- Email: [email protected]
Changes in the Microstructure of Feeding After Administration of Enterostatin into the Paraventricular Nucleus and the Amygdala
Peptides, Vol. 19, No. 3, pp. 557–562, 1998 Copyright © 1998 Elsevier Science Inc. Printed in the USA. All rights reserved 0196-9781/98 $19.00 1 .00
PII S0196-9781(97)00460-9
Changes in the Microstructure of Feeding After Administration of Enterostatin into the Paraventricular Nucleus and the Amygdala LING LIN AND DAVID A. YORK1 Pennington Biomedical Research Center, Louisiana State University, Baton Rouge, LA 70808 Received 12 September 1997; Accepted 15 November 1997 LING, L. AND D. A. YORK. Changes in the microstructure of feeding after administration of enterostatin into the paraventricular nucleus and the amygdala. PEPTIDES 19(3) 557–562, 1998.—The effects of enterostatin (ENT) injected into the paraventricular nucleus (PVN) and the amygdala (AMYG) on the microstructure of feeding was studied by using an automated feeding apparatus. In rats adapted to a 6-h meal feeding regime, ENT reduced the size and duration of the first meal after injection in both the PVN and the AMYG. Similar effects were observed when ENT was given at the beginning of the dark cycle in rats fed ad libitum although the onset of feeding was also delayed in this situation. The number of meals and the size of subsequent meals was unaffected by ENT but the eating rate within the first meal was reduced after ENT injection into the AMYG of meal-fed rats. Enterostatin injected into the AMYG at a dose that suppressed feeding did not produce a conditioned taste aversion. ENT given centrally therefore appears to reduce food intake by delaying the initiation of feeding and/or advancing meal termination suggesting that it affects both appetite and satiation mechanisms. © 1998 Elsevier Science Inc. Enterostatin Satiety
Paraventricular nucleus
Amygdala
IT HAS been suggested that the pentapeptide enterostatin (ENT) may be a feedback signal to regulate the intake of dietary fat (10,33). Exogenous ENT selectively inhibits fat intake when given either peripherally or centrally to rats allowed to select diets or macronutrients (17,23). The peripheral mechanism involves a vagal signaling pathway, since either transection of the hepatic vagus or capsaicin treatment abolishes the effect of peripheral ENT on feeding and c-fos expression in specific brain nuclei (31,33). The central effects of enterostatin may be mediated through a pathway that involves both an opioid component and a serotonergic component, since both the opioid kappa1-agonist U50,488 and the 5HT antagonist metergoline block the ENT induced feeding suppression (1,20). Enterostatin is the activation peptide of procolipase, a protein which is synthesized in both the exocrine pancreas (9,10,33) and the gastric
Meal size
Conditioned taste aversion
mucosa (24). ENT concentrations in the circulation and the intestinal lumen are elevated after a high fat meal (5,21). Further colipase activity, a marker for enterostatin secretion, has been correlated with dietary fat consumption both within and between rat strains (23), high levels of activity being associated with low levels of fat ingestion. Although specific receptors for ENT have not yet been identified, mapping studies have shown that enterostatin is effective after local injection into the paraventricular nucleus (PVN) and the amygdala (AMYG), but not the ventromedial hypothalamus (VMH) or nucleus tractus solitarius (NTS) (19). The suppression of food intake induced by peripheral ENT has been linked to the promotion of an early satiety without other changes in the satiety sequence or other behaviors (18). To date there is no information on the behavioral effects of enterostatin administered directly into
1
Requests for reprints should be addressed to: David A. York, Pennington Biomedical Research Center, Louisiana State University, 6400 Perkins Road, Baton Rouge, LA 70808. E-mail: [email protected]
557
558
the CNS. The availability of automated recording techniques enables the continuous monitoring of feeding and subsequent analysis of the microstructure of the meal pattern, including the size and number of meals and the rate of ingestion. A number of anorexic and orexigenic compounds have been investigated for their effects on meal structure. Fenfluramine, cholecystokinin (CCK) and insulin reduce meal size without changing the meal frequency (12,29,32), opioid antagonists advance meal termination (13), whereas d-amphetamine delays the onset of the feeding and reduces meal frequency (12). In contrast NPY increases food intake by increasing the size and duration of meals (28). The primary aim of this study was to understand the effect of central ENT on the microstructure of feeding in rats. ENT was injected into the PVN and AMYG, two sites at which ENT is known to inhibit food intake. The possibility that ENT reduced food intake through the development of a conditioned taste aversion was also examined. METHOD Animals and Diet A total of 44 male Harlan Sprague–Dawley rats, body weight 235 6 2 grams at the beginning of the studies, were used in these experiments. Rats were housed in hanging stainless steel cages in a temperature controlled room (22– 23°C) with a 12/12 light/dark cycle (lights on at 0700 h). Rats were fed a high fat diet (56% of energy as fat, 4.78 kcal/g) ad lib during the initial 14-day adaptation period, after which 11 rats were adapted to a 6-h meal feeding paradigm in which food was only provided between 1000 – 1600 h daily and the remainign rats were maintained on ad lib feeding regimen. The composition of this diet has been described previously (17). Surgery Under pentobarbital (40 mg/kg body weight, IP) anesthesia, rats were stereotaxically implanted with a chronic, unilateral 26-gauge guide cannula made from stainless steel tubing (Small Parts, Inc., Miami Lakes, FL). The cannula tip was aimed at PVN and AMYG to allow the injector (silica capillary, O.D.150 mm, I.D. 40 mm) to extend 2.0 mm beyond the guide cannula to penetrate the tissue at the target site. The coordinates (in mm posterior to bregma/lateral to midsagittal/ventral to skull surface) were: PVN, 1.9/0.4/6.0 and AMYG, 2.4/3.8/6.0. Animals were allowed to recover from surgery for at least 7 days before experimentation. Only rats that regained their presurgical weight and normal food intake in their home cages and experimental cages (see below) were used. Peptide and Injections ENT was synthesized by the Core Laboratory at Louisiana State University Medical School (New Orleans, LA). The purity of ENT was at least 95% as established by HPLC and
LIN AND YORK
Mass Spectrometry. The peptide or artificial cerebrospinal fluid vehicle (aCSF, Harvard Apparatus, South Natick, MA) were administered in a volume of 0.1 ml over a 30-s period. Injections were given every 3 days alternating between vehicle or ENT treatment. The doses of ENT used (0.1 or 0.01 nmol) have been shown previously to suppress feeding after injection into the PVN or AMYG respectively (19). Testing Procedure The test chambers were rat cages identical to home cages that were modified to allow access to a slotted metal tunnel, under which the food cup was placed on a balance. The rats were acclimated to these chambers for at least 3 days before experimentation. Output generated from each balance was continually monitored by computer. Crumbs and spillage were caught by the balance platform and not recorded as ingested food. ENT or vehicle were administered immediately before the start of feeding in meal fed rats or immediately before lights off (1900 h) in ad lib fed animals. Conditioned Taste Aversion (CTA) Before the start of the CTA test, 22 rats maintained on the high fat diet ad lib were water deprived once to adapt them to the experimental protocol. Twelve rats had amygdala cannulas and ten rats without cannulas were used for intraperitoneal injections. On experimental day one, 18-h water deprived rats were provided with a novel 0.1% sodium saccharin solution (Sigma, St. Louis, MO) for 30 min in the absence of food and the consumption of saccharin solution was measured by weighing the bottles. The rats consumed 8.0 6 0.9 grams of saccharin solution. Immediately after saccharin exposure, ENT (0.01 nmol; n 5 6) or aCSF vehicle (n 5 6) were injected into the AMYG or LiCl (63 mg/kg body weight, Sigma Chemical Co., St. Louis, MO; n 5 5) or saline vehicle (n 5 5) were given intraperitoneally, after which food was provided. Food intake was measured to ensure the treatment effect. Three days later, CTA was assessed by measuring 3 h consumption of 0.1% sodium saccharin solution or tap water from the two available bottles after either ENT (0.01 nmol) or vehicle injection into the AMYG or peripheral injection (IP) of either LiCl or vehicle. The saccharin preference ratio was calculated as the amount of saccharin consumed divided by total consumption of both fluids. Histology After completion of all experimental trials, the rat brains were removed from anesthetized rats that had been transcardially perfused with 0.9% (w/v) saline followed by 10% (v/v) buffered formalin solution. One hundred mm thick frozen coronal sections were cut through the injection site and stained with Cresyl violet. The locations of the tips of the injection needles were determined according to the atlas of Paxinos and Watson (25). Animals in which the cannula
ENTEROSTATIN AND FEEDING BEHAVIOR
FIG. 1. Effects of ENT on (A) individual meal size (B) duration of the first meal (C) number of meals in 6 hour meal feeding rats. ENT was injected immediately before the food was presented in PVN (0.1 nmol; n 5 5 in each treatment) or AMYG (0.01 nmol; n 5 4). Data was presented as mean 6 SEM. *p , 0.05; **p , 0.01 compared to aCSF vehicle group.
placement was not within or on the dorsal surface of the PVN (n 5 3) or AMYG (n 5 2) were eliminated from further data analysis. Data Analysis All data are presented as mean 6 SEM. A meal was defined as an intake of at least 0.2 gram of food separated from the next eating episode by a 10 min (meal feeding paradigm) or 20 min (ad lib feeding) interval. The data were analyzed by Students’ t-test (two tail). Significant differences between means were defined by p , 0.05. RESULTS The effects of ENT administered into PVN and AMYG on cumulative food intake of overnight fasted rats have been
559
reported previously (19). The ENT doses used for the PVN and AMYG injections in this experiment (0.1 and 0.01 nmol respectively) were those shown to be optimal for inhibition in this previous study. 6-H meal feeding paradigm. Figure 1 shows the effects of ENT on the meal size, time spent eating the first meal and number of the meals in meal fed rats. The rats took three individual meals during the 6-h feeding period, the first meal being the largest in both vehicle treatment groups (PVN and AMYG) (Fig. 1A). ENT (0.1 nmol) injected into the PVN decreased the size of the first meal by 42%, whereas ENT (0.01 nmol) injected into the AMYG reduced the intake of the first meal by 62%. There were no differences in the size of subsequent meals between the ENT treatment and control groups for either PVN or AMYG site injections. The time spent eating the first meal was reduced by ENT treatment in both PVN and AMYG cannulated animals (Fig. 1B). All rats accessed the food immediately it became available after the injection. The rats spent 78 6 17 min or 42 6 1 min to eat the first meal after vehicle injections into the PVN and AMYG respectively, whereas after ENT treatment the times spent on the first meal were 37 6 3 minutes (PVN) and 25 6 2 minutes (AMYG) (Fig. 1B). There were no significant changes in the rate of eating within the first meal after ENT injection in the PVN but eating rate was reduced after ENT injection into the AMYG (Table 1). There was no difference in the number of meals between the treatment groups during the 6-h feeding period. Ad lib feeding during the dark period. The effects of ENT on the size of each individual meal are shown in Figure 2A. The size of the first meal was reduced by 61% by PVN ENT and by 60% in AMYG ENT groups compared with their respective vehicle treatment controls. Neither the size of subsequent meals nor the number of meals (Fig. 2C) was altered by ENT during the 12-h dark period. Both PVN and AMYG ENT significantly delayed the onset of feeding, increasing the latency to the onset of eating the first meal (Fig. 2B). Although the time spent eating the first meal was significantly decreased by ENT (Fig. 2B), the eating rate of the first meal was not affected (Table 1). TABLE 1 EATING RATE WITHIN FIRST MEAL (G/MIN)
PVN ad lib Meal fed AMYG ad lib Meal fed
aCSF
ENT
p
0.08 6 0.01 (4) 0.09 6 0.01 (5)
0.06 6 0.02 (4) 0.10 6 0.01 (5)
NS NS
0.05 6 0.01 (4) 0.21 6 0.02 (4)
0.04 6 0.01 (4) 0.13 6 0.01 (4)
NS ,0.05
Values represent means 6 SEM for the number of rats shown in parenthesis. NS, no statistically significant difference.
560
LIN AND YORK
decrease hunger. Our previous studies had shown that peripheral ENT reduced food intake through initiation of an early satiety response (18). In addition, the absence of a conditioned taste aversion to AMYG ENT observed in this study is consistent with another report where ENT was administered intracerebroventricularly (22). Neuropeptides and drugs may alter food intake through effects on various components of meal structure. Changes in the microstructure of feeding have been observed after food deprivation, vagotomy, capsaicin treatment, nucleus accumbens lesions and in genetically obese Zucker rats (6,7,8,15,27). A number of anorectic compounds produce different feeding patterns although they all reduce total caloric intake; CCK decreases the time spent eating (29), insulin reduces meal size (32), satietin increases the intermeal interval (2), nitric oxide reduces meal duration and meal number (27), amphetamine delays the onset of the feeding and reduces the meal frequency, whereas d-fenfluramine reduced the eating rate and meal size (12). The current experiments have shown that ENT not only decreased the time spent eating, but also reduced the size of the first meal immediately after injection without changing the size of following meals. ENT did not alter the number of meals or eating rate except after injection into the AMYG in meal-fed rats. Blundell et al. (3) defined satiation as the processes leading to the natural termination of eating, and satiety as the state after complete satiation that is inferred by measures such as intermeal interval (3,4). In contrast, abnormal inhibition of eating, such as that induced by nausea or a conditioned taste aversion, decreases eating rate and the number of meals without changing the duration or size of FIG. 2. Effects of ENT on (A) individual meal size (B) latency to first meal and duration of the first meal (C) number of meals in ad lib fed rat in 12-h dark cycle. ENT was injected before lights off. The doses were same as described in Fig. 1. n 5 4 for PVN and n 5 4 for AMYG. *p , 0.05; **p , 0.01 compared to aCSF vehicle group.
Conditioned taste aversion (CTA). The saccharin preference ratios for ENT and LiCl are presented in Figure 3. The preference ratios between AMYG ENT and vehicle injected rats were not significantly different, whereas the animals that received the intraperitoneal LiCl injection significantly reduced the preference (0.12 6 0.03) for saccharin compared to vehicle injection (0.71 6 0.05, p , 0.001). DISCUSSION The data presented here clearly demonstrate that feeding suppression caused by acute central injection of ENT at local sites resulted from decreasing the size and duration of the first meal in both meal fed and ad lib fed rats and delayed the onset of the feeding in ad lib fed rats. It suggests that ENT acts to promote both the state of satiety and to
FIG. 3. Effects of ENT on the saccharin preference ratio. Preference ratio was defined as consumption of 0.1% saccharin solution divided by total fluid (saccharin plus water). ENT at dose of 0.01 nmol (n 5 6 in each group) injected into AMYG while LiCl (63 mg/kg, n 5 5 in each group) was given IP. ***p , 0.001 compared to the saline vehicle.
ENTEROSTATIN AND FEEDING BEHAVIOR
561
the meals. The ENT effects on meal structure described in this paper fit with those proposed by Blundell for naturally occurring factors that affect the process of satiation. Inhibitory effects on food ingestion induced by a peptide or drug can reflect direct actions on the pathways regulating food intake, can result from the induction of alternative behaviors or can reflect the induction of sickness, nausea or a conditioned taste aversion. Thus the inhibitory effect of corticotropin releasing hormone on food intake has been associated with an enhancement of grooming activity (14) while glucagon-like peptide-l induces a conditioned taste aversion (CTA) that might be responsible for its anorectic properties (30). In contrast the inhibition of feeding induced by leptin is not linked to other behaviors or to a CTA (30). The observation that the AMYG was very responsive to ENT and that ENT may reduce the eating rate prompted concern that it might induce a CTA at this site. Although a previous report had suggested that intracerebroventricular ENT did not induce a CTA that could account for the inhibition of food intake, the association of CTAs with amygdala activity (11,26) prompted the further investigation of this possibility in the current experiments. The current experiments rule out this possibility since ENT, at a dose that inhibits feeding, did not induce any aversion as measured in the 2 bottle saccharin preference paradigm. In contrast, peripheral administration of LiCl induced the classical aversion response. We have reported previously that peripheral ENT induced an early state of satiety (18). This was also observed in the current experiments after central administration of ENT. However, ENT also delayed the onset of the feeding in normally feeding rats at the start of the dark cycle suggesting that ENT may also suppress hunger in addition to promoting satiety. It is possible that the drive to eat was so great in the 6-h meal-feeding rats that this effect was not
observed in that experimental paradigm. Although acute injection of enterostatin only reduced the size of the initial meal, studies in which enterostatin have been chronically infused have shown that ICV ENT does reduce food intake and fat selection over a prolonged period (17). This suggests that tonic levels of ENT have effects on all meals to suppress the appetite for dietary fat and reduce the daily level of fat ingestion. It is possible that the single bolus injection of ENT only effected the size of the first meal because ENT has a relatively short half-life. No direct measures of enterostatin half-life have yet been made to confirm this hypothesis. To date, the central mechanism through which ENT affects feeding has not been clearly defined. Both kappa1opioidergic and serotonergic components appear to be involved in the pathway through which enterostatin mediates its response (1,20). We have shown that ENT changes serotonin turnover in certain brain areas and that the 5-HT antagonist, metergoline, partially blocks the feeding inhibition by ENT (20). Some serotonergic agents have a similar profile to ENT on the microstructure of individual meals. Metergoline completely reverses the increased latency to feed and partially reverses the reduced meal size induced by the serotonergic agent fluoxetine (16). To our knowledge there is no data on the effects of kappa1-opioid agonists or antagonists on meal microstructure. In conclusion, the present experiments suggest that the inhibitory effects of acute central injections of ENT result from both suppression of hunger and the early termination of the first meal and are not related to induction of a conditioned taste aversion. ACKNOWLEDGEMENTS The authors gratefully acknowledge the help of Ms. Maudrie Eldridge in preparing the manuscript. Supported by NIH Grant NIDDK 45278.
REFERENCES 1. Barton, C.; Lin L.; York, D. A.; Bray, G. A. Differential effects of enterostatin, galanin and opioids on high-fat diet consumption. Brain Res. 702:55– 60; 1995. 2. Bellinger, L. L.; Mendel, V. E. Blood profile and balance study of rats given the putative anorectic agent satietin. Am. J. Physiol. 268:R1–7; 1995. 3. Blundell, J. E.; Rogers, P. J. Behavioural structure and mechanisms of anorexia: calibration of natural and abnormal inhibition of eating. Brain Res. Bull. 15:371–376; 1985. 4. Blundell, J. E. Pharmacological approaches to appetite suppression. Trends Pharmacol. Sci. 12:147–157; 1991. 5. Bowyer, R. C.; Rowston, W. M.; Jehanli, A. M. T.; Lacey, J. H.; Hermon–Taylor, J. Effect of a satiating meal on the concentrations of procolipase propeptide in the serum and urine of normal and morbidly obese subjects. Gut. 34:1520 – 1525; 1993. 6. Castonguay, T. W.; Bellinger, L. L. Capsaicin and its effects upon meal patterns, and glucagon and epinephrine suppression of food intake. Physiol. Behav. 40:337–342; 1987. 7. Clifton, P. G.; Somerville, E. M. Disturbance of meal pattern-
8.
9. 10. 11.
12.
ing following nucleus accumbens lesions in the rat. Brain Res. 667:123–128; 1994. Erecius, L. F.; Dixon, K. D.; Jiang, J. C.; Gietzen, D. W. Meal patterns reveal differential effects of vagotomy and tropisetron on responses to indispensable amino acid deficiency in rats. J. Nutr. 126:1722–1731; 1996. Erlanson–Albertsson, C.; Mei, J.; Okada, S.; York, D.; Bray, G. A. Pancreatic procolipase propeptide, enterostatin, specifically inhibits fat intake. Physiol. Behav. 49:1191–1194; 1991. Erlanson–Albertsson, C.; York, D. A. Enterostatin—A peptide regulating fat intake. Obesity Res. 5:300 –372; 1997. Galaverna, O. G.; Seeley, R. J.; Berridge, K. C.; Grill, H. J.; Epstein, A. N.; Schulkin, J. Lesions of the central nucleus of the amygdala. I: Effects on taste reactivity, taste aversion learning and sodium appetite. Behav. Brain Res. 59:11–17; 1993. Grinker, J. A.; Drewnowski, A.; Enns, M.; Kissileff, H. Effects of d-amphetamine and fenfluramine on feeding patterns and activity of obese and lean Zucker rats. Pharmacol. Biochem. Behav. 12:265–275; 1980.
562 13. Kirkham T. C.; Blundell, J. E. Dual action of naloxone on feeding revealed by behavioural analysis: separate effects on initiation and termination of eating. Appetite. 5:45–52; 1984. 14. Krahn, D. D.; Gosnell, B. A.; Levine, A. S.; Morley, J. E. Behavioral effects of corticotropin-releasing factor: localization and characterization of central effects. Brain Res. 443: 63– 69; 1988. 15. Larue-Achagiotis, C.; Le Magnen, J. Changes of meal patterns induced by food deprivation: metabolic correlates. Neurosci. Biobehav. Rev. 4:25–27; 1980. 16. Lee, M. D.; Clifton, P. G. Partial reversal of fluoxetine anorexia by the 5-HT antagonist metergoline. Psychopharmacology. 107:359 –364; 1992. 17. Lin, L.; Chen, J.; York, D. A. Chronic ICV enterostatin preferentially reduced fat intake and lowered body weight. Peptides. 18:657– 661; 1997. 18. Lin, L.; McClanahan, S.; York, D. A.; Bray, G. A. The peptide enterostatin may produce early satiety. Physiol. Behav. 53: 789 –794; 1993. 19. Lin, L.; York, D. A. Enterostatin actions in the amygdala and PVN to suppress feeding in the rat. Peptides. 18:1341–1347; 1997. 20. Lin, L. and York, D. A. Metergoline blocks the feeding response to enterostatin. Appetite. 23:313; 1994. 21. Mei, J.; Bowyer, R. C.; Jehanli, A. M.; Patel, G.; Erlanson– Albertsson, C. Identification of enterostatin, the pancreatic procolipase activation peptide in the intestine of rat: effect of CCK-8 and high-fat feeding. Pancreas. 8:488 – 493; 1993. 22. Mei, J.; Erlanson–Albertsson, C. Effect of enterostatin given intravenously and intracerebroventricularly on high-fat feeding in rats. Regul. Pept. 41:209 –218; 1992. 23. Okada, S.; York, D. A.; Bray, G. A.; Erlanson–Albertsson, C. Differential inhibition of fat intake in two strains of rat by the peptide enterostatin. Am. J. Physiol. 262:R1111–R1116; 1992.
LIN AND YORK 24. Okada, S.; York, D. A.; Bray, G. A. Procolipase mRNA: tissue localization and effects of diet and adrenalectomy. Biochem. J. 292:787–789; 1993. 25. Paxinos, G.; Watson C. The rat brain in stereotaxic coordinates. Sydney: Academic Press, 1982. 26. Roldan, G.; Bures, J. Tetrodotoxin blockade of amygdala overlapping with poisoning impairs acquisition of conditioned taste aversion in rats. Behav. Brain Res. 65:213–219; 1994. 27. Stricker–Krongrad, A.; Beck, B.; Burlet, C. Nitric oxide mediates hyperphagia of obese Zucker rats: relation to specific changes in the microstructure of feeding behavior. Life Sci. 58:PL9 –15; 1996. 28. Stricker–Krongrad, A.; Max, J. P.; Musse, N.; Nicolas, J. P.; Burlet, C.; Beck, B. Increased threshold concentrations of neuropeptide Y for a stimulatory effect on food intake in obese Zucker rats— changes in the microstructure of the feeding behavior. Brain Res. 660:162–166; 1994. 29. Strohmayer, A. J.; Greenberg, D. Devazepide alters meal patterns in lean, but not obese, male Zucker rats. Physiol. Behav. 56:1037–1039; 1994. 30. Thiele, T. E.; Van Dijk, G.; Campfield, L. A.; Smith, F. J.; Burn, P.; Woods, S. C.; Bernstein, I. L.; Seeley, R. J. Central infusion of GLP-1, but not leptin, produces conditioned taste aversions in rats. Am. J. Physiol. 272:R726 –30; 1997. 31. Tian, Q.; Nagase, M.; York, D. A.; Bray, G. A. Vagal-central nervous system interactions modulate the feeding response to peripheral enterostatin. Obesity Res. 2:527–534; 1994. 32. VanderWeele, D. A. Insulin is a prandial satiety hormone. Physiol. Behav. 56:619 – 622; 1994. 33. York, D. A.; Lin, L. Enterostatin and fat intake: current insights into possible mechanisms. In: Angel, A.; Anderson, H.; Bouchard, C.; et al., Eds. Progress in obesity research, Vol. 7, New York: John Libbey & Company Ltd.; 1996;495–501.
PII S0196-9781(97)00460-9
Changes in the Microstructure of Feeding After Administration of Enterostatin into the Paraventricular Nucleus and the Amygdala LING LIN AND DAVID A. YORK1 Pennington Biomedical Research Center, Louisiana State University, Baton Rouge, LA 70808 Received 12 September 1997; Accepted 15 November 1997 LING, L. AND D. A. YORK. Changes in the microstructure of feeding after administration of enterostatin into the paraventricular nucleus and the amygdala. PEPTIDES 19(3) 557–562, 1998.—The effects of enterostatin (ENT) injected into the paraventricular nucleus (PVN) and the amygdala (AMYG) on the microstructure of feeding was studied by using an automated feeding apparatus. In rats adapted to a 6-h meal feeding regime, ENT reduced the size and duration of the first meal after injection in both the PVN and the AMYG. Similar effects were observed when ENT was given at the beginning of the dark cycle in rats fed ad libitum although the onset of feeding was also delayed in this situation. The number of meals and the size of subsequent meals was unaffected by ENT but the eating rate within the first meal was reduced after ENT injection into the AMYG of meal-fed rats. Enterostatin injected into the AMYG at a dose that suppressed feeding did not produce a conditioned taste aversion. ENT given centrally therefore appears to reduce food intake by delaying the initiation of feeding and/or advancing meal termination suggesting that it affects both appetite and satiation mechanisms. © 1998 Elsevier Science Inc. Enterostatin Satiety
Paraventricular nucleus
Amygdala
IT HAS been suggested that the pentapeptide enterostatin (ENT) may be a feedback signal to regulate the intake of dietary fat (10,33). Exogenous ENT selectively inhibits fat intake when given either peripherally or centrally to rats allowed to select diets or macronutrients (17,23). The peripheral mechanism involves a vagal signaling pathway, since either transection of the hepatic vagus or capsaicin treatment abolishes the effect of peripheral ENT on feeding and c-fos expression in specific brain nuclei (31,33). The central effects of enterostatin may be mediated through a pathway that involves both an opioid component and a serotonergic component, since both the opioid kappa1-agonist U50,488 and the 5HT antagonist metergoline block the ENT induced feeding suppression (1,20). Enterostatin is the activation peptide of procolipase, a protein which is synthesized in both the exocrine pancreas (9,10,33) and the gastric
Meal size
Conditioned taste aversion
mucosa (24). ENT concentrations in the circulation and the intestinal lumen are elevated after a high fat meal (5,21). Further colipase activity, a marker for enterostatin secretion, has been correlated with dietary fat consumption both within and between rat strains (23), high levels of activity being associated with low levels of fat ingestion. Although specific receptors for ENT have not yet been identified, mapping studies have shown that enterostatin is effective after local injection into the paraventricular nucleus (PVN) and the amygdala (AMYG), but not the ventromedial hypothalamus (VMH) or nucleus tractus solitarius (NTS) (19). The suppression of food intake induced by peripheral ENT has been linked to the promotion of an early satiety without other changes in the satiety sequence or other behaviors (18). To date there is no information on the behavioral effects of enterostatin administered directly into
1
Requests for reprints should be addressed to: David A. York, Pennington Biomedical Research Center, Louisiana State University, 6400 Perkins Road, Baton Rouge, LA 70808. E-mail: [email protected]
557
558
the CNS. The availability of automated recording techniques enables the continuous monitoring of feeding and subsequent analysis of the microstructure of the meal pattern, including the size and number of meals and the rate of ingestion. A number of anorexic and orexigenic compounds have been investigated for their effects on meal structure. Fenfluramine, cholecystokinin (CCK) and insulin reduce meal size without changing the meal frequency (12,29,32), opioid antagonists advance meal termination (13), whereas d-amphetamine delays the onset of the feeding and reduces meal frequency (12). In contrast NPY increases food intake by increasing the size and duration of meals (28). The primary aim of this study was to understand the effect of central ENT on the microstructure of feeding in rats. ENT was injected into the PVN and AMYG, two sites at which ENT is known to inhibit food intake. The possibility that ENT reduced food intake through the development of a conditioned taste aversion was also examined. METHOD Animals and Diet A total of 44 male Harlan Sprague–Dawley rats, body weight 235 6 2 grams at the beginning of the studies, were used in these experiments. Rats were housed in hanging stainless steel cages in a temperature controlled room (22– 23°C) with a 12/12 light/dark cycle (lights on at 0700 h). Rats were fed a high fat diet (56% of energy as fat, 4.78 kcal/g) ad lib during the initial 14-day adaptation period, after which 11 rats were adapted to a 6-h meal feeding paradigm in which food was only provided between 1000 – 1600 h daily and the remainign rats were maintained on ad lib feeding regimen. The composition of this diet has been described previously (17). Surgery Under pentobarbital (40 mg/kg body weight, IP) anesthesia, rats were stereotaxically implanted with a chronic, unilateral 26-gauge guide cannula made from stainless steel tubing (Small Parts, Inc., Miami Lakes, FL). The cannula tip was aimed at PVN and AMYG to allow the injector (silica capillary, O.D.150 mm, I.D. 40 mm) to extend 2.0 mm beyond the guide cannula to penetrate the tissue at the target site. The coordinates (in mm posterior to bregma/lateral to midsagittal/ventral to skull surface) were: PVN, 1.9/0.4/6.0 and AMYG, 2.4/3.8/6.0. Animals were allowed to recover from surgery for at least 7 days before experimentation. Only rats that regained their presurgical weight and normal food intake in their home cages and experimental cages (see below) were used. Peptide and Injections ENT was synthesized by the Core Laboratory at Louisiana State University Medical School (New Orleans, LA). The purity of ENT was at least 95% as established by HPLC and
LIN AND YORK
Mass Spectrometry. The peptide or artificial cerebrospinal fluid vehicle (aCSF, Harvard Apparatus, South Natick, MA) were administered in a volume of 0.1 ml over a 30-s period. Injections were given every 3 days alternating between vehicle or ENT treatment. The doses of ENT used (0.1 or 0.01 nmol) have been shown previously to suppress feeding after injection into the PVN or AMYG respectively (19). Testing Procedure The test chambers were rat cages identical to home cages that were modified to allow access to a slotted metal tunnel, under which the food cup was placed on a balance. The rats were acclimated to these chambers for at least 3 days before experimentation. Output generated from each balance was continually monitored by computer. Crumbs and spillage were caught by the balance platform and not recorded as ingested food. ENT or vehicle were administered immediately before the start of feeding in meal fed rats or immediately before lights off (1900 h) in ad lib fed animals. Conditioned Taste Aversion (CTA) Before the start of the CTA test, 22 rats maintained on the high fat diet ad lib were water deprived once to adapt them to the experimental protocol. Twelve rats had amygdala cannulas and ten rats without cannulas were used for intraperitoneal injections. On experimental day one, 18-h water deprived rats were provided with a novel 0.1% sodium saccharin solution (Sigma, St. Louis, MO) for 30 min in the absence of food and the consumption of saccharin solution was measured by weighing the bottles. The rats consumed 8.0 6 0.9 grams of saccharin solution. Immediately after saccharin exposure, ENT (0.01 nmol; n 5 6) or aCSF vehicle (n 5 6) were injected into the AMYG or LiCl (63 mg/kg body weight, Sigma Chemical Co., St. Louis, MO; n 5 5) or saline vehicle (n 5 5) were given intraperitoneally, after which food was provided. Food intake was measured to ensure the treatment effect. Three days later, CTA was assessed by measuring 3 h consumption of 0.1% sodium saccharin solution or tap water from the two available bottles after either ENT (0.01 nmol) or vehicle injection into the AMYG or peripheral injection (IP) of either LiCl or vehicle. The saccharin preference ratio was calculated as the amount of saccharin consumed divided by total consumption of both fluids. Histology After completion of all experimental trials, the rat brains were removed from anesthetized rats that had been transcardially perfused with 0.9% (w/v) saline followed by 10% (v/v) buffered formalin solution. One hundred mm thick frozen coronal sections were cut through the injection site and stained with Cresyl violet. The locations of the tips of the injection needles were determined according to the atlas of Paxinos and Watson (25). Animals in which the cannula
ENTEROSTATIN AND FEEDING BEHAVIOR
FIG. 1. Effects of ENT on (A) individual meal size (B) duration of the first meal (C) number of meals in 6 hour meal feeding rats. ENT was injected immediately before the food was presented in PVN (0.1 nmol; n 5 5 in each treatment) or AMYG (0.01 nmol; n 5 4). Data was presented as mean 6 SEM. *p , 0.05; **p , 0.01 compared to aCSF vehicle group.
placement was not within or on the dorsal surface of the PVN (n 5 3) or AMYG (n 5 2) were eliminated from further data analysis. Data Analysis All data are presented as mean 6 SEM. A meal was defined as an intake of at least 0.2 gram of food separated from the next eating episode by a 10 min (meal feeding paradigm) or 20 min (ad lib feeding) interval. The data were analyzed by Students’ t-test (two tail). Significant differences between means were defined by p , 0.05. RESULTS The effects of ENT administered into PVN and AMYG on cumulative food intake of overnight fasted rats have been
559
reported previously (19). The ENT doses used for the PVN and AMYG injections in this experiment (0.1 and 0.01 nmol respectively) were those shown to be optimal for inhibition in this previous study. 6-H meal feeding paradigm. Figure 1 shows the effects of ENT on the meal size, time spent eating the first meal and number of the meals in meal fed rats. The rats took three individual meals during the 6-h feeding period, the first meal being the largest in both vehicle treatment groups (PVN and AMYG) (Fig. 1A). ENT (0.1 nmol) injected into the PVN decreased the size of the first meal by 42%, whereas ENT (0.01 nmol) injected into the AMYG reduced the intake of the first meal by 62%. There were no differences in the size of subsequent meals between the ENT treatment and control groups for either PVN or AMYG site injections. The time spent eating the first meal was reduced by ENT treatment in both PVN and AMYG cannulated animals (Fig. 1B). All rats accessed the food immediately it became available after the injection. The rats spent 78 6 17 min or 42 6 1 min to eat the first meal after vehicle injections into the PVN and AMYG respectively, whereas after ENT treatment the times spent on the first meal were 37 6 3 minutes (PVN) and 25 6 2 minutes (AMYG) (Fig. 1B). There were no significant changes in the rate of eating within the first meal after ENT injection in the PVN but eating rate was reduced after ENT injection into the AMYG (Table 1). There was no difference in the number of meals between the treatment groups during the 6-h feeding period. Ad lib feeding during the dark period. The effects of ENT on the size of each individual meal are shown in Figure 2A. The size of the first meal was reduced by 61% by PVN ENT and by 60% in AMYG ENT groups compared with their respective vehicle treatment controls. Neither the size of subsequent meals nor the number of meals (Fig. 2C) was altered by ENT during the 12-h dark period. Both PVN and AMYG ENT significantly delayed the onset of feeding, increasing the latency to the onset of eating the first meal (Fig. 2B). Although the time spent eating the first meal was significantly decreased by ENT (Fig. 2B), the eating rate of the first meal was not affected (Table 1). TABLE 1 EATING RATE WITHIN FIRST MEAL (G/MIN)
PVN ad lib Meal fed AMYG ad lib Meal fed
aCSF
ENT
p
0.08 6 0.01 (4) 0.09 6 0.01 (5)
0.06 6 0.02 (4) 0.10 6 0.01 (5)
NS NS
0.05 6 0.01 (4) 0.21 6 0.02 (4)
0.04 6 0.01 (4) 0.13 6 0.01 (4)
NS ,0.05
Values represent means 6 SEM for the number of rats shown in parenthesis. NS, no statistically significant difference.
560
LIN AND YORK
decrease hunger. Our previous studies had shown that peripheral ENT reduced food intake through initiation of an early satiety response (18). In addition, the absence of a conditioned taste aversion to AMYG ENT observed in this study is consistent with another report where ENT was administered intracerebroventricularly (22). Neuropeptides and drugs may alter food intake through effects on various components of meal structure. Changes in the microstructure of feeding have been observed after food deprivation, vagotomy, capsaicin treatment, nucleus accumbens lesions and in genetically obese Zucker rats (6,7,8,15,27). A number of anorectic compounds produce different feeding patterns although they all reduce total caloric intake; CCK decreases the time spent eating (29), insulin reduces meal size (32), satietin increases the intermeal interval (2), nitric oxide reduces meal duration and meal number (27), amphetamine delays the onset of the feeding and reduces the meal frequency, whereas d-fenfluramine reduced the eating rate and meal size (12). The current experiments have shown that ENT not only decreased the time spent eating, but also reduced the size of the first meal immediately after injection without changing the size of following meals. ENT did not alter the number of meals or eating rate except after injection into the AMYG in meal-fed rats. Blundell et al. (3) defined satiation as the processes leading to the natural termination of eating, and satiety as the state after complete satiation that is inferred by measures such as intermeal interval (3,4). In contrast, abnormal inhibition of eating, such as that induced by nausea or a conditioned taste aversion, decreases eating rate and the number of meals without changing the duration or size of FIG. 2. Effects of ENT on (A) individual meal size (B) latency to first meal and duration of the first meal (C) number of meals in ad lib fed rat in 12-h dark cycle. ENT was injected before lights off. The doses were same as described in Fig. 1. n 5 4 for PVN and n 5 4 for AMYG. *p , 0.05; **p , 0.01 compared to aCSF vehicle group.
Conditioned taste aversion (CTA). The saccharin preference ratios for ENT and LiCl are presented in Figure 3. The preference ratios between AMYG ENT and vehicle injected rats were not significantly different, whereas the animals that received the intraperitoneal LiCl injection significantly reduced the preference (0.12 6 0.03) for saccharin compared to vehicle injection (0.71 6 0.05, p , 0.001). DISCUSSION The data presented here clearly demonstrate that feeding suppression caused by acute central injection of ENT at local sites resulted from decreasing the size and duration of the first meal in both meal fed and ad lib fed rats and delayed the onset of the feeding in ad lib fed rats. It suggests that ENT acts to promote both the state of satiety and to
FIG. 3. Effects of ENT on the saccharin preference ratio. Preference ratio was defined as consumption of 0.1% saccharin solution divided by total fluid (saccharin plus water). ENT at dose of 0.01 nmol (n 5 6 in each group) injected into AMYG while LiCl (63 mg/kg, n 5 5 in each group) was given IP. ***p , 0.001 compared to the saline vehicle.
ENTEROSTATIN AND FEEDING BEHAVIOR
561
the meals. The ENT effects on meal structure described in this paper fit with those proposed by Blundell for naturally occurring factors that affect the process of satiation. Inhibitory effects on food ingestion induced by a peptide or drug can reflect direct actions on the pathways regulating food intake, can result from the induction of alternative behaviors or can reflect the induction of sickness, nausea or a conditioned taste aversion. Thus the inhibitory effect of corticotropin releasing hormone on food intake has been associated with an enhancement of grooming activity (14) while glucagon-like peptide-l induces a conditioned taste aversion (CTA) that might be responsible for its anorectic properties (30). In contrast the inhibition of feeding induced by leptin is not linked to other behaviors or to a CTA (30). The observation that the AMYG was very responsive to ENT and that ENT may reduce the eating rate prompted concern that it might induce a CTA at this site. Although a previous report had suggested that intracerebroventricular ENT did not induce a CTA that could account for the inhibition of food intake, the association of CTAs with amygdala activity (11,26) prompted the further investigation of this possibility in the current experiments. The current experiments rule out this possibility since ENT, at a dose that inhibits feeding, did not induce any aversion as measured in the 2 bottle saccharin preference paradigm. In contrast, peripheral administration of LiCl induced the classical aversion response. We have reported previously that peripheral ENT induced an early state of satiety (18). This was also observed in the current experiments after central administration of ENT. However, ENT also delayed the onset of the feeding in normally feeding rats at the start of the dark cycle suggesting that ENT may also suppress hunger in addition to promoting satiety. It is possible that the drive to eat was so great in the 6-h meal-feeding rats that this effect was not
observed in that experimental paradigm. Although acute injection of enterostatin only reduced the size of the initial meal, studies in which enterostatin have been chronically infused have shown that ICV ENT does reduce food intake and fat selection over a prolonged period (17). This suggests that tonic levels of ENT have effects on all meals to suppress the appetite for dietary fat and reduce the daily level of fat ingestion. It is possible that the single bolus injection of ENT only effected the size of the first meal because ENT has a relatively short half-life. No direct measures of enterostatin half-life have yet been made to confirm this hypothesis. To date, the central mechanism through which ENT affects feeding has not been clearly defined. Both kappa1opioidergic and serotonergic components appear to be involved in the pathway through which enterostatin mediates its response (1,20). We have shown that ENT changes serotonin turnover in certain brain areas and that the 5-HT antagonist, metergoline, partially blocks the feeding inhibition by ENT (20). Some serotonergic agents have a similar profile to ENT on the microstructure of individual meals. Metergoline completely reverses the increased latency to feed and partially reverses the reduced meal size induced by the serotonergic agent fluoxetine (16). To our knowledge there is no data on the effects of kappa1-opioid agonists or antagonists on meal microstructure. In conclusion, the present experiments suggest that the inhibitory effects of acute central injections of ENT result from both suppression of hunger and the early termination of the first meal and are not related to induction of a conditioned taste aversion. ACKNOWLEDGEMENTS The authors gratefully acknowledge the help of Ms. Maudrie Eldridge in preparing the manuscript. Supported by NIH Grant NIDDK 45278.
REFERENCES 1. Barton, C.; Lin L.; York, D. A.; Bray, G. A. Differential effects of enterostatin, galanin and opioids on high-fat diet consumption. Brain Res. 702:55– 60; 1995. 2. Bellinger, L. L.; Mendel, V. E. Blood profile and balance study of rats given the putative anorectic agent satietin. Am. J. Physiol. 268:R1–7; 1995. 3. Blundell, J. E.; Rogers, P. J. Behavioural structure and mechanisms of anorexia: calibration of natural and abnormal inhibition of eating. Brain Res. Bull. 15:371–376; 1985. 4. Blundell, J. E. Pharmacological approaches to appetite suppression. Trends Pharmacol. Sci. 12:147–157; 1991. 5. Bowyer, R. C.; Rowston, W. M.; Jehanli, A. M. T.; Lacey, J. H.; Hermon–Taylor, J. Effect of a satiating meal on the concentrations of procolipase propeptide in the serum and urine of normal and morbidly obese subjects. Gut. 34:1520 – 1525; 1993. 6. Castonguay, T. W.; Bellinger, L. L. Capsaicin and its effects upon meal patterns, and glucagon and epinephrine suppression of food intake. Physiol. Behav. 40:337–342; 1987. 7. Clifton, P. G.; Somerville, E. M. Disturbance of meal pattern-
8.
9. 10. 11.
12.
ing following nucleus accumbens lesions in the rat. Brain Res. 667:123–128; 1994. Erecius, L. F.; Dixon, K. D.; Jiang, J. C.; Gietzen, D. W. Meal patterns reveal differential effects of vagotomy and tropisetron on responses to indispensable amino acid deficiency in rats. J. Nutr. 126:1722–1731; 1996. Erlanson–Albertsson, C.; Mei, J.; Okada, S.; York, D.; Bray, G. A. Pancreatic procolipase propeptide, enterostatin, specifically inhibits fat intake. Physiol. Behav. 49:1191–1194; 1991. Erlanson–Albertsson, C.; York, D. A. Enterostatin—A peptide regulating fat intake. Obesity Res. 5:300 –372; 1997. Galaverna, O. G.; Seeley, R. J.; Berridge, K. C.; Grill, H. J.; Epstein, A. N.; Schulkin, J. Lesions of the central nucleus of the amygdala. I: Effects on taste reactivity, taste aversion learning and sodium appetite. Behav. Brain Res. 59:11–17; 1993. Grinker, J. A.; Drewnowski, A.; Enns, M.; Kissileff, H. Effects of d-amphetamine and fenfluramine on feeding patterns and activity of obese and lean Zucker rats. Pharmacol. Biochem. Behav. 12:265–275; 1980.
562 13. Kirkham T. C.; Blundell, J. E. Dual action of naloxone on feeding revealed by behavioural analysis: separate effects on initiation and termination of eating. Appetite. 5:45–52; 1984. 14. Krahn, D. D.; Gosnell, B. A.; Levine, A. S.; Morley, J. E. Behavioral effects of corticotropin-releasing factor: localization and characterization of central effects. Brain Res. 443: 63– 69; 1988. 15. Larue-Achagiotis, C.; Le Magnen, J. Changes of meal patterns induced by food deprivation: metabolic correlates. Neurosci. Biobehav. Rev. 4:25–27; 1980. 16. Lee, M. D.; Clifton, P. G. Partial reversal of fluoxetine anorexia by the 5-HT antagonist metergoline. Psychopharmacology. 107:359 –364; 1992. 17. Lin, L.; Chen, J.; York, D. A. Chronic ICV enterostatin preferentially reduced fat intake and lowered body weight. Peptides. 18:657– 661; 1997. 18. Lin, L.; McClanahan, S.; York, D. A.; Bray, G. A. The peptide enterostatin may produce early satiety. Physiol. Behav. 53: 789 –794; 1993. 19. Lin, L.; York, D. A. Enterostatin actions in the amygdala and PVN to suppress feeding in the rat. Peptides. 18:1341–1347; 1997. 20. Lin, L. and York, D. A. Metergoline blocks the feeding response to enterostatin. Appetite. 23:313; 1994. 21. Mei, J.; Bowyer, R. C.; Jehanli, A. M.; Patel, G.; Erlanson– Albertsson, C. Identification of enterostatin, the pancreatic procolipase activation peptide in the intestine of rat: effect of CCK-8 and high-fat feeding. Pancreas. 8:488 – 493; 1993. 22. Mei, J.; Erlanson–Albertsson, C. Effect of enterostatin given intravenously and intracerebroventricularly on high-fat feeding in rats. Regul. Pept. 41:209 –218; 1992. 23. Okada, S.; York, D. A.; Bray, G. A.; Erlanson–Albertsson, C. Differential inhibition of fat intake in two strains of rat by the peptide enterostatin. Am. J. Physiol. 262:R1111–R1116; 1992.
LIN AND YORK 24. Okada, S.; York, D. A.; Bray, G. A. Procolipase mRNA: tissue localization and effects of diet and adrenalectomy. Biochem. J. 292:787–789; 1993. 25. Paxinos, G.; Watson C. The rat brain in stereotaxic coordinates. Sydney: Academic Press, 1982. 26. Roldan, G.; Bures, J. Tetrodotoxin blockade of amygdala overlapping with poisoning impairs acquisition of conditioned taste aversion in rats. Behav. Brain Res. 65:213–219; 1994. 27. Stricker–Krongrad, A.; Beck, B.; Burlet, C. Nitric oxide mediates hyperphagia of obese Zucker rats: relation to specific changes in the microstructure of feeding behavior. Life Sci. 58:PL9 –15; 1996. 28. Stricker–Krongrad, A.; Max, J. P.; Musse, N.; Nicolas, J. P.; Burlet, C.; Beck, B. Increased threshold concentrations of neuropeptide Y for a stimulatory effect on food intake in obese Zucker rats— changes in the microstructure of the feeding behavior. Brain Res. 660:162–166; 1994. 29. Strohmayer, A. J.; Greenberg, D. Devazepide alters meal patterns in lean, but not obese, male Zucker rats. Physiol. Behav. 56:1037–1039; 1994. 30. Thiele, T. E.; Van Dijk, G.; Campfield, L. A.; Smith, F. J.; Burn, P.; Woods, S. C.; Bernstein, I. L.; Seeley, R. J. Central infusion of GLP-1, but not leptin, produces conditioned taste aversions in rats. Am. J. Physiol. 272:R726 –30; 1997. 31. Tian, Q.; Nagase, M.; York, D. A.; Bray, G. A. Vagal-central nervous system interactions modulate the feeding response to peripheral enterostatin. Obesity Res. 2:527–534; 1994. 32. VanderWeele, D. A. Insulin is a prandial satiety hormone. Physiol. Behav. 56:619 – 622; 1994. 33. York, D. A.; Lin, L. Enterostatin and fat intake: current insights into possible mechanisms. In: Angel, A.; Anderson, H.; Bouchard, C.; et al., Eds. Progress in obesity research, Vol. 7, New York: John Libbey & Company Ltd.; 1996;495–501.