricularly (ICV) include increased grooming, movement, decreased exploration of a new environment, and decreased food intake, all behaviors typical of the behaviors seen during the HPA activation [13]. Of more importance for the present experiment, CRF has also been shown to induce a reduction in food intake when injected ICV into the paraventricular nucleus (PVN) of the hypothalamus [13], as well as decreased body weight shown by decreased food intake when injected into genetically obese Zucker rats [1]. It has been suggested that the decrease in food intake and subsequent decrease in body weight seen with injection of CRF or CRF agonists could be the result of appetite loss as well as increased metabolic rate due to the increase in locomotion and cardiac output seen with the stress response induced by CRF [11]. Of specific interest for the current study was the isolation of a CRF like peptide known as urocortin (UCN). Urocortin, like CRF, leads to secretion of ACTH from the pituitary, although it is not yet known if UCN accesses the pituitary through the peripheral blood stream or through a feedback mechanism via the pituitary [23]. Localization of the specific UCN receptors have shown that UCN binds to both the CRF1 receptor and CRF2 receptor binding sites, although it binds with higher affinity to the CRF2 receptor site. The distribution of CRF1 receptors is diffuse, and is found throughout the central nervous system, including structures such as the cerebral cortex and cerebellum. However, the CRF2 receptors have been found to be more localized to subcortical structures such as the limbic system and hypothalamic nuclei. The localization of the type 2 CRF receptor in regions of the brain that have been linked to ingestive behaviors has led researchers to hypothesize that the CRF2 receptor mediates stress-induced satiety. In addition, UCN has been found to be a potent agonist of the type 2 receptor, therefore researchers have proposed that UCN may be a member of the CRF family responsible for the anorectic effects associated with the stress response. Smagin, Howell, Ryan, De Souza, and Harris (1998) found that injection of antisense oligonucleotides to the CRF2 receptor mRNA reduced the anorectic effect of UCN in rats. It has also been proposed that UCN produced an anxiogenic effect as well as an effect on food intake. A study performed by Moreau et al. (1997) found that, at similar doses, UCN produced very similar anxiogenic-like behaviors equivalent to those found with injections of CRF. They concluded that UCN shared

comparable anxiety-producing properties, as well as anorectic properties, with CRF. Studies focusing on the satiety-inducing properties of UCN have shown that ICV injections of UCN at varying dosages produce satiety in food deprived rats as shown by decreased food consumption in their home cages. These researchers also provided evidence that the animals did not exhibit anxiogenic behaviors when tested in the elevated plus-maze [21]. More recently, a study by Asakawa, Inui, Ueno, Makino, and Fujino (1999) showed that intraperitoneal (IP) injections of UCN produced much larger decreases in food intake in genetically obese mice than any other peptides examined, including CCK and Leptin. Therefore, the next logical step in the investigation of the effects of UCN on feeding and satiety was to examine the operant responding of food-deprived rats injected IP with UCN. In order to investigate this, an operant reward system was utilized (shown to be effective by Babcock et al. 1985, and Bowers and Herzog, 1991). It was hypothesized that peripheral injections of UCN would produce a dose-dependent decrease in feeding behavior, as indicated by a reduction in the behavior that produces food reinforcement, i.e. bar press responding in an operant behavioral task. This would indicate decreased motivation for food intake.

2. Methods 2.1. Subjects Twelve adult male Sprague-Dawley rats obtained from Charles River Laboratories were used as subjects in this study. The animals were approximately 100 days old weighing between 250 grams and 300 grams when they arrived. All subjects were individually housed in hanging plastic cages with bedding material. The colony room in which the animals were housed was maintained at 32°C and on a twelve hour light/dark cycle (light phase began at 7:00 am). Subjects were given free access to water throughout the experiment in their home cages. Upon arrival subjects were given free access to food in their home cages for three weeks, with individual subject weight recorded each day. Following the three weeks of free feeding, subjects were then maintained at an 85% free feeding weight on a twentythree hour food deprivation schedule. The 85% free feeding weight was derived from the largest weight each animal obtained in order to account for growth throughout the remainder of the experiment. Subjects were weighed daily, then given supplemental feedings to maintain their individual 85% free feeding weight. 2.2. Apparatus Subjects were fed Purina rat chow in their home cages, and Noyes .45 gram pellets in the operant chambers. All injections were performed with Beckton-Dickenson 1 cc

J.W. Kinney et al. / Peptides 22 (2001) 583–587

syringes and Beckton-Dickenson 27 1/2 gauge needles. The saline injected, as well as used to mix the urocortin was 0.9% NaCl. The operant chambers used were manufactured by BRS/LVE and were equipped with two response levers, each 3 cm from the grid floor and 3 cm from each of the side walls. The chamber also contained a houselight mounted directly between the response levers and raised 9 cm, a food hopper equidistant from each of the response levers, and three cue lights mounted above each of the response levers. The operant chambers were placed in BRS/LVE sound attenuation chambers that allowed observation of the subjects through one-way mirrors without interfering with behavior. The urocortin used in this study was purchased from Bachem pharmaceuticals in powder form. The drug was mixed with saline to achieve concentrations of 5 ␮g/ml (low dose), and 10 ␮g/ml (high dose). 2.3. Procedure Prior to the subjects reaching their 85% free feeding weight, they were handled and habituated to the intraperitoneal (IP) injections used throughout the study. The IP injections were administered at a volume of 1 ml/Kg animal body weight. Once the subjects reached their 85% free feeding weight, chamber habituation and magazine training were initiated. Subjects were placed in the operant chambers individually and allowed to explore the chamber and retrieve food rewards for two forty minute sessions. Following the magazine training, subjects were shaped to press the left response lever at the back of the chamber. The method employed to shape the bar press response was differential reinforcement of successive approximations. Throughout the shaping procedure, and all remaining sessions, a white cue light over the left response lever was illuminated when a session was underway. Once the animals had been shaped to press the bar for a food reinforcer, a ratio training protocol was initiated. Subjects were progressively moved from a continuous reinforcement schedule of reinforcement to a fixed ratio five schedule of reinforcement. The ratio training consisted of increasing the ratio size by one each day, following which another eight sessions were used to stabilize the response rates of the subjects. The criteria for response rate being stable was the number of responses in a thirty-minute time frame for each subject not being significantly different across training sessions. Once the subjects’ response rates were stable, the experimental sessions began. The experimental session were carried out over a five-day period. On day’s one, three, and five subjects were injected with one of the three concentrations of urocortin. On days two and four all subjects were injected with saline, and exposed to the same procedures as all other experimental sessions. All subjects were administered each treatment once, with a session between each experimental session to maintain responding. The concentrations of drug used were; high dose UCN (10 ␮g/ml/Kg), low dose UCN

585

(5 ␮g/ml/Kg), and saline (1 ml/Kg). The selection of which treatment was administered on each of the experimental sessions was counterbalanced to ensure experience in the chamber did not affect responding. In addition the experimenters were blind to which treatments contained the drug. Subjects were placed into the operant chambers fifteen minutes following the IP injection of one of the three solutions during the experimental sessions. The subjects were allowed thirty minutes in the chamber on each of the experimental sessions to press the response lever for food. Following the thirty minutes in the chamber, the subjects were weighed and given supplemental feeding to maintain their individual 85% free feeding weight. The number of responses for each subject in the thirty minutes for each of the treatments injected was collected. All behavioral testing was carried out at the same time of day throughout the experiment to ensure consistent motivation for the subjects. The data collected were analyzed via the repeated measures analysis of variance within subject design. Any significant differences observed in the mean number of responses for each treatment were further analyzed via a Tukey post-hoc statistic to determine where the differences existed.

3. Results One subject was removed from the study following the second session of the five-day experiment. This animal ceased responding while in the operant chamber when injected with saline. No connection between drug administration and the cessation of responding in this animal could be made, as the first treatment was saline. The remaining eleven subjects completed the five session experimental phase. The collected data indicated a dose dependent relationship between the number of responses and the solution injected (see Fig. 1). A repeated measures analysis of variance performed on the operant responding data revealed a significant effect due to the solution administered, F [2,20] ⫽ 11.965 p ⫽ .0004. Tukey post-hoc comparisons were performed to determine which groups reliably differed from each other. Significant differences were obtained between the mean number of bar presses when subjects were injected with 10 ␮g/ml/Kg versus injected with saline ( p ⱕ .05), as well as significant differences between the mean number of bar presses when subjects were injected with 5 ␮g/ml/Kg and injected with saline ( p ⱕ .05). In addition, a significant difference was obtained between the 10 ␮g/ml/Kg dose and the 5 ␮g/ml/Kg treatments ( p ⱕ .05). No reliable subject effect was produced in the present experiment, F [10,22] ⫽ 1.582 p ⫽ .1772. No significant interaction of treatment and time during the thirty minute test sessions was obtained in the current study.

586

J.W. Kinney et al. / Peptides 22 (2001) 583–587

Fig. 1. Subjects’ mean number of bar press responses for each of the three treatments injected intraparitoneally. Drug treatments were 10 ␮g/ml/Kg urocortin (high dose), 5 ␮g/ml/Kg urocortin (low dose), and 0 ␮g/ml/Kg (Saline). Subjects were injected fifteen minutes prior to being placed in the operant chambers for thirty minutes. Significant dose dependant differences between treatments were obtained (* ⫽ p ⬍ .05).

4. Discussion The data clearly indicated that the IP administration of urocortin significantly decreased the mean number of operant responses performed by the food deprived subjects. These findings were consistent with previous work which investigated the satiating effect UCN produces [21]. In addition, the data indicated that the differences in responding due to the concentration injected were dose dependent. The animals produced the fewest number of responses when injected with the high dose of urocortin, then an intermediate number of responses when injected with the low dose of urocortin, and finally, produced the greatest number of responses when injected with saline. Because the experiment utilized a within design, and no reliable differences were obtained among subjects, the strength of the claims that can be made regarding the suppressing effect the administration of the drug produced in operant responses was strengthened. Each subject received an injection of each of the concentrations of urocortin on separate days, which resulted in

fewer responses as the concentration of urocortin increased. The researchers interpreted the reduction in responses as a reduction in motivation to obtain the food reward, thus indicating satiety. Throughout the experiment general behavioral observations were taken to determine if any changes in responding might be due to malaise, or increased sympathetic arousal. The data gathered did not indicate any reliable changes in behavior relating to illness (malaise) when the drug was administered, or any pyloerection (indicative of sympathetic arousal) when subjects were injected with urocortin. The reduction in responding therefore probably reflected decreases in the subject’s motivation to eat, thus indicating satiety. Although no evidence was gathered indicating an increase in sympathetic arousal following the peripheral injections, the possibility exists that the urocortin affected responding by eliciting activation of sympathetic efferents. A follow up procedure to the current study should directly examine glucocorticoid levels following the IP administra-

J.W. Kinney et al. / Peptides 22 (2001) 583–587

tion of the UCN. In addition, methods similar to Moreau et al. (1997) should be employed to determine if UCN produced anxiogenic-like behaviors in the current project. Previous research investigating the satiating effect of UCN has principally focused on the changes in food intake following either peripheral or central administration. However, the measure recorded as an indicator of satiety in this experiment differed from the previous research investigating UCN. Because motivation to obtain a food reward was the measure recorded, the data from the current study suggests that the alteration in ingestive behavior produced by UCN is not likely due to an alteration in the animals’ ability to engage in ingestive behaviors. The subjects in the present experiment did show a reduction in responding following the UCN administration; however, the subjects did maintain some responding. If the alterations in food intake were simply due to illness, we would have expected cessation of responding following the injection of UCN. The data from the current experiment clearly indicated that the administration of the UCN decreased the number of operant responses of the subjects to obtain food, but did not abolish them. The absence of any indication of sympathetic arousal and malaise suggested that the most likely cause of the decreased responding was a reduction in the animal’s motivation to obtain food. Such results suggest a graded and dose-dependent relationship between UCN and motivation to obtain food. These findings are consistent with an interpretation that the urocortin may elicit a satiating effect on the subjects. A final note to consider regarding the data gathered in the present experiment is the mechanism by which the UCN elicited an alteration in behavior. Because the administration of UCN was done peripherally, and alterations in behavior were seen, further information needs to be gathered regarding the site of action of UCN. Previous work with both CRF and UCN has demonstrated that central administration affects eating behavior. It has been proposed that the UCN binding to sites on the pituitary produce the reduction in feeding. The finding that peripheral administration also alters feeding behavior as well as motivation merits attention. The crucial question that must be addressed is whether the central and peripheral administration of UCN act via the same receptor sites, or if the peripheral administration is mediated by peripheral receptors. In addition, if systemic UCN administration alters ingestive behavior, research should attempt to determine if there is an endogenous UCN like peptide that may affect behavior in a similar fashion, not necessarily via HPA mechanisms. References [1] Arase K, Shargill N, Bray G. Effects of corticotropin releasing factor on genetically obese (fatty) rats. Physiol Behav 1989;45:565–70.

587

[2] Asakawa A, Inui A, Ueno N, Makino S, Fujino M, Kasuga M. Urocortin reduces food intake and gastric emptying in lean and ob/ob obese mice. Gastroenterology 1999;116:1287–92. [3] Babcock A, Livosky M, Avery D. Cholecystokinin and bombesin suppress operant responding for food reward. Pharmacol Biochem Behav 1985;22:893–5. [4] Bernardis L, Bellinger L. The lateral hypothalamic area revisited: ingestive behavior. Neurosci Biobehav Rev 1996;20(2):189 –287. [5] Bhakthavatsalam P, Leibowitz F. Morphine-elicited feeding: diurnal rhythm, circulating corticosterone and macronutrient selection. Pharmacol Biochem Behav 1985;24:911–7. [6] Billington C, Levine A, Morley J. Are peptides truly satiety agents? A method of testing for neurohumoral satiety effects. Am J Physiol 1983;245(6):920 – 6. [7] Blundell J. Serotonin manipulations and the structure of feeding behaviour. Appetite 1986;7:39 –56. [8] Bowers R, Herzog C. Bombesin effects on responding maintained by a fixed interval schedule of food reinforcement. Peptides 1991;12(6): 1435– 6. [9] Flynn F. Applications of taste reactivity to the study of the neuralhormonal controls of ingestive behavior. Neurosci Biobehav Rev 1995;19(1):109 –20. [10] Gibbs J, Smith G. The actions of bombesin-like peptides on food intake. Ann NY Acad Sci 1988;547:210 – 60. [11] Heinrichs S, Lapsansky J, Behan D, Chan R, Sawchenko P, Lorang M, Ling N, Vale W, DeSouza E. Corticotropin-releasing factorbinding protein ligand inhibitor blunts excessive weight gain in generally obese Zucker rats and rats during nicotine withdrawal. Proc Natl Acad Sci, Pharmacol 1996;93:15475– 80. [12] Krahn D, Gosnell B, Grace M, Levine A. CRF antagonist partially reverses CRF and stress induced effects on feeding. Brain Res Bull 1986;17(3):285–9. [13] Krahn D, Gosnell B, Levine A, Morley J. Behavioral effects of corticotropin-releasing factor: localization and characterization of central effects. Brain Res 1988;443:63–9. [14] Lee M, Schiffman S, Pappas T. Role of neuropeptides in the regulation of feeding behavior: a review of cholecystokinin, bombesin, neuropeptide Y, and galanin. Neurosci Biobehav Rev 1994;18(3): 313–23. [15] Lynch W, Babcock A. Effects of bombesin on temporal patterns of ingestion in the rat. Physiol Behav 1993;53(6):1223– 6. [16] Marks-Kaufman R. Increased fat consumption induced by morphine administration in rats. Pharmacol Biochem Behav 1982;16:949 –55. [17] Moreau J, Kilpatrick G, Jenck F. Urocortin, a novel neuropeptide with anxiogenic-like properties. NeuroReport 1997;8:1697–1701. [18] Schwartz D, Hoebel B. Effect of phenylpropanolamine on diet selection in rats. Pharmacol Biochem Behav 1988;31(3):721–3. [19] Shor-Posner G, Azar A, Insinga S, Leibowitz S. Deficits in the control of food intake after hypothalamic paraventricular nucleus lesions. Physiol Behav 1985;35(6):883–90. [20] Smagin G, Howell L, Ryan D, DeSouza E, Harris R. The role of CRF2 receptors in corticotropin-releasing factor- and urocortin-induced anorexia. NeuroReport 1998;9:1601– 6. [21] Spina M, Merlo-Pich E, Chan R, Basso A, Rivier J, Vale W, Koob G. Appetite-suppressing effects of urocortin, a CRF-related neuropeptide. Science 1996;273:1561– 4. [22] Thibault L, Nagai K, Hashida A, Yanaihara N, Nakagawa H. Satiation due to CCK-8 derivative infusion into VMH is related to a specific macronutrient selection. Physiol Behav 1990;47:911–5. [23] Vaughan J, Donaldson C, Bittencourt J, Perrin M, Lewis K, Sutton S, Chan R, Turnbull A, Lovejoy D, Rivier C, Rivier J, Sawchenko P, Vale W. Urocortin, a mammalian neuropeptide related to fish urotensin I and to corticotropin-releasing factor. Nature 1995;378:287– 92.