Lesions of the area postrema and underlying solitary nucleus fail to attenuate the inhibition of feeding produced by systemic injections of cholecystokinin in Syrian hamsters

Physiology & Behavior, Vol. 38, pp. 855-860. Copyright ¢ Pergamon Journals Ltd., 1986. Printed in the U.S.A.

0031-9384/86 $3.00 + .00

Lesions of the Area Postrema and Underlying Solitary Nucleus Fail to Attenuate the Inhibition of Feeding Produced by Systemic Injections of Cholecystokinin in Syrian Hamsters I M A R I O O. MICELI, .2 C Y N T H I A A. P O S T ' t A N D D E R E K VAN DER K O O Y *

*Departments o f A n a t o m y , ~fPhysiology and Psychology, University o f Toronto Toronto, Ontario, Canada M S S IA8 R e c e i v e d 15 M a y 1986 MICELI, M. O., C. A. POST AND D. VAN DER KOOY. Lesions of the area postrerna and underlying solitary nucleus fail to attenuate the inhibition of feeding produced by systemic injections of cholecystokinin in Syrian hamsters. PHYSIOL BEHAV 38(6) 855-860, 1986.--A large body of evidence indicates that the intestinal hormone cholecystokinin (CCK) may serve as a signal for satiety. The abdominal vagus has been shown to be important for the satiety response to exogenous, and by inference, endogenous, CCK in rats and hamsters. Thus, it appears that stimulation of CCK receptors on afferent fibers of the abdominal vagus activates a gut-brain pathway to signal satiety. The present study was undertaken to further trace this viscerosensory pathway by examining food intake after administration of one of two doses (2.0 and 8.0/~g/kg) of CCK-octapeptide to intact hamsters and to hamsters sustaining lesions of the area postrema (AP) and underlying nucleus of the solitary tract (NST), regions containing neurons postsynaptic to vagal afferent fibers. As lesions of the AP/NST result in many alterations in ingestive behaviour and body weight regulation in rats, various aspects of feeding and drinking behaviour (spontaneous food intake, body weight maintenance, and responsiveness to a palatable drinking solution and osmotic stimulation) were also examined in lesioned hamsters. Aside from producing transient hypophagia and weight loss immediately after surgery, AP/NST lesions had no effects on these various parameters of ingestive behaviour. The lack of lesion effects on these particular parameters may be explained on the basis that hamsters are generally unresponsive to many of the stimuli for feeding and drinking which purportedly act on the vagus and/or AP/NST. Hamsters with AP/NST lesions were as responsive to the two tested doses of CCK as intact animals. These results suggest that, relative to effective lesions in rats, ablation of greater amounts of hamster NST may be necessary for the blockade of the feeding inhibitiory response to CCK. ARernatively, CCK may have acted on a CCK-receptive system that is independent of the vagusAP/NST to reduce feeding in lesioned hamsters. Area postrema Vagus nerve

Body weight

Cholecystokinin

Hamster

Nucleus of the solitary tract

Satiety

central nervous system [24,25]. A central site of action of intestinally released C C K appears unlikely, as systemically administered radiolabelled C C K does not readily cross the blood-brain barrier [17]. Recent investigations suggest, instead, that CCK produces its effect on feeding by acting at a peripheral site(s). In a series of experiments, Smith and his colleagues have shown that a section of the abdominal vagus nerve in the rat abolishes the feeding inhibitory response to systemic CCK injections [20]. Further work by these inves-

IT is now firmly established that systemic injections of the gut hormone, cholecystokinin (CCK), reduce feeding in fasted animals (see [7] for review). This and other lines of evidence [1,7] have suggested that prandially released intestinal C C K may act as a satiety stimulus. Although the effects of C C K on feeding have long been known, workers have only recently begun to investigate sites and mechanisms of C C K action. C C K binding sites are distributed in both the gut and

1Supported by a grant from the Natural Sciences and Engineering Research Council of Canada to D.v.d.K. and a Medical Research Council of Canada postdoctoral fellowship to M.O.M. We gratefully acknowledge R. Selberg for assistance with the statistical analyses and the preparation of the manuscript. ZRequests for reprints should be addressed to Dr. Mario Miceli, Department of Anatomy, University of Toronto, Medical Sciences Building, Toronto, Ontario, Canada M5S 1A8.

855

856

MIC[:t_I t:~ A I

FIG. 1. Schematic representations of medulla sections of intact (A) and lesioned (B) hamsters. Top to bottom=rostral to caudal. The thin lines in B indicate the area of damage in each of the lesioned hamsters. Abbreviations: AP--area postrema; DMN-dorsal motor nucleus of the vagus; MVN--medial vestibular nucleus; NC--nucleus cuneatus; NG--nucleus gracilis; NST-nucleus of the solitary tract; ST--solitary tract; XIl--hypoglossal nucleus.

tigators has narrowed down the critical site o f C C K action to afferent fibers of the gastric branch of the rat abdominal vagus [20,21]. C C K may act in one of two (or both) ways to elicit a vagally-carried satiety signal. C C K may act directly on vagal receptors [25] to activate the afferent limb of a satiety reflex [12,20]. Alternatively, CCK may activate satiety-related vagal afferents indirectly by inhibiting gastric emptying (through its action on pyloric smooth muscle receptors [22]) which facilitates gastric distention, a vagallytransmitted satiety stimulus [11,12]. Studies designed to further trace the pathway transmitting a CCK-elicited satiety signal have involved testing C C K ' s feeding inhibitory effect in rats with lesions of central nervous system neurons postsynaptic to afferent fibers of the abdominal vagus. The central projections of abdominal vagal

afferent fibers include the area postrema (AP) and the medial and commissural division of the nucleus of the solitary t r a c t (NST) [14, 16, 19]. We and other i n v e s ~ a t o r s have independently reported that lesions of these AP/NST second order sensory neurons in the rat have the predicted effect of abolishing or attenuating the feeding inhibitory response to C C K [3, 4, 23]. Syrian hamsters (Mesocricetus auratus) also reduce their food intake after systemic CCK-8 injections [11,13]. As in rats, total abdominal vagotomy in hamsters can attenuate the feeding inhibitory response after CCK injections [I2]. There are, however, several important differences in this between rats and hamsters. First, urda'ke vagotomized rats which appear to be totally unresponsive to a wide range o f CCK doses [10,20], vagotomized hamsters are unresponsive

AP/NST L E S I O N S A N D C C K I N H I B I T I O N O F FOOD I N T A K E

857

to low dose (<~6.0 /xg/kg) CCK injections, but respond to higher doses (I>8.0 /xg/kg) to the same degree as controls [12]. Second, selective removal of the gastric branch of the vagus appears to be necessary and sufficient for blockade of the feeding inhibitory response to CCK in rats [20]. Although selective section of the hamster gastric vagus can also attenuate responsiveness to low dose CCK injections, this type of nerve section is not as effective as removal of all the abdominal branches [12], suggesting that other branches of the vagus (hepatic and/or celiac) are also important for the feeding inhibitory response to CCK in this species. In a preliminary study we found that selective (ibotenic acid) lesions of cell bodies in the hamster A P and subadjacent NST did not alter feeding suppression in response to CCK (unpublished observations). These data, however, were difficult to interpret owing to the small sample sizes and the difficulty in determining the full extent of the lesion. In the present study we wished to examine the feeding inhibitory response to CCK in hamsters after more radical (aspiration) lesions of the AP/NST. In addition to abolishing the feeding inhibitory response to systemic CCK injections, lesions of the AP and NST in rats produce many other alterations in ingestive behaviour including: severe hypophagia and weight loss immediately after the lesions [8]; mild hypophagia and maintenance of body weight at lower levels thereafter [8]; overconsumption of highly palatable foods [5]; absence of feeding in response to glucoprivic challenges [2]; and chronic hyperdipsia and exaggerated drinking in response to intracellular and extracellular dehydration [6,15]. Thus, we were also interested in determining whether AP/NST lesions have similar effects in hamsters.

METHOD

Animals Adult (90 days and over) male and female hamsters of Lakeview origin ( L A K : L V G ) were purchased from Charles River of Canada.

Housing and Maintenance The hamsters were individually housed in polycarbonate tub cages with wood shavings for bedding. The animal room was maintained at 22°C with a reversed 14:10 hr light:dark illumination cycle. Except when otherwise, indicated, animals had ad lib access to food (Purina rat chow pellets) and water.

Surgery The hamsters were anesthetized with sodium pentobarbitol and placed in a stereotaxic intrument head holder. A midline incision was made in the dorsal neck musculature which was retracted by blunt dissection to expose the occipital plate and cisterna magna. With visual guidance from a dissecting microscope, the dural sheath over the cisterna magna was excised, exposing the dorsal surface of the medulla and overlying cerebellum. The AP and variable amounts of the underlying N S T were aspirated with a 22 ga needle under vacuum pressure. Sham operates were treated indentically to the point of excising the dura over the cisterna magna. Surgical wounds were closed with silk sutures and stainless steel surgical clips.

FIG. 2. Photomicrographs of the dorsal medulla of an intact hamster (A), of a hamster with an AP lesion, but with relatively little damage to the underlying NST (B), and of a hamster with an AP lesion which also extensively damaged the NST (C). Hamsters with the large lesions (as in C) lost the most weight after surgery, but were as responsive to CCK-8 as hamsters with small lesions or intact hamsters. Abbreviations: cNST--commissural division of the nucleus of the solitary tract; 1NST--lateral division of the nucleus of the solitary tract; mNST--medial division of the nucleus of the solitary tract.

Food Intake and Body Weight The animals were weighed on every fifth day beginning on the fifteenth day before surgery. F o o d was replenished on the days when the animals were weighed. Food intakes dur-

MICI-tA I).1 AI

858

1.2 Controls

5O

1.0 ~C d:

40

n=8

A P/NST Lesioned

N

n =10

0.8

i

2c 30

_z 0 . 6 o

•

AP/NST

Lesioned

n=10

2O

L

0 0

t~

0.4 •

Controls t

t

n= 8 i

i

i

140

0.2

i o~

130

0

2

8

DOSE (,ug/kg) C C K - 8

~ 12o

FIG. 4. Mean (_+s.e.m.) food intakes of lesioned and intact hamsters after IP injections of CCK-8 or saline vehicle. o

m 110

-10

0

'

b'

1

0 Days

30

0

50

Postsurgery

FIG. 3. Mean (-+s.e.m.) food intakes (top) and body weights (bottom) during the course of the experiment.

on days 45 and 46 postsurgery. On days 47 and 48 postsurgery, the animals' water was replaced with a 10% sucrose solution. Intake of sucrose solution was recorded daily during this period.

Data Analysis ing the preceding five days were measured by subtracting let~-over pellets from the amount previously provided. Except for three hr intervals on three separate days (see below), animals had ad lib access to food throughout the experiment.

Feeding Response to CCK Injections Beginning on day 15 postsurgery, the animals underwent a series of feeding tests with CCK-octapeptide (CCK-8; a generous donation from Squibb). The hamsters were prepared for each test by a three hr fast. At the end of the fast, the animals were given 2.0 p,g/kg, 8.0/zg/kg CCK-8, or equal volumes o f isotonic saline, by intraperitoneal injection. Five min after the injection the animals were given a weighed ration of food which was reweighed 1 hr later. Treatments were administered to each animal in random order at five day intervals.

Drinking Response to Hypertonic Saline On days 35 and 40 postsurgery, the hamsters' water intakes were measured after subcutaneous injection of isotonic (0.9%) or hypertonic (12.0%) saline (0.8 ml/100 g). Each concentration was in a solution containing 2.5% procaine hydrochloride. Water intakes were measured over the first three hr after the injection. Half of the animals were tested with isotonic saline on day 35 postsurgery and with hypertonic saline on day 40 postsurgery. The remaining animals received treatments in reversed order.

Intake of a Palatable Solution Daily water consumption for each animal was recorded

Data for each dependent measure underwent two-way (group × treatment or group x time) analysis of variance with repeated measures on the treatment or time factor. In those instances where group x time or group × treatment interactions were significant, tests were made on simple main effects.

Histology Upon completion of the behavioural testing, the animals were deeply anesthetized with pentobarbital and were perfused intracardially with saline followed by buffered 10% formalin. The brains were removed from the skull, blocked in the region of the lesion and stored in formalin until sectioning on a freezing microtome. Fifty micron frontal sections of the caudal brainstem were mounted onto chrom-alum coated slides which were later stained with thionin. Evaluation of the location and extent of brain damage was conducted without knowledge of the animals' behaviour. RESULTS

Histology The A P was completely ablated in all ten lesioned hamsters. In addition there were variable amounts of damage to the underlying NST. In each case there was some amount of damage to the commissural NST, immediately adjacent to the AP. In five cases, the lesions encroached on the medial NST and dorsal motor nucleus unilaterally. In three hamsters there was extensive bilateral damage to the medial NST and dorsal motor nucleus with some amount of danmge to the dorsal aspect of the hypoglossal nucleus (see Figs. 1 and 2).

AP/NST LESIONS AND CCK INHIBITION OF FOOD INTAKE

Body Weights Body weights on the day of surgery and on days 15, 30, and 45 postsurgery were used in the statistical analysis. Although animals in both groups lost some weight after surgery, F(3,48) = 10.1, p <0.01, weight loss was much more pronounced in lesioned hamsters, F(3,48)=4.7, p<0.05 (see Fig. 3, bottom). There was no main effect of surgical treatment on body weight. However, tests on simple main effects revealed that sham-operates were significantly heavier than lesioned animals on day 15 postsurgery (p<0.05), but not on the day of surgery, nor on days 30 and 45 postsurgery.

Food Intake Food intakes during the five day block preceding surgery, and during the first, third, and fifth five day blocks postsurgery were used in the statistical analyses. As in the case of body weight, there was no overall main effect of surgery on food intake. Food intakes were depressed after surgery, F(3,48)=23.9, p<0.01, and this effect too was more pronounced in lesioned hamsters, F(3,48)= 13.1, p <0.01 (see Fig. 3, top). These effects were attributed to the sharp reduction in food intake by lesioned hamsters during the first five day time block. It was only during this time block that intakes differed significantly (p<0.05) between the groups.

Feeding Responses to CCK-8 Lesioned hamsters tended to eat less than intact animals after each of the three tests, including the test after injection of saline vehicle. This trend was statistically reliable, F(1,16)=6.2, p<0.05. CCK-8 injections produced a doserelated suppression of food intake in both groups of animals, F(2,32)=12.9, p<0.01, but the magnitude of the response was consistent between the groups (see Fig. 4).

Drinking Response to Hypertonic Saline Hamsters in both groups drank more after hypertonic saline injections than after isotonic saline injections (9.1 ±0.6 vs. 4.0--_1.0 ml, respectively for intact animals and 11.2±0.9 vs. 4.5---0.2 ml, respectively for lesioned hamsters), F(1,16)=77.8, p <0.01. The magnitude of this increment did not vary between the groups.

Ingestion of Palatable Liquids Fluid intakes were significantly higher (approximately 94% for control animals and 70% for lesioned animals) on the days when 10% sucrose was available (mean_s.e.m. for control hamsters on the first and second days of availability=25.6±3.5 and 34.6±3.0; mean±s.e.m, for lesioned hamsters=22.4±2.7 and 30.0___3.2) than when water was available (mean-s.e.m. for controls=16.8±0.9 and 14.2--- 1.1; mean±s.e.m, for lesioned animals= 15.5___1.1 and 15.2±0.9), F(3,48)=34.3, p<0.01. The magnitude of this increment was also consistent between the lesioned and intact groups. DISCUSSION

The AP and underlying NST in rats have been demonstrated to be important components of the neural substrate underlying feeding and drinking behaviour and body weight maintenance. Ablation of the AP/NST in rats results in a syndrome characterized by hypophagia and weight loss [8];

859

loss of responsiveness to glucoprivic feeding stimuli [2]; overresponsiveness to certain drinking stimuli [6,15] and palatable foods [5]; and attenuation of CCK's feeding inhibitory effect [3, 4, 23]. In contrast to the above syndrome demonstrated in the rat, we report here that hamsters with comparable lesions do not show chronic hypophagia nor long term deficits in body weight regulation; do not overconsume a palatable drinking solution nor overconsume water in response to intracellular dehydration; and remain responsive to the feeding inhibitory effect of systemic CCK injections. Evidently, lesioned hamsters are similar to their rat counterparts only in showing transient hypophagia and weight loss immediately after surgery. In rats, lesions of the AP/NST mimic many of the effects of abdominal vagotomy (including mild, chronic hypophagia and maintenance of body weight at lower levels [8]) and in this respect these lesions may be considered "central vagotomies." In view of the fact that abdominal vagotomy in hamsters does not produce severe alterations in ingestive behaviour and body weight regulation [12] along with observations that hamsters are not responsive to many of the feeding and drinking stimuli which purportedly act on the vagus and/or AP/NST [18], it is not surprising that AP/NST lesions are not as debilitating in hamsters as they appear to be in rats. What remains to be explained, however, are the seemingly paradoxical findings that transection of the hamster abdominal vagus, but not lesions of central neurons postsynaptic to afferent fibers of this nerve, can attenuate the feeding inhibitory response to CCK. The projections of the abdominal vagus to the lower brainstem have been studied in rats [16,19] and hamsters [14]. These neuroanatomical studies revealed a similar distribution of vagal afferent terminals within the dorsal vagal complex of each species, with the AP and the commissural and medial divisions of the caudal NST receiving the densest innervation. The precise localization of those secondorder sensory neurons necessary for feeding suppression in response to systemic CCK in rats has been subject to some debate. Work conducted in our laboratory [23] indicated that lesions of the rat AP which encroached variably, but in a relatively minimal manner, on the subadjacent NST were sufficent to block feeding suppression after CCK injections, thus suggesting that the AP may be critically important. Other workers have suggested that the NST alone is critical for the response [3], and yet another group of workers has reported that ablation of both the AP and medial division of the NST is necessary for blockade of the feeding inhibitory response to CCK [4]. In the present study, lesions of the hamster AP with minimal or extensive damage to those zones of the underlying NST innervated by the sensory components of the abdominal vagus [14, 16, 19] were both ineffective at blocking or attenuating the feeding inhibitory response to CCK. In fact, some of our ineffective lesions involved more extensive damage to the NST than the effective lesions reported in the rat studies [3,23]. It is possible that abdominal vagal afferents carrying a CCK-initiated satiety signal in the hamster may have a wider distribution within the dorsal vagal complex than in the rat and that ablation of greater amounts of hamster NST tissue is required for the blockade or attenuation of the feeding inhibitory response to CCK. Partial sparing of these widely distributed second-order sensory neurons may thus account for why lesioned hamsters remained responsive to CCK. Although this appears to be the most parsimonious interpretation of our negative findings, the fact that there was no trend toward

860

MIC?E}~I t / t /

diminished responsiveness to CCK in the hamsters with near-complete damage to the central terminations of the abdominal vagus gives rise to the possibility for alternative explanations. As noted earlier, hamsters with abdominal vagotomy will show a significant feeding suppression in response to relatively large doses of CCK [12]. Although "real feeding" (see below) vagotomized rats appear to be totally unresponsive to even large doses of C C K [10,20], " s h a m feeding" (liquid food passes out of an oesophageal or gastric fistula and does not enter the glandular stomach) vagotomized rats are as responsive to a range of C C K doses as " s h a m feeding" intact rats [9]. The fact that, under certain circumstances, vagotomized hamsters and rats will reduce feeding in response to C C K requires sites of action other than the abdominal vagus or sites innervated by the nerve. Therefore, it is also possible that CCK may have acted on extravagal receptors to reduce feeding in hamsters with AP/NST lesions. Within this hypothetical framework, the apparent dissociation between the effects of abdominal vagotomy and AP/NST lesions may be explained on the basis that direct damage to the relevant neurons in the dorsal vagal complex lowers the activation threshold of these putative extravagal receptors and that abdominal vagotomy, which only partially denervates neurons of the dorsal vagal complex, does not. It is interest-

ing to note that in one study of the rat [4]. animals with extensive lesions of the AP/NST were totally unresponsive to relatively low (<8.0/zg/kg) doses of CCK-8, yet were ax responsive as intact animals to a larger (8.0/zg/kg) dose. This finding is also consistent with the idea that CCK can act on a system(s) independent of the vagus-AP/NST. The differences in effective doses reported for vagotomized or AP/NST lesioned rats and hamsters may reflect that these putative extravagal CCK receptors may have a lower activation threshold in hamsters than in rats. In summary, lesions of the AP/NST in hamsters do not produce many of the effects on ingestive behaviour and body weight regulation previously demonstrated in the rat. Our observations that hamsters sustaining such lesions remain responsive to the feeding inhibitory effects of CCK do not necessarily argue against a role for these second-order sensory neurons in processing and further transmission of a CCK-elicited, vagally-carried satiety signal in the intact hamster. The present results along with those in our previous study of vagotomized hamsters [12] suggest that systemically injected CCK may also act at extravagal sites. Further work is clearly needed to delineate what role the AP/NST may have in processing and transmitting a CCK-generated feeding inhibitory response in this species.

REFERENCES 1. Baile, C. A., C. L. McLaughlin and M. A. Della-Fera. Role of cholecystokinin and opioid peptides in the control of food intake. Physiol Rev 66: 172-234, 1986. 2. Contreras, R. J., E. Fox and M. L. Drugovitch. Area postrema lesions produce feeding deficits in the rat: Effects of preoperative dieting and 2-deoxy-D-glucose. Physiol Behav 29: 875-884, 1982. 3. Crawley, J. N. and J. S. Schwaber. Nucleus tractus solitarius lesions block the behavioral actions of cholecystokinin. Peptides 4: 743-747, 1983. 4. Edwards, G. L., E. E. Ladenheim and R. C. Ritter. Dorsal hindbrain participation in cholecystokinin induced satiety. Soc Neurosci Abstr 11: 37, 1985. 5. Edwards, G. L. and R. C. Ritter. Ablation of the area postrema causes exaggerated consumption of perferred foods in the rat. Brain Res 216: 265-276, 1981. 6. Edwards, G. L. and R. C. Ritter. Area postrema lesions increase drinking to angiotensin and extraceUular dehydration. Physiol Behav 29: 943-947, 1982. 7. Gibbs, J. and G. P. Smith. The neuroendocrinology of postprandial satiety. In: Frontiers in Neuroendocrinology, vol 8, edited by L. Martini and W. F. Ganong. New York; Raven, 1984, pp. 223-245. 8. Hyde, T. M. and R. R. Miselis. Effects of area postrema/caudal medial nucleus of solitary tract lesions on food intake and body weight. A m J Physiol 244: R577-R587, 1983. 9. Kraly, F. S. Vagotomy does not alter cholecystokinin's inhibition of sham feeding. A m J Physiol 246: R829-R831, 1984. 10. Lorenz, D. N. and S. A. Goldman. Vagal mediation of the cholecystokinin satiety effect in rats. Physiol Behav 29: 59% 604, 1982. 11. Miceli, M. O. Role of gastric distenstion in cholecystokinin's satiety effect in golden hamsters. Physiol Behav 35: 945-953, 1985. 12. Miceli, M. O. Abdominal vagus and regulation of ingestive behavior and body weight in golden hamsters. Am J Physiol 248: R686-R697, 1985. 13. Miceli, M. O. and C. W. Malsbury. Effects of putative satiety peptides on feeding and drinking behavior in golden hamsters. Behav Neurosci 99:1192-1207, 1985.

14. Miceli, M. O. and C. W. Malsbury. Brainstem origins and projections of the cervical and abdominal vagus in the golden hamster: A horseradish peroxidase study. J Comp Neurol 237: 65-76, 1985. 15. Miselis, R. R., T. M. Hyde and R. E. Shapiro. Area postrema and adjacent solitary nucleus in water and energy balance. Fed Proc 43: 296%2971, 1984. 16. Norgren, R. and G. P. Smith. The central distribution of the vagal subdiaphragmatic branches in the rat. Soc Neurosci Abstr 9: 611, 1983. 17. Passaro, E., E. Debas, W. Oldendorf and T. Yamada. Rapid appearance of intraventricularly administered neuropeptides in the peripheral circulation. Brain Res 241: 338-351, 1982. 18. Rowland, N. E. Ingestive behaviour of Syrian hamsters: Advantages of the comparative approach. Brain Res Bull 15: 417-423. 1985. 19. Shapiro, R. E. and R. R. Miselis. The central organization of the vagus nerve innervating the stomach of the rat. J Comp Neurol 238: 473-488, 1985. 20. Smith, G. P., C. Jerome, C. Eterno, B. J. Cushin and K. J. Simansky. Abdominal vagotomy abolishes the satiety effect of cholecystokinin in the rat. Science 213: 1036-1037, 1981: 21. Smith, G. P., C. Jerome and R. Norgren. Afferent axons in the abdominal vagus mediate satiety effect of cholecystokinin in rats. A m J Physiol 249: R638--R641, 1985. 22. Smith, G. T., T. H. Moran, J. T. Coyle, M. J. Kuhar, T. L. O'Donohue and P. R. McHugh. Anatomical localization of cholecystokinin receptors to the pyloric sphincter. Am J Physiol 246 (Reg lnt Comp Physiol): RI27-R130, 1984. 23. van der Kooy, D. Area postrema: Site where cholecystokinin acts to decrease food intake. Brain Res 295: 345-347, 1984. 24. Zarbin, M. A., R. B. Innis, J. K. Wamsely, S. H. Snyder and M. J. Kuhar. Autoradiographic localization of cholecystokinin receptors in rodent brain. J Neurosei 3: 877-906, 1983. 25. Zarbin, M. A., J. K. Wamsley, R. B. Innis and M. J. Kuhar. Cholecystokinin receptors: Presence and axonal flow in the rat vagus nerve. Life Sci 29: 697-705, 1981.