Physiology&Behavior,Vol. 51, lap.543-547, 1992

0031-9384/92 $5.00 + .00 Copyright© 1992 PergamonPressLtd.

Printed in the USA.

Intraperitoneally Injected CholecystokininOctapeptide Activates Pica in Rats B R U C E M c C U T C H E O N , l M I C H E L E B A L L A R D 2 A N D R O B E R T J. M c C A F F R E Y

Department of Psychology, The University at Albany, State University of New York, Albany, N Y 12222 R e c e i v e d 17 J u n e 1991 McCUTCHEON, B., M. BALLARD AND R. J. McCAFFREY. lntraperitoneallyinjectedcholecystokinin-octapeptide activates pica in rats. PHYSIOL BEHAV 51(3) 543-547, 1992.--Cholecystokinin-octapeptide (CCK-8) was injected intraperitoneally into rats to see if it could cause them to eat kaolin (clay)--a pica behavior which has been shown to indicate gastric distress. In the first study, a single large dose of CCK-8 (20 t~g/kg) failed to produce pica. In the second study, 4 smaller doses of CCK-8 (8 #g/ kg), 30 min apart, produced significant ingestion of kaolin compared to the baseline condition of vehicle injections. The pica was comparable to that observed in another group of rats given a toxic dose of LiCI ( 127 mg/kg, IP). It is concluded that interperitoneal injections of CCK-8 can induce a state of gastric distress in the rat. Cholecystokinin

Ingestion

Pica

Rats

Kaolin

CHOLECYSTOKININ (CCK) is an intestinal hormone and putative brain neuromodulator/neurotransmitter that has been proposed to have a role in satiety for food ingestion (10,11,15,20). Over several years, a debate has continued over the possibility that the suppression of eating caused by peripheral injection of CCK is due to gastric distress or nausea, not accelerated satiety (2,3,4,16,17). One of the kinds of evidence used to support the distress hypothesis comes from taste aversion conditioning (3,4). However, as Booth has shown, taste aversion conditioning may not unarguably be indicative of a state of distress; a novel taste associated with the satiety condition of the end of a meal may also control a suppression of ingestion (1). We reasoned, therefore, that a less ambiguous test should be used to assess the capacity of CCK to produce gastric distress. Such a test is provided by giving the CCK-injected animal an opportunity to ingest clay (kaolin). Rats made ill by LiCI injection (9,19), other toxins (8,9), or physical rotation (6,7) will eat a significant amount of clay (or soil). This ingestive response, called pica, is putatively an index that the rat has been sickened by the injection or physical rotation (viz., motion sickness). Because the only other known reason for significant ingestion of clay is in response to a mineral deficit (9), there is strong justification for interpreting this response as reflecting a state of gastric distress.

ually housed in hanging wire cages and kept in a temperatureand humidity-controlled colony room. Lights were on at 0700 h and off at 1900 h. Rat chow and water were available ad lib unless otherwise specified.

Kaolin The pica substance was made from a mixture of kaolin and an acacia gum base in a ratio of 99:1. Enough distilled water was added to the mixture to make it the consistency of cake icing. Using a cake decorating bag, we extruded the mixture onto plastic sheets in long strands, about 3-mm thick. After drying overnight, the strands were broken into 1 cm pieces (pellets) for easy handling by the rats.

Drugs CCK-8 was obtained from Bristol-Myers Squibb. It was dissolved in 0.15 M NaHCO3 (as recommended by Squibb for ease of solubility) and injected IP at a dose of 20 ~g/kg in a volume of 12 ml/kg. LiCI (reagent grade) was dissolved in distilled water to a concentration of 0.15 M. It was injected IP at a volume of 20 ml/kg (127 mg/kg).

Food

EXPERIMENT 1: TEST FOR PICA WITH A SINGLE 20 /~g/kg DOSE O F CCK-8

Liquid test food was chocolate Sustacal (Mead Johnson), presented in 100-ml graduated cylinders fitted with sipper tubes.

METHOD

Animals

Procedure

Ten, 3-month-old Sprague-Dawley males (purchased from Blue Spruce) were used in this experiment. They were individ-

Familiarization. For 5 days, rats were offered ad lib access to kaolin pellets that were placed in a small metal can fastened

Request for reprints should be addressed to Bruce McCutcheon. 2 Michele Ballard performed these studies as part of her research requirement for The Graduate Program in Biopsychology, SUNY Albany. 543

544

McCUTCHEON, BALLARD AND M('CAFFREY 2B z0 v

E

IqN,ea Plcex,llluea[~ccr~]

18

0O 16

.~:~ 12 O ~

8

•.o

6

0

30 rain

60 rain

90 rain

--

IP.O min

Total

FIG. 1. Mean intake of liquid food, in milliliters, is shown at 30-min intervals for a total access of 120 min. Successivetreatment conditions were NaCI, CCK-8, LiC1, and CCK-8. Error bars denote standard error of the mean. to the back corner of the cage. Then, for two 24-h periods, separated by 2 days of rat chow and water, the rats were deprived of chow overnight (22 h) and given access to liquid food and kaolin pellets for 24 h. Baseline. One week after familiarization, rats were again deprived of chow for 22 h, then injected with 0.15 M NaCI (12 ml/k~) and offered liquid food (at 1450 h) and approximately 40 g of kaolin pellets. Every 30 min for a total of 2 h, intake of liquid food was recorded (to the nearest 0.5 ml). At the end of 24 h, the kaolin pellets were removed, spillage was collected from beneath the cage; the pellets were air dryed for 2 days, and weighed to determine the amount ingested (to the nearest 0.1 g). One week later, this baseline test was repeated. Treatment. One week following the second baseline test, the rats were tested for the effect of CCK-8 IP injection in the same

manner as baseline testing. CCK-8 is commonly injected at doses ranging from 2-8 #g/kg to cause significant suppression of eating (15). At the lower end of the dose range, CCK-8 and LiC1 have been compared for food intake suppression (15). Because pica in response to LiCI has a much higher threshold than either food suppression or conditioned taste aversion [about 2× and 10× respectively; see (19)], we chose a dose of CCK-8 considerably higher than that commonly used for demonstratinga suppression of eating in hungry rats. Thus, each rat was injected with 20 t~g/ kg CCK-8 (12 ml/kg) and observations of liquid food intake were taken every 30 min for a total of 2 h. Kaolin ingestion was measured at the end of 24 h. One week later, these same rats were tested in the same manner with 0.15 M LiC1 IP injection (20 ml/kg), a dose (127 nag/ kg) known to cause pica (19).

1.6 1.4 ~.. 1.2

~

1.0

"-~ 0.8 O

j

0.6

e~ 0.4 0.2 0.0

NaCI

CCK 1

LiCI

CCK 1

Treatment FIG. 2. Mean 24-h intake of kaolin, in grams, is shown for four successivetests---followingNaCI, CCK-8, LiC1, and CCK-8 IP injection. Error bars denote standard error of the mean. Asterisk over the LiCI bar denotes a significantdifference(p < 0.05) from all other conditions.

CCK ACTIVATES PICA

545

1.6 1.4 1.2 ~1.0 0.8 O

0.6

eq 0.4

T

T

0.2 0.0

NaHCO 3

CCK

NaHCO 3

LiC1

t

I

[

I

Treatment FIG. 3. Mean 24-h intake of kaolin, in grams, is shown following two successive tests---vehicle (NaHCO3) and drug (CCK-8 or LiCI). One group received CCK-8 and the other received LiC1 as IP injections. Error bars denote standard error of the mean. Asterisks over the CCK and LiCI bars denote significant differences (p < 0.05) from their respective control conditions.

Finally, a week later, the rats were retested with CCK-8 as in the first treatment. This was to see if, having demonstrated pica with LiCI, the rats would be sensitized to respond to CCK with pica.

Analysis One-way, within group ANOVA's were performed for significance testing of the drug effects on liquid food intake and kaolin ingestion. Individual comparisons were performed with the Tukey test.

as a cause of suppressed eating. Data from tests with LiC1 show, for example, that feeding is suppressed at about half the dose required to induce a reliable pica (19). That is, pica is not sensitive to all levels of gastric distress. A major difference between LiC1- and CCK-suppressed eating is the duration of the suppression. CCK is rapidly metabolized by the liver so that the rat begins to return to ingesting food after about 45 min of quietude. In contrast, LiC1 caused a prolonged, almost 2-h suppression of all behavior, with eating returning steadily after that. In the next study, we attempted to duplicate the duration of suppression, as well as the magnitude of suppression by giving lower, repeated doses of injected CCK.

RESULTS

Figure l shows the results of NaCl, CCK, and LiCl on ingestion of liquid food. A significant overall ANOVA [F(3,27) = 42.5 l, p < 0.005] was decomposed by the Tukey test to show that both CCK and LiCl treatments caused reliable reduction of total intake compared to NaCl baseline [F(1,9) = 74.1 l, p < 0.05], and LiCl caused more reduction than either CCK test [F(l,9) = 17.24, p < 0.05]. The amount of the reduction was 50% and 60% for the two CCK tests and 80% for the LiCl test. Kaolin pellet consumption was activated by LiCl treatment [see Fig. 2; F(3,27) = 6.67, p < 0.005], but not by either CCK test compared to NaCl [F(l,9) = 0.002, p > 0.10]. The increase was approximately 260% of baseline. DISCUSSION

A single large dose of CCK did not cause pica, but suppressed eating by 50%. LiCI, on the other hand, caused pica and suppressed eating by 80%. Even after recovering from a toxic dose of LiCI, and displaying pica to that toxin, the rats retested with CCK were again unresponsive in the pica test. However, negative data from a pica test cannot rule out aversive reactions to CCK

EXPERIMENT 2: TEST FOR PICA WITH FOUR SUCCESSIVE 8 #g/kg DOSES OF CCK-8 METHOD

Animals Twenty four male Sprague-Dawley rats, approximately 3 months old, participated in this study. They had previously served as untreated control animals in another study and had no history of drug exposure. They were maintained as described in Experiment 1.

Kaolin, Liquid Food, Drug Kaolin pellets and liquid food were obtained as described in Experiment 1. CCK-8, from the same batch used in Experiment 1, was given IP in four spaced injections at a dose of 8/~g/kg per injection (first dose at 20 ml/kg then remaining doses at 1 ml/kg) and 0.15 M LiCI was injected IP in a single dose of 127 mg/kg (20 ml/kg).

Familiarization This was done as described in Experiment 1.

546

McCUTCHEON, BALLARD AND M('CAFFRE~

34,18

:5.27

22 [ ] NaI-ICO3 [ ] LiC1 • CCK i 20 18 16

~ 10

4

30 rain

60 win

90 min

120 rain

150 rain

180 min

Total

FIG. 4. Mean intake of liquid food, in milliliters, is shown at 30-min intervals for a total access of 180 min. Successive treatment conditions were vehicle (NaHCO3) and drug (LiCI or CCK-8). One group received LiCI and the other CCK-8 as IP injections. Error bars denote standard error of the mean. At each time point, the first NaHCO3 bar gives the baseline results from the LiCI group, and the second NaHCO3 bar gives the baseline results from the CCK group.

Baseline Testing for liquid food ingestion and kaolin ingestion after food deprivation (22 h) was done as in Experiment l, with the following differences. The baseline injection was 0.15 M NaCO3 (20 ml/kg), the vehicle used to dissolve CCK-8. However, instead of a single IP injection, as in Study l, four injections were given: the first was 20 ml/kg, and the three subsequent ones were l ml/kg, 30 min apart. Food ingestion was measured at 30 min intervals for a total of 3 h. Kaolin ingestion was measured at the end of 24 h, as in Experiment 1.

Treatment One week following baseline measures, the rats were divided into two groups: a CCK group and a LiC1 group. Each group had 12 rats, which were distributed to the groups by matching on amount of kaolin consumed during baseline testing. Both groups were tested in the same manner as during baseline testing, but with CCK or LiCl injection instead of vehicle injections. The CCK group received the first injection of CCK at a dose of 8 #g/kg in a volume of 20 ml/kg, and three subsequent injections, 30 min apart, at the same dose, but in a volume of I ml/kg. The LiCi group received their first injection of0.15 M LiCl at a dose of 127 mg/kg in a volume of 20 ml/kg, and three subsequent injections, 30 min apart, of 0.15 M N a O in a volume of I ml/ kg. Liquid food ingestion and kaolin ingestion were measured as during baseline testing.

Analysis Reliability was tested by an ANOVA for between groups (CCK vs. LiCl) within treatments (vehicle vs. drug). Specific comparisons were made with the Tukey test.

RESULTS Both CCK and LiC1 caused a significant ingestion of kaolin over 24 h compared to the vehicle baseline condition [F(1,20) = 7.28, p < 0.05; see Fig. 3. Note that group size was reduced by the loss of one rat per group because of illness]. For LiC1, the increase was 500% over baseline, and for CCK, the increase was 300% over baseline (the difference between CCK and LiCI was not significant [F(1,20) = 0.65, p > 0.10]. As predicted, CCK and LiC1 caused a similar pattern of suppression of eating over the 3-h observation period (Fig, 4). During the first 30 min, intake dropped more than 50% of baseline. Suppression of consumption continued for the next 90 min, then began to rise over the last 30 rain of observation (150-180 rain). DISCUSSION Intraperitoneally injected CCK-8 caused pica. By dosing rats with CCK-8 in such a way as to equate both magnitude and duration of suppression of eating with that caused by an effective dose of LiCl, the rats ate kaolin. Our inference, therefore, is that the CCK injections induced gastric distress. Because Study l showed that a single large dose of CCK does not cause pica, and given the 30-min spacing of injections and relatively rapid inactivation of CCK, this pica cannot be explained as the result of a cumulatively large dose (i.e., greater than reached shortly after a single injection of 20/zg/kg CCK). Rather, it is likely that the prolongation of CCK action over a few hours is necessary to engage the pica response, just as the prolongation of LiCl toxic response may be necessary for pica to occur. In no way should these data imply that all CCK studies generate an artifact of gastric distress. Some of the more recent efforts to study the action of endogenous CCK by way of receptor antagonists are particularly helpful because eating behavior is

CCK ACTIVATES PICA

547

increased rather than suppressed (5). What we do suggest is that those studies which use peripheral injections of C C K may, in fact, cause an aversive state, especially if the dose of the injection raises circulating C C K well above physiological levels (13,14). Such a state is implied by published data showing hormonal changes indicative of a stress reaction (12,18). Unless one works within the physiological range of endogenous CCK, which in the rat peaks at levels at least 10 times lower than required to obtain suppression of eating by intravenous injection (14), it would be prudent to eschew the approach of artifically elevating the level of circulating CCK. Instead, manipulating endogenous

C C K action by way of selective receptor antagonists would be a more illuminating approach to the study of CCK's involvement in satiety. ACKNOWLEDGEMENTS The authors are pleased to acknowledge the generous contribution of CCK-8 from S. J. Lucania and information pertaining to CCK metabolism from Amy Tetervin, both of the Research Chemicals Division, Bristol-Myers Squibb. We also wish to thank Dr. John Hannigan, formerly of the Center for Behavioral Teratology (SUNY-Albany), for his gift of Sustacal.

REFERENCES 1. Booth, D. A. The physiology of appetite. Br. Med. Bull. 37:135140; 1987. 2. Davidson, T. L.; Flynn, F. W.; Grill, H. J. Comparison of the introceptive sensory consequences of CCK, LiC1, and satiety in rats. Behav. Neurosci. 102:134-140; 1988. 3. Deutsch, J. A.; Hardy, W. T. Cholecystokinin produces bait shyness in rats. Nature 266:196; 1977. 4. Deutsch, J. A.; Thiel, T. R.; Greenburg, L. H. Duodenal motility after cholecystokinin injection or satiety. Behav. Biol. 24:393-399; 1978. 5. Dourish, C. T.; Rycroft, W.; Iverson, S. D. Postponement of satiety by blockade of brain cholecystokinin (CCK-8) receptors. Science 245:1509-151 l; 1989. 6. McCaffrey, R. J. Appropriateness of kaolin consumption as an index of motion sickness in the rat. Physiol. Behav. 35:151-156; 1985. 7. McCaffrey, R. J.; Graham, G. Age-related differences for motion sickness in the rat. Exper. Aging Res. 6:555-561; 1980. 8. Mitchell, D.; Beatty, E. T.; Cox, P. K. Behavioral differences between two populations of wild rats: Implications for domestication research. Behav. Bio. 19:206-216; 1976. 9. Mitchell, D.; Wells, C,; Hoch, N.; Lind, K.; Woods, S. C.; Mitchell, L. K. Poison induced pica in rats. Physiol. Behav. 17:691-697; 1976. 10. Moran, T. H.; McHugh, P. R. Cholecystokinin suppresses food intake by inhibitinggastric emptying. Am. J. Physiol. 242:R491-R497; 1982. I I. Muurahainen, N.; Kissileff, H. R.; Derogatis, A. J.; Pi-Sunyer, F. X. Effects ofcholecystokinin-octapeptide(CCK-8) on food intake and gastric emptying in man. Physiol. Behav. 44:645-649; 1988. 12. Parrott, R. F.; Ebenezer, 1. S.; Baldwin, B. A.; Forsling, M. L. Central and peripheral doses of cholecystokinin that inhibit feeding in pigs also stimulate vasopressin and cortisol release. Exp. Physiol. 76:525531; 1991.

13. Reidelberger, R. D.; Kalogeris, T. J.; Solomon, T. E. Plasma CCK levels after food intake and infusion of CCK analogues that inhibit feeding in dogs. Am. J. Physiol. Regul. Integ. Comp. Physiol. 256(25): R1148-R1154; 1989. 14. Reidelberger, R. D.; Solomon, T. E. Comparative effects of CCK-8 on feeding, sham feeding, and exocrine pancreatic secretion in rats. Am. J. Physiol. ReguL Integ. Comp. Physiol. 251(20):R97-R105; 1986. 15. Smith, G. P.; Gibbs, J.; Kulkosky, P. J. Relationship between braingut peptides and neurons in the control of food intake. In: Hoebel, B. G.; Novin, D., eds. The neural basis of feeding and reward. Brunswick, ME: Haer Institute for Electrophysiological Research; 1982:149-165. 16. VanderWeele, D. A.; Granja, J. A.; Deems, D. A. Discomfort or satiety: The spontaneous meal pattern may serve as a predictor. In: Hoebel, B. G.; Novin, D., eds. The neural basis of feeding and reward. Brunswick, ME: Haer Institute for Electrophysiological Research; 1982:167-173. 17. Verbalis, J. G.; McCann, M. J.; McHale, C. M.; Stricker, E. M. Oxytocin secretion in response to cholestokinin and food: Differentiation of nausea and satiety. Science 232:1417-1420; 1986. 18. Verbalis, J. G.; Richardson, D. W.; Stricker, E. M. Vasopressin release in response to nausea-producing agents and cholecystokinin in monkeys. Am. J. Physiol. Regulatory Integrative Comp. Physiol. 252(21):R749-R753, 1987. 19. Watson, P. J.; Leitner, C. Patterns of increased and decreased ingestive behavior after injections of lithium chloride and 2-deoxy-dglucose. Physiol. Behav. 43:697-704; 1988. 20. Zhang, D.; Bula, W.; Stellar, E. Brain cholecystokinin as a satiety peptide. Physiol. Behav. 36:1183-1186; 1986.