An alternative explanation for apparent satiating properties of peripherally administered bombesin and cholecystokinin in domestic fowls

Physiology & Behavior, Vol. 39, pp. 191-202.Copyright©PergamonJournals Ltd., 1987.Printedin the U.S.A.

0031-9384/87$3.00 + .00

An Alternative Explanation for Apparent Satiating Properties of Peripherally Administered Bombesin and Cholecystokinin in Domestic Fowls C. J O H N S A V O R Y Agricultural and F o o d Research Council's Poultry Research Centre Roslin, Midlothian EH25 9PS, Scotland R e c e i v e d 21 O c t o b e r 1985 SAVORY, C. J. An alternative explanation for apparent satiating properties of peripherally administered bombesin and cholecystokinin in domesticfowls. PHYSIOL BEHAV 39(2) 191-202, 1987.--This paper describes a series of experiments with domestic fowls designed to investigate different effects of peripherally administered bombesin (BBS) and cholecystokinin octapeptide (CCK8), with a view to gaining a better understanding of their proposed roles as physiological satiety signals. Following intravenous injections of 1-10 txg/kg BBS or CCK8, short periods of complete inhibition of feeding coincided with periods of abnormal gizzard motility, and longer periods of reduced feeding were associated with periods of abnormal gastrointestinal (GI) motility. Increases in heart rate, measured in a separate experiment, coincided with the periods of altered GI and feeding activities. Effects of the peptides on GI motility and feeding were strongly related to dose with CCK8, but not with BBS, and effects of (2 ~g/kg) BBS and CCK8 on feeding were additive. Evidence from conditioned avoidance tests suggested that consequences of (10/zg/kg) BBS and CCK8 injections may be mildly aversive, perhaps more so with BBS, and in another experiment inhibition of feeding by (8/~g/kg) BBS (but not CCKS) was almost abolished in birds tranquillised with a reserpine derivative drug. CCK8 was more potent than BBS at suppressing feeding only when relatively high doses (8 and 10/xg/kg) were injected. It is suggested that peripherally administered BBS and CCK8 may act on feeding in similar ways, with animals being distracted by possible abdominal discomfort associated with the abnormal GI responses. The results indicate that immediate discomfort may be more severe with BBS, but that discomfort associated with CCK8 may last longer. It is further suggested that satiating properties of the peptides are less apparent in situations where animals are less easily distracted by discomfort, when their arousal is reduced, when their perception of discomfort is reduced and when their motivation to feed is increased. Bombesin Heart rate

Cholecystokinin Food intake Feeding activity Satiety Conditioned avoidance Tranquillisers Domestic fowl

BOMBESIN (BBS) and cholecystokinin (CCK) have received much attention in recent years as putative physiological satiety factors. Both are among the growing number of polypeptides that have a dual distribution in the brain and in neurons and endocrine cells in the alimentary tract. They have a wide variety of biological actions, and can function in endocrine, paracrine, neurotransmitter and neuroendocrine roles [11]. As with various mammals, they have short-term inhibitory effects on feeding when administered centrally (intracerebroventricularly) or peripherally (intravenously) in domestic fowls ([48, 50-52]; Savory and Gentle, unpublished data). The purpose of this paper is to examine in more detail the nature and interpretation of effects of peripheral administration of exogenous BBS and CCK in fowls. In the avian gut, there is high BBS-like immunoreactivity in the proventriculus and gizzard (glandular and muscular stomachs), and high CCK-like immunoreactivity in the intestine from the antrum to the ileum [10, 20, 46, 59]. The

Gastrointestinal motility

assumption is that BBS and CCK released in the gut in response to passage of food may provide preabsorptive signals contributing to termination of meals. However, this implies an intermittent rather than a regular flow of food through the stomach and upper intestine, which in herbivorous birds seems unlikely in view of their brief and frequent meals, and the presence of a crop which allows food to be stored in the oesophageal region [15,47]. With mammals, the claims that (peripheral) BBS and CCK may act as satiety signals are based on evidence that their effects are dose-related, they elicit behaviour characteristic of satiety, treated animals drink normal amounts after water deprivation, do not appear ill and are not hyperthermic [1, 23, 24, 26, 29, 30, 39]. Some authors, however, have reported that BBS and CCK produce conditioned (taste and place) aversions in rats [7, 9, 58], though others have found no such effect [29, 30, 36]. There is also evidence that rats habituate to repeated injections and become tolerant to

191

192

SAVORY TABLE 1 MEAN FOOD INTAKE (g MASH) AFTER NON-INJECTION, SALINE, BBS AND CCK8 TREATMENTS (ALL DOSES COMBINED) Non-

10.00-10.15 10.15-10.30 10.30-11.00

Injection

Saline

BBS

CCK8

SED~

SEDz

9.2* 5.0*+ 7.3*

9.3* 5.0*t 6. It

8. It 4.7* 7.5*

7.8t 5.6t 7.7*

0.6 0.6 0.6

0.4 0.4 0.4

*tin each time period, treatment means having the same superscript do not differ significantly (/9>0.05, by analyses of variance). SED~ is the standard error of the difference between non-injection and any other mean, and SED2 is the standard error of the difference betweeen saline, BBS and CCK8 means (both SEDs have 72 degrees of freedom).

chronic infusion of CCK [5, 41,62]. After injections of CCK, patterns of gastrointestinal (GI) myoelectric activity and duodenal motility are quite unlike those seen during normal satiety [8,63], exogenous CCK may suppress food intake by inhibiting gastric emptying [43], and in man abdominal cramps and nausea have been reported after injections of quite a small dose [57]. Other authors, too, have suggested that the reduced feeding observed following CCK administration is due to aversive consequences and not satiety [21]. Like CCK, exogenous BBS also causes abnormal GI motility in mammals [2, 40, 64], and it is even possible that BBS may stimulate release of CCK [19]. Domestic fowls do not really show behaviour that is characteristic of satiety, and do not appear ill after injections of BBS or CCK. In turkeys and fowls, however, intravenous (IV) injections of 0.5-15 /zg/kg body weight of CCK octapeptide (CCK8) cause gizzard inhibition together with increased activity in the upper intestine, and it has been suggested that this abnormal motility may cause abdominal discomfort sufficient to distract birds from feeding [49]. There is evidence that such discomfort may be less distracting when birds have greater motivation to eat [50]. In order to obtain more information on how BBS and CCK might act on feeding behaviour, this paper describes a series of experiments designed to examine associations between various responses to IV injections of the peptides, and possible aversive properties.

E X P E R I M E N T 1 (DOSE R E S P O N S E TEST) Although it has been established that IV injections of CCK8 inhibit feeding in a dose-related way in fowls [48,50], a similar dose response test has not been done with BBS. The first experiment, therefore, tested feeding responses to three doses of BBS and CCK8 in order to compare dose effects and relative potencies of the two peptides. Procedure

Ten immature medium-hybrid hens (Rhode Island Red x Light Sussex) were housed and tested in individual cages in a room illuminated from 07.00-21.00 hr and maintained at about 20°C. They were given access to a standard layers' mash diet (for composition see [52], 16% protein, 11 MJ/kg metabolisable energy) from 10.00-16.00 hr each day, having

been trained to this schedule for a week before testing started. They were tested for two weeks from 15 to 17 weeks of age, when their mean body weights increased from 1.36_+0.05 (SE) to 1.50_+0.06 kg. Food consumption was measured in the periods 10.00-10.15, 10.15-10.30 and 10.30-11.00 hr every day from Monday to Friday. On these days at 10.00 hr, immediately before receiving its food, each bird was injected by wing vein with either 1, 5 or 10 tzg/kg of BBS (Sigma, London) or CCK8 (Squibb Institute for Medical Research, N J), both dissolved in 0.9% NaCI solution (10 /xg/ml), or equivalent volumes of the 0.9% saline, or was handled as for injection but not injected. Every bird received each of the 10 treatments (three each of BBS, CCK8 and saline, and one non-injection) once, in random order, according to a Latin square arrangement. Results

In the first 15 min access to food, after 18 hr deprivation, IV injections of BBS and CCK8 caused significant reductions in intake compared to that after saline and noninjection treatments (Table 1). There was no significant difference between BBS and CCK8 treatments in the first 15 min, but birds ate more after CCK8 than after BBS in the next 15 min, and less after saline than other treatments in the last 30 min of testing. In Fig. 1, food intakes in the first 15 min, with the three doses of saline, BBS and CCK8, are compared proportionally with the non-injection treatment. The reductions in feeding with BBS and CCK8 increased with dose. The CCK8 response then was strongly related to dose (linear regression coefficient, p<0.01, Table 2), but the BBS response only weakly so (p<0.1). The CCK8 regression coefficient differed significantly from the saline one in the first 15 min, but the BBS coefficient did not (Table 2).

E X P E R I M E N T 2 (ADDITIVITY TEST) Effects on feeding of peripherally administered BBS and CCK have been shown to be completely additive in rats [26,56]. The second experiment tested whether this is also the case in fowls. Procedure

The same 10 birds that were used in Experiment 1 were tested in the week after Experiment 1, when they were 18

E F F E C T S OF SATIETY PEPTIDES IN FOWLS

c~

o

193

30 ,, C C K 8

BBS

2O

TABLE 2 LINEAR REGRESSIONCOEFFICIENTSOF FOODINTAKESON IV DOSES OF SALINE(EQUIVALENTVOLUMES), BBS AND CCK8(1, 5 AND 10/zg/kg)

.=

Saline

BBS

CCK8

SE

SED

-0.08* -0.05* 0.15"

-0.15*t 0.01" 0.06*t

-0.24t 0.04* 0.01t

0.08 0.08 0.08

0.07 0.07 0.07

r-

.~

10

O"

o . . . . . 7 "~"

/

J

Saline

0

*tin each time period, regression coefficients having the same superscript do not differ significantly (p>0.05, by analyses of variance). SE is the standard error of the regression coefficient, and SED is the standard error of the difference between coefficients (both have 72 degrees of freedom).

~- 4o I,,,

0

I

I

5

10

10.00-10.15 10.15-10.30 10.30-11.00

Dose ( # g / k g )

FIG. 1. Mean (n= 10) percentage reductions in food intake in the first 15 min (10.00-10.15 hr), relative to the non-injectiontreatment, with three doses of saline (equivalent volumes), BBS and CCK8 (1, 5 and 10/xg/kg).

TABLE 3 MEAN FOOD INTAKE(g MASH)AFTERNON-INJECTION, SALINE, BBS (2 ptg/kg), CCK8(2 tLg/kg) AND COMBINEDPEPTIDE(2 ttg/kg BBS + 2 tzg/kgCCK8) TREATMENTS

10.00-10.15 10.15-10.30 10.30-11.00

NonInjection

Saline

BBS

CCK8

BBS + CCK8

SED

8.0* 4.7* 6.4*

8.3* 4.3* 6.0*

6.9t 4.3* 6.1"

7.2t 4.7* 6.5*

6.0~t 4.8* 7.0*

0.4 0.3 0.6

*t~In each time period, treatment means having the same superscript do not differ significantly (/9>0.05, by analyses of variance). SED is the standard error of the difference between means (with 32 degrees of freedom).

weeks old and weighed 1.56---0.06 kg. They were kept in the same conditions as before, with the same 10.00-16.00 hr feeding schedule, and food intake was measured from 10.00-10.15, 10.15-10.30 and 10.30-11.00 hr every day from Monday to Friday. On these days at 10.00 hr, immediately before receiving its food, each bird was injected by wing vein with either 2/zg/kg BBS or 2/xg/kg CCK8, both dissolved in 0.9% saline (5/zg/ml), or 2/xg/kg BBS plus 2/zg/kg CCK8 in an equal mixture of the two peptides ( 10 /zg/ml), or an equivalent volume of saline, or was handled but not injected. On a body weight basis, therefore, all injections were isovolaemic. Every bird received each of the five treatments once, in random order, according to a Latin square arrangement, with two birds receiving each treatment on each day. Results As in Experiment 1, BBS and CCK8 caused significant reductions in food intake in the first 15 min after IV injection, compared with saline and non-injection treatments (Table 3). There were no differences in intake between treatments in the remaining 45 min of testing. In the first 15 rain the combined peptide treatment (2 + 2 /zg/kg) suppressed feeding more than did BBS or CCK8 alone (both 2 /zg/kg). The additivity of BBS and CCK8 was tested by analysis of vari-

ance, with the null hypothesis that there was no interaction between the two peptides when combined, and hence that the reduction in intake (relative to the saline or non-injection controls) with the combined treatment was equal to the sum of the reductions with BBS and CCK8 alone. This null hypothesis was not rejected with either the saline treatment comparison, t(32)=0.35, NS, or the non-injection comparison ( t = - 0 . 1 4 ) , indicating that, as with rats, effects of BBS and CCK8 on feeding (with this dose) are additive in fowls. EXPERIMENT 3 (GI MOTILITY AND F E E D I N G ACTIVITY RESPONSES) The ways in which the avian gut respond to IV injections of CCK have already been described [49], but GI responses to BBS have not. The third experiment investigated the association between feeding and GI responses to the peptides in two ways, by observing the alimentary tract with an X-ray image intensifier and by making simultaneous recordings of GI motility and feeding activity. RADIOGRAPHICVIDEORECORDINGS Procedure Four immature medium-hybrid hens, aged 14 weeks and

194

SAVORY

TABLE 4 NUMBERS OF GIZZARDCONTRACTIONSAND DURATIONS OF ABNORMALINTESTINALMOTILITYON VIDEORECORDINGS OF GI RESPONSES OF FOUR BIRDSTO TWO IV DOSES OF BBS AND CCK8 Time Before and After Injection (min)

2/xg/kg BBS

8 tzg/kg BBS

2/xg/kg CCK8

8/xg/kg CCK8

Numbers of Gizzard Contractions -3 -2

1 1

1 2

3 3

4 3

-I

1

1

3

4

2 2 3 2

13 13 6 4 2

0 1 3 3 3

0f 2 2 3 3

2 3 4 5 6

,

changes in intestinal motility for 7 and 12 min (Table 4). The abnormal activity of the gizzard appeared as 'pulsating,' or incomplete contractions, and that of the intestine as 'rushing' of material through the duodenum and ileum, first in one direction and then the other, without any apparent coordinated propulsive contractions. This rushing was more or less continuous, and was accompanied by occasional nonpropulsive 'segmenting' contractions [28] in the upper ileum. The bird that received 8/zg/kg BBS was noticeably restless in the first minute after injection. Increased GI activity following treatment with BBS has also been found with rats and dogs [2, 40, 64]. IV injections of 2 and 8/xg/kg CCK8 caused reduced gizzard motility for 3-4 min and abnormal intestinal motility for 8 and 14 min. This abnormal motility was similar to that described previously in [49] and appeared mainly as constriction of the duodenum with segmenting contractions in the ileum, and some rushing of material (as with BBS) in the first few minutes after injection (referred to as 'refluxes' in [49]).

Duration of Abnormal Intestinal Motility (rain) 7

12

8

14

*Incomplete contractions--'pulsating.' tRemained in contracted shape.

weighing 1.20-1.48 kg, were each implanted with a permanent indwelling 8 cm cannula in the right jugular vein three days before testing, using the method described in [53]. Food was withdrawn from the birds at least 20 hr before testing. At the start of testing each bird was infused with 20 ml liquid barium sulphate contrast medium into its crop, a polyethylene tubing extension was attached to its jugular cannula for injecting peptides remotely, and it was placed in a narrow cardboard box in which it could not turn round. The box was placed on a platform in front of an X-ray machine (Siemens Triplex Optimatic Generator, energising 125 kV) linked to an image intensifier with a direct recording facility to a video cassette recorder (Panasonic NV-8200). Recordings were made of either lateral or dorso-ventral aspects of the birds' abdominal regions, both aspects being equally good for analysing GI motility. Each bird was injected 10-15 min after being put in the box, and recording continued for 20-30 min after injection. Treatments were allocated to the four birds at random. One bird received 2/zg/kg BBS, one received 8/zg/kg BBS, one 2/zg/kg CCK8 and one 8/zg/kg CCK8. The concentration of both peptide solutions was 10/xg/ml and sufficient saline was injected immediately afterwards to ensure that all the peptide entered the bird. Measurements were made later from the video cassette of numbers of gizzard contractions seen in each of the three mintues before and six minutes after injection, and of durations of abnormal (altered) intestinal motility after injection (this last measurement was somewhat subjective since abnormal motility ceased gradually rather than abruptly). With only one bird per treatment, the aim of this experiment was simply to provide information with which to interpret the subsequent GI motility recordings. Results IV injections of 2 and 8/~g/kg BBS caused increases in gizzard motility in the first 2 and 4 min respectively, and

RECORDINGSOF GI MOTILITYAND FEEDINGACTIVITY Procedure GI motility was recorded by suturing one extraluminal strain gauge transducer (SGT) to the surface of the gizzard and another to the mid-proximal duodenum in each of seven immature medium-hybrid hens, when they were aged between 12 and 14 weeks. One bird was implanted and tested at a time, over a period of many months. Preparation of SGTs (Micro-Measurements, type SA-06-125AC-350), their implantation and use in several bird species have been described in [12,14], and examples of recording traces from turkeys and fowls can be seen in [12, 14, 49]. The birds were also implanted with permanent indwelling 8 cm jugular cannulae (as in [53]) to allow remote injection of peptide solutions. When they were not being tested the birds were kept in the same housing conditions and fed ad lib on the same mash diet as described for Experiment 1. Subjects were deprived of food for 18 hr (overnight) before each recording session, and there was only one such session in a day. At the start of each test the subject's two external leads to its SGTs were linked to a four-channel recorder (Elcomatic) with conditioning amplifiers (Fylde), and an extension was attached to its jugular cannula. Wires and the tubing extension were brought out through the top of the test cage to allow the bird freedom of movement. This cage, which was in a different room, was provided with a water dispenser and had a hole in its side to allow the bird access to a food pan. The pan contained the same mash as before and it rested on a springy metal bar to which was attached another SGT; this in turn was linked to the four-channel recorder. The bird's pecking at food in the pan caused the metal bar to move, and feeding could thus be recorded simultaneously with motility of the gizzard and duodenum. After connecting up the bird the recorder chart was set to run at 5 cm/min, the bird was given 5 min to settle down, it was then given access to the food in the pan and allowed to feed for 10 rain before being injected remotely with the test solution (time of injection being indicated with an event marker), and recording was terminated 30 rain after injection. Doses of BBS and CCK8 injected were 1, 5 and 10/xg/kg (10/xg/ml), saline was injected immediately afterwards to ensure all the peptide entered the bird, and a control saline treatment was

E F F E C T S OF SATIETY PEPTIDES IN FOWLS BBS 5 p g / k g

BBS l p g / k g

0L

L5

I

8 oL ~120[

Saline

lO,nnH OO O

~: c~ 1.01-

~ .~10"0t

BBS 10 pg/kg

I

I

~2"0 f

195

I

I

I

I

II

nnnnnnnnnnnl nO'Innnnnnn

llnqllnn : 'ln,nnnnn I

nn

8 ok -6.

. 0.

.6 . 12. 18

24 - 6

nn

I

' nnnnnnnnnn

0. . 6. .12 . 18 24 - 6 0' 6' 12 ' 18 ' 24' - 6 Time before and after injection (min)

0

6

12 '

18 2 4

FIG. 2. Mean (n=5) times spent feeding and frequencies of gizzard and duodenal contractions, in each 3-min period in the 6 min before and 24 rain after IV injections of three doses of BBS and one dose of saline.

TABLE 5 ASSOCIATIONSBETWEENFEEDINGACTIVITYAND GI RESPONSES, FROMSIMULTANEOUSRECORDINGS,AFTERIV INJECTIONSOF BBS, CCK8 (THREE DOSES) AND SALINE Total Time Feeding in First 15 min

Time to Start of Feeding

Time to First Gizzard Contraction

Time to Normal Gizzard Contractions

Time to First Duodenal Contraction

Time to Normal Duodenal Contractions

2.7* 3.3* 4.6*

3.1" 3.3* 2.9*

0.6*# 0.4*t 0.4*

3.1"¢ 2.9"~t 3.5*

0.4* 0.4* 0.3*

10.9"? 15.6" 9.2"t

2.5* 4.0* 1.3t 5.7* 1.9

1.8" 5.5* 11.9t 1.1" 2.6

1.7"I 2.0t 3.8~ 0.3" 0.7

2.4"~ 5.0*t 7.9t 0.3~ 1.4

1.2t 1.4t 0.9*t 0.7*t 0.3

8.5t 11.8*t 14.8"t 1.2~; 3.0

BBS

1/~g/kg 5/zg/kg 10/zg/kg CCK8 1/zg/kg 5 p.g/kg 10 p.g/kg Saline SED

All mean (n=5) values are in minutes. Other details as in Table 3. The SEDs have 9 degrees of freedom.

equivalent in volume to the largest dose of peptide. Subjects weighed between 1.08 and 1.72 kg at the time of testing, and treatments were applied in random order. Owing to SGTs ceasing to function, however, some of the birds did not receive all seven treatments and data were eventually collected from five birds for each treatment. From the recordings, total times spent feeding and frequencies of gizzard and duodenal contractions were measured in each 3-min period in the six minutes before and 24

min after injections. The times after each injection until the start of the first subsequent feeding bout, the first subsequent gizzard and duodenal contractions, and the returns to 'normal' gizzard and duodenal contractions were also measured. These last two measurements were somewhat subjective, and were based on returns to the form and amplitude of contractions before injection, rather than on returns to preinjection frequencies. Contractions of the proventriculus, gizzard and duodenum are usually highly synchronised [18],

196

SAVORY

CCK8 lpg/kg

CCK8 51ug/kg

I

~ 20 f

CCK8 lOpg/kg

Saline

I

I

Hn l'HHHnnHnn" Hfi'nnnn HH'nnnnnnHH

.~ 1"0I-

~-~ ol

H,~nnnnnOO

nOMn

,

I

I

I

c 4"0I

nnHHHUH HH HH' onnnnnnHHHHHHHH I

~._N .~_ ~.o

~o I

~ .~'°°f D

8

I

nF1

ot i

flnnn i

-6

I

0

I

I

I I I I

i

An nnnnnnnn

I-I I

I

6 12 18 24-6

0

I

I

I

I

I

I

I

I

I

I

i

6 12 18 24 -6 0 6 12 18 24 -6 Time before and after injection (min)

I

0

I

I

I

I

6 12 18 24

FIG. 3. Details as in Fig. 2., with three doses of CCK8 and one of saline.

280 f

I

Saline 240 t" f l O f l 320f

~os

FIn~n~nnFlnn

I

** • **

320[

~r]fln.

oos

lfl...

,

f HiiiiiiiiOnnnno.r,.nH o 'llfinr]nnn

5pg/kg240L

flflfl

240 t.

320f

I

320[

I',,,

I

5"pg/kg280f f l f l f l 240t. .

-6

10JJg/kg280 240 t .

O

.

I

,,

.

6

.

12

18

,

.

flflfl

~lHF]~]F-~flflflfl 0

,

,

24 30 -6 O Time before and after injection (min)

,

6

,

12

,

18

.

24

.

30

FIG. 4. Mean (n=8) heart rates (beats/min) in each 3-min period in the 9 min before and 30 min after IV injections of two doses of BBS and CCK8 and one dose of saline. Post-injection values were compared with overall mean preinjection values by paired t-test. *p<0.05; **p<0.01; ***p<0.001.

EFFECTS OF SATIETY PEPTIDES IN FOWLS and for every contraction of the gizzard there is normally a burst of one to three duodenal contractions [49], as confirmed from electrical activity [13]. Results

With the control saline treatment, feeding activity declined gradually over the 30 min data period (Fig. 2), presumably reflecting normal satiety on return to ad lib feeding after 18 hr deprivation. The saline injections caused interruptions in feeding for a minute or so (Table 5), but had no effect on gizzard or duodenal motility. Injections of BBS caused breaks in feeding in the first three minutes which coincided (Table 5, Fig. 2) with increased gizzard motility (the 'pulsating' seen on the video recordings), and longer periods of reduced feeding appeared to coincide (Fig. 2) with increased duodenal motility (the 'rushing' seen on video). Effects of BBS on feeding and GI motility did not differ significantly between doses, and only two (GI) responses differed significantly from the saline treatment (Table 5). This was because the sample was small and because variation between birds was high. Mean times spent feeding in 3-min periods after injections of BBS were correlated (p<0.01) negatively with corresponding values of duodenal frequency, but were not correlated with gizzard frequencies (three doses combined, n=24). Times spent feeding after injections of saline were correlated (p <0.05) positively with duodenal frequencies only (n= 8). Injections of CCK8 caused breaks in feeding and inhibition of gizzard motility that were related significantly to dose, and increases in duodenal motility that were not (Table 5, Fig. 3). As with BBS, the return to feeding coincided more closely with recovery of normal gizzard contractions than with duodenal motility (Table 5). Unlike BBS, times spent feeding after CCK8 injections were correlated (p<0.01) positively with gizzard frequencies, and were not correlated with duodenal frequencies (n=24).

E X P E R I M E N T 4 (HEART RATE RESPONSES) Since it has been shown that domestic fowls respond to a variety of potentially alarming stimuli (auditory, visual, handling) by increasing their heart rates [17,33], the fpurth experiment tested the possibility that responses to IV injections of peptides have similarly alarming properties by measuring changes in heart rate. Procedure

Radiotelemetry devices, designed to transmit electrocardiogram (ECG) signals, were implanted subcutaneously over the pectoral muscles [17] of eight adult medium-hybrid hens weighing between 1.96 and 2.59 kg. The devices were fabricated as thin-film hybrid microcircuits and encapsulated in silicone rubber [22]. The same birds were also fitted with permanent indwelling jugular cannulae as in Experiment 3. When not being tested they were kept in the same conditions and fed ad lib on the same diet as in Experiment 1. The birds were tested one at a time in the same cage that was used for the previous experiment (GI motility and feeding activity), and it was surrounded with a wire aerial for receiving the ECG signal. After fitting a tubing extension to its jugular cannula, the subject was shut in the cage, its ECG signal checked on an oscilloscope (Hewlett Packard) and a tape

197 recorder (Racal) switched on to record the signal. The bird was given 15--20 min to settle down and was then injected remotely with BBS or CCK8 (5 and 10 tzg/kg, concentration 10/zg/ml), or 0.9% saline equivalent in volume to the largest dose of peptide. Tape recording was terminated 30 min after injection. Each of the eight birds received all five treatments in random order, each was tested only once in a day and was not deprived of food either before or during testing. ECG signals on the tape recordings were analysed on a computer (Digital Minc-23) by an autocorrelation technique [3] to obtain numbers of heart beats in each minute. Mean numbers of beats per minute were then calculated for each 3-min period in the nine minutes before and 30 rain after injections. To test the significance of effects of injections, paired t-tests were calculated between mean heart beat frequencies in each 3 min after injection and the overall mean frequencies in the nine minutes before injection. Results

IV injections of the two doses of BBS and CCK8 caused significant increases in heart rate for periods up to 18 min or so, whereas injections of saline caused a reduction 6-9 min afterwards (Fig. 4). The times taken for heart rate to return to preinjection (mean) levels did not differ significantly between peptides or doses (by analysis of variance, overall mean 20.4 min).

EXPERIMENT 5 (CONDITIONED A V E R S I O N TESTS) Possible aversive consequences of peripheral administration of both BBS and CCK8 have been indicated in rats by conditioned avoidance of associated tastes or places (see Introduction). Such results are far from consistent, however, and their interpretation is disputed by some [6,25]. In the fifth experiment a similar procedure was used to test aversive properties of BBS and CCK8 in fowls. Only the highest dose used previously (10/xg/kg) was tested, the assumption being that if that was not aversive then lower doses would not be either. Procedure

It was intended originally to use two novel flavours, 0.03 M citric acid and 1 g/l saccharin sodium solutions, as cues for the conditioned aversion tests. However, as the fowls showed a marked preference for saccharin over citric acid, it was decided instead to use coloured solutions, red and blue, for which there was no inherent preference. There were four aversion tests, each lasting 5 days and each with a different group of eight immature light-hybrid (White Leghorn) hens, aged 16--19 weeks and weighing I. 19-1.78 kg at the time of testing. T w ~ o f the tests were with BBS and two with CCK8. In each test the eight birds were housed in the same conditions and fed ad lib on the same diet as in Experiment 1. On days 1-4 (Monday to Thursday) they were deprived of water for 3 hr from I 1.00 to 14.00 hr, then given the coloured solutions to drink for 15 min, then injected by wing vein with either 10/zg/kg of peptide (10/xg/ml) or an equivalent volume of 0.9% saline, then given the coloured solutions for a further 15 min, and then returned to their normal drinking water. In each test half the birds received an injection of peptide on days 1 and 3 and of saline on days 2 and 4, and the other half vice versa; and half the

198

SAVORY TABLE 6 MEAN PROPORTION(%) OF PEP'rIDE-AND SALINE-ASSOCIATEDCOLOURED SOLUTIONSCONSUMEDIN A 15-MINCHOICE PERIOD(DAY 5) n

Peptide Colour

Saline Colour

SED

BBS (10 ~xg/kg) BBS on days 1 and 3* BBS on days 2 and 4 All data

8 8 16

49.4 27.3 38.4

50.6 72.7 61.6

20.9 24.2 16.7

NS (p=0.1) NS

CCK8 (10 ~g/kg) CCK8 on days 1 and 3 CCK8 on days 2 and 4 All data

8 8 16

46.6 39.0 42.8

53.4 61.0 57.2

25.6 26.7 17.9

NS NS NS

*This refers to the order in which peptide and saline injections were given on days 1-4 (see the Procedure section). SED is the standard error of the difference between means. NS, not significant (p>0.1, by paired t-test).

TABLE 7 MEAN FOOD INTAKE(g MASH)AFTERSALINE, DIAZEPAM(0.35 mg/kg), BBS (8/xg/kg), CCK8(8 ~g/kg) AND COMBINED PEPTIDE/DIAZEPAMTREATMENTS

10.00-10.15 10.15-10.30 10.30-11.00

Saline

Diazepam

BBS

BBS/ Diazepam

CCK8

CCK8/ Diazepam

SED

12.7" 5.9* 6.7*

11.2"* 7.3t 8.5*?

10.5% 7.6t 7.5"?

10.0% 7.4t 7.9*?

7.4§ 7.0*? 8.3"t

8.8~:§ 7.5* 9.4*

0.9 0.6 1.1

Details as in Table 3. The SEDs have 50 degrees of freedom.

birds received a red solution (0.5 g/1 ponceau de xylidine) associated with the peptide injection and a blue solution (0.0125 g/l methylene blue) with the saline injection, and the other half vice versa. The coloured solutions were made more conspicuous by presenting them in white dispensers, and amounts consumed in the first 15 min access varied from 5 to 59 ml. On day 5 the birds were deprived of water from 11.00 to 14.00 hr, as before, and each was then given a simultaneous choice of red and blue solutions in two dispensers in front of its cage. The dispensers were removed after 15 min, amounts of each colour that had been consumed were measured, and these were expressed as percentages of the total amount consumed by each bird. Judging from the results and the rather different experimental designs of some previous conditioned taste aversion tests with rats [7,9], it was considered that the order in which injections of peptide and saline were given might influence the degree to which birds discriminated between associated colour cues in the present experiment. To test this possibility, therefore, the data from those birds receiving peptide injections on days 1 and 3, and those receiving them on days 2 and 4, were analysed both separately and together. Data from the two tests with each peptide were combined, and mean percentage intakes of the peptide- and salineassociated coloured solutions were compared by paired t-tests.

Results Total amounts of the red and blue solutions consumed during the 15 min choice period on day 5 varied from 8 to 71 ml. Those birds that received saline injections on the day before the colour choice, and peptide injections on days 1 and 3, showed no discrimination between the peptide- and saline-associated colours with either BBS or CCK8 (Table 6). Those that received peptide injections on days 2 and 4, however, showed greater avoidance of the peptideassociated colour. This avoidance was almost significant in the case of BBS, but not with CCK8, variation in choice between birds being high, and the differences between injection orders were not significant either (by t-test). EXPERIMENT 6 ( T R A N Q U I L L I Z E R TESTS) Inhibitory effects of BBS and CCK on food intake in rats have been reversed when the tranquillizer drug diazepam was injected simultaneously with the peptide [38, 44, 65]. This response could be associated with diazepam's known action as an appetite stimulant in various mammals [38], its apparent ability to block gastric abnormalities induced by BBS [64], decreased arousal and hence possible reduced awareness of aversive effects of the peptides, or a combination of these properties. Diazepam and related compounds

E F F E C T S O F SATIETY PEPTIDES IN F O W L S

199

TABLE 8 MEAN FOODINTAKE(g MASH)AFTER SALINE, BBS (8/~g/kg)AND CCK8(8/J.g/kg)TREATMENTSWHEN BIRDS WERE GIVEN NORMAL WATER (CONTROL) OR PACITRANSOLUTION (0.6 g/l) TO DRINK Control

10.00-10.15 10.15-10.30 10.30-11.00

Pacitran

Saline

BBS

CCK8

Saline

BBS

CCK8

17.4" 6.3* 6.6*

13.5t 8.5t 7.3*

10.0:~ 10.1t 7.5*

16.8" 6.3* 6.6*

15.6"~

10.8~:

6.5* 7.4*

7.1*t 8.0*

SED 1.3

0.8 0.8

Details as in Table 3. The SEDs have 50 degrees of freedom.

are not recognised drugs for birds, although chlordiazepoxide and chlorpromazine cause decreases in arousal and in frequency of stereotypies in canaries [34]. Nevertheless, in the sixth experiment, effects of diazepam and of a recognised avian tranquillizer, the reserpine derivative 'Pacitran' [16], were tested to see if they blocked the inhibition of feeding by BBS and CCK8 in fowls.

DIAZEPAM Procedure Preliminary trials showed that fowls are highly sedated by IV injections of doses of diazepam that block effects of BBS and CCK in rats (1-5 mg/kg). With 0.5 mg/kg birds still lay down and were slightly drowsy for 5-10 min after injection, and 0.35 mg/kg was the highest dose that had no obvious sedative effect. This dose was used for the experiment. Twelve immature light-hybrid hens, aged 15-17 weeks and weighing 1.01-1.43 kg at the time of testing, were housed and tested in the same conditions and fed on the same mash diet as in Experiment 1. They had access to food from 10.00-16.00 hr each day, having been trained to this schedule for a week before testing started. F o o d consumption was measured in the periods 10,00-10.15, 10.15--10.30 and 10.30-11.00 hr on Monday, Wednesday and Friday in two consecutive weeks. On these days at 10.00 hr, immediately before receiving its food, each bird was injected by wing vein with either BBS, CCK8 (both 8 tzg/kg, 10 ~g/ml), an equivalent volume of 0.9% saline, 0.35 mg/kg diazepam (Roche Products Ltd., 0.5 mg/0.5 ml propylene glycol + 0.5 ml saline), or combined injections of BBS and diazepam or CCK8 and diazepam (same doses as the individual treatments). Every bird received each of the six treatments once, in random order, according to a Latin square arrangement, with two birds receiving each treatment on each day. Results As in Experiments 1 and 2, IV injections of BBS and CCK8 caused significant reductions in food intake, compared with saline, in the first 15 min access to food after 18 hr deprivation (Table 7). Injections of 0.35 mg/kg diazepam did not cause increased feeding in the first 15 min, and simultaneous injections of diazepam together with the peptides did not block the inhibition of feeding then by either BBS or CCK8.

PACITRAN Procedure The same 12 birds that were used for the diazepam test were used again for another tranquillizer test. At the end of the diazepam test half of them were given a solution of 0.6 g/l pacitran (metoserpate hydrochloride, Ciba Laboratories Ltd., dissolved in tap water) to drink ad lib instead of their normal drinking water. This concentration did not affect the birds' behaviour in any obvious way, but counts showed that it reduced frequencies of their spontaneous 'headshakes,' a response to alerting stimuli [31], by 54% (mean of 12 birds), and it reduced heart rate by 24% in two birds implanted with radiotelemetry devices (cf. Experiment 4). All 12 birds were kept on the 10.00-16.00 hr feeding schedule as before, and their food intake was measured in the periods 10.00-10.15, 10.15-10.30 and 10.30-11.00 hr on Monday, Wednesday and Friday in the week after the start of the pacitran treatment. On these days at 10.00 hr, immediately before receiving its food, each bird was injected by wing vein with either BBS, CCK8 (both 8/zg/kg, 10/~g/ml) or an equivalent volume of saline. Every bird received each of the three treatments once, in random order, according to a Latin square arrangement, with two birds on pacitran solution and two on normal drinking water (control) receiving each treatment on each day. After the food intake measurements on the Friday, the six birds on pacitran were returned to normal water and the six with normal water were given the pacitran solution instead. The three injection treatments were administered on the following Monday, Wednesday and Friday in the same way as before, and food intake was measured as before. Thus all 12 birds received the three injections on both pacitran solution and normal water. Results With normal drinking water, IV injections of BBS and CCK8 caused significant reductions in food intake compared with the saline treatment in the first 15 min access to food (Table 8), as before. With the pacitran solution, however, intake was reduced with CCK8, but not with BBS. Although the BBS values did not differ significantly between control and pacitran treatments in the first 15 min, the difference between saline and BBS was slightly greater with control drinking water than with pacitran solution (p =0.1, by paired t-test), suggesting that effects of BBS may be reduced in tranquillized birds. Measurements of amounts of food and

200

SAVORY

water consumed between 10.00 and 16.00 hr showed that pacitran had no effect on food intake (compared with the control treatment) but it reduced water intake by 33%.

GENERAL DISCUSSION The results of Experiments 1-4 indicate that IV injections of BBS and CCK8 cause short-term inhibition of feeding in fowls, their effects are additive when administered together, total cessation of feeding coincides with periods of abnormal gizzard motility, inhibition of feeding is associated with a period of increased heart rate, and recovery of normal heart rate and normal feeding activity are associated with recovery of normal GI motility. The conditioned avoidance tests in Experiment 5 suggest that consequences of BBS and CCK8 injections may be mildly aversive, perhaps more so with BBS, whereas in Experiment 6 a tranquillizer alleviated the inhibitory effects of BBS only, suggesting that CCK8 may be more potent. CCK8 has also been shown to be more potent than BBS at inhibiting feeding in rats [24,56] and fowls [52], but in the present study it suppressed food intake more than did BBS only in Experiment 6 (Tables 7 and 8), and not in Experiments 1 and 2 (Tables 1 and 3). This apparent inconsistency may be accounted for by differences in the doses tested. In the instances with fowls where CCK8 was more potent (see also effects on gizzard motility in Table 5) only relatively high doses (8 and 10/zg/kg) were used. When 1 to 10/zg/kg were tested in Experiment 1, effects of CCK8 were strongly related to dose but those of BBS only weakly so (Table 2). Unlike fowls, effects of BBS appear to be closely related to dose in rats [24,56]. Presumably differences in responses between BBS and CCK8, and between doses, reflect variation in both magnitude and duration of effects. In Experiment 3 the birds had all been deprived of food for 18 hr before testing, and this prior deprivation may well have affected the timing of their feeding and GI responses to injections of BBS and CCK8. It is known that normal GI motility recovers sooner in fasted fowls than in undeprived ones after IV injections of CCK8 [49], and birds may be less easily distracted by any associated abdominal discomfort when motivation to eat is high. These two consequences of deprivation probably account for the finding that IV CCK8 is most effective at inhibiting feeding in undeprived birds, and becomes less effective with increasing deprivation [50]. The reason why birds were deprived before Experiment 3 (and before most other peptide experiments) was to increase the probability that birds would be feeding at about the time of the peptide injections. Thus the observed responses in Experiment 3 might have lasted longer had there been less or no prior deprivation. The heart rate responses in Experiment 4, however, lasted for about the same times as the inhibition of feeding and abnormal GI motility in Experiment 3, despite the fact that birds in Experiment 4 were not deprived of food before testing. Apart from this difference, conditions in Experiments 3 and 4 were the same. Judging from the known effects of potentially alarming stimuli on heart rate in fowls [17,33], it seems reasonable to interpret the results of Experiment 4 as reflecting alarm or arousal associated with internal changes caused by the peptide injections. However, the possibility that BBS and CCK8 also have direct effects on heart rate cannot be ruled out. Apart from the possible discomfort associated with abnormal GI responses to BBS and CCK8, CCK was originally named

for its contractile effect on the gall bladder [32], and users of pancreozymin (a proprietary preparation of CCK, Boots, UK) for testing gall-bladder function in humans are warned that too rapid injection (
E F F E C T S OF SATIETY PEPTIDES IN FOWLS

201

CCK may be wholly or partly accounted for by distracting feelings of internal discomfort, particularly at higher doses. The proposal that BBS and CCK act on feeding in similar ways is further supported by the fact that their effects are additive (Experiment 2, [26,56]). If internal discomfort is indeed the main cause of reduced feeding after injections of BBS or CCK, then animals should behave in a more normal way if their perception of this discomfort is reduced or abolished. They do respond less to BBS and CCK, as predicted, when treated simultaneously with tranquillisers or antiemetics (see above), and peripheral administration of CCK does not inhibit feeding or exploratory behaviour in vagotomised rats [4, 37, 45, 54] whose sensory feedback from affected areas may be abolished. However, conflicting evidence comes from the findings that abdominal vagotomy does not alter CCK's satiety effects in sham-fed rats [35], nor does it alter satiety effects of BBS [45,55]. The reason for these differences is unclear. In fowls, vagotomy at the level of the proventriculus does not abolish satiety effects of either BBS or CCK8 (or the associated increases in heart rate, Savory, unpublished data), but this could be because afferent information from areas affected by the peptides does not reach the CNS via the vagus nerve in birds [52]. The altered patterns of GI motility that occur after peripheral administration of doses of BBS and CCK8 that cause suppression of feeding are quite unlike anything seen during normal, spontaneous, contraction sequences or during normal satiety (Experiment 3, [8, 49, 63, 65]). This suggests that levels of circulating BBS and CCK8 following such injections are probably supraphysiological. So far it has not been possible to quantify the normal range of concentrations of endogenous BBS or CCK which could then be re-

lated to hunger and satiety. In a pilot trial using osmotic minipumps (Alzet, model 1701), chronic infusion of a low dose of CCK8 (1/zg/hr) into the jugular veins of fowls caused significant reductions in food intake (not seen with control saline infusions) only on the first day after implanting the pumps, with full recovery on the second day and thereafter, despite the fact that the pumps continued to infuse active CCK8 for a week (Savory and Gentle, unpublished data). Such tolerance of CCK8 was also found in a similar experiment with rats [5]. Despite the claim here that feeding responses to peripheral injections of BBS or CCK do not reflect real satiety, these are just two of at least 10 different peptides that have been found to inhibit feeding. It is quite possible, therefore, that the combined effect of all GI peptides together, that are released endogenously in response to vagal stimulation during feeding [60] and in response to passage of ingesta, could be sufficient to play a part in normal satiety. It is also possible that BBS and CCK (and other peptides) in the brain may contribute towards normal satiety.

ACKNOWLEDGEMENTS I wish to thank S. J. Lucania of the Squibb Institute for Medical Research for kindly providing the synthetic CCK8, Roche Products Ltd. for the gift of diazepam and G. Wenham for use of the Rowett Research Institute's X-ray image intensifier. I am also grateful to C. McCorquodale, D. Waddington and M. Yeomans for help with statistical analyses, J. Hodgkiss and D. Shaw for help with Experiment 3, I. Duncan, A. Carlisle, S. Clark and G. Slee for help with Experiment 4, L. Glasgow for help with Experiment 5 and E. Seawright for help with all the experiments.

REFERENCES

1. Antin, J., J. Gibbs, J. Holt, R. C. Young and G. P. Smith. Cholecystokinin elicits the complete behavioural sequence of satiety in rats. J Comp Physiol Psychol 89" 784-790, 1975. 2. Caprilli, R., P. Melchiorri, G. Improta, P. Vernia and G. Frieri. Effects of bombesin and bombesin-like peptides on gastrointestinal myoelectric activity. Gastroenterology 68: 1228-1235, 1975. 3. Clark, J. S. B. and J. H. Filshie. Autocorrelation techniques for measuring avian heart rates. Med Biol Eng Cornput 15: 656-665, 1977. 4. Crawley, J. N., S. E. Hays and S. M. Paul. Vagotomy abolishes the inhibitory effects of cholecystokinin on rat exploratory behaviors. Eur J Pharmacol 73: 379-380, 1981. 5. Crawley, J. N. and M. C. Beinfeld. Rapid development of tolerance to the behavioural actions of cholecystokinin. Nature 302: 703-706, 1983. 6. Deutsch, J. A. Bornbesin--satiety or malaise? Nature 285: 592, 1980. 7. Deutsch, J. A. and W. T. Hardy. Cholecystokinin produces bait shyness in rats. Nature 266: 196, 1977. 8. Deutsch, J. A., T. R. Thiel and L. H. Greenberg. Duodenal motility after cholecystokinin injection or satiety. Behav Biol 24: 393-399, 1978. 9. Deutsch, J. A. and S. L. Parsons. Bombesin produces taste aversion-in rats. Behav Neural Biol 31: 110-113, 1981. 10. Dockray, G. J. Cholecystokinin-like peptides in avian brain and gut. Experientia 35: 628-630, 1979. 11. Dockray, G. J. and R. A. Gregory. Relations between neuropeptides and gut hormones. Proc R Soc Lond [Biol] 210: 151-164, 1980. 12. Duke, G. E. and T. E. Kostuch. The use of strain gauge transducers to study gastroduodenal motility in turkeys. Poult Sci 54: 1472-1478, 1975.

13. Duke, G. E., T. E. Kostuch and O. A. Evanson. Gastroduodenal electrical activity in turkeys. Am J Dig Dis 20: 1047-1058, 1975. 14. Duke, G. E., O. A. Evanson and P. T. Redig. A cephalic influence on gastric motility upon seeing food in domestic turkeys (Melagris gallopavo), great-horned owls (Bubo virginianus) and red-tailed hawks (Buteo jamaicensis). Poult Sci 55: 2155-2165, 1976. 15. Duncan, I. J. H., A. B. Home, B. O. Hughes and D. G. M. Wood-Gush. The pattern of food intake in female Brown Leghorn fowls as recorded in a Skinner box. Anirn Behav 18: 245-255, 1970. 16. Duncan, I. J. H. and D. G. M. Wood-Gush. The effect of a Rauwolfia tranquillizer on stereotyped movements in frustrated domestic fowl. Appl Anita Ethol 1: 67-76, 1974. 17. Duncan, I. J. H. and J. H. Filshie. The use of telemetry devices to measure temperature and heart rate in domestic fowl. In: A Handbook on Biotelemetry and Radio Tracking, edited by C. J. Almaner and D. W. MacDonald. Oxford: Pergamon Press, 1980, pp. 579-588. 18. Dziuk, H. E. and G. E. Duke. Cineradiographic studies of gastric motility in turkeys. Am J Physiol 222: 159-166, 1972. 19. Erspamer, V., G. Improta, P. Melchiorri and N. Sporanzi. Evidence of cholecystokinin release by bombesin in the dog. Br J Pharmacol 52: 227-232, 1974. 20. Erspamer, V. and P. Melchiorri. Amphibian skin polypeptides active on the gut. J Endocrinol 70: 12P-13P, 1976. 21. Ettenberg, A. and G. F. Koob. Different effects of cholecystokinin and satiety on lateral hypothalamic self-stimulation. Physiol Behav 32: 127-130, 1984.

202 22. Filshie, J. H., I. J. H. Duncan and J. S. B. Clark. Radiotelemetry of avian electrocardiogram. Med Biol Eng Comput 18: 633-637, 1980. 23. Gibbs, J., R. C. Young and G. P. Smith. Cholecystokinin decreases food intake in rats. J Comp Physiol Psycho184: 488-495, 1973. 24. Gibbs, J., D. J. Fauser, E. A. Rowe, B. J. Rolls, E. T. Rolls and S. P. Maddison. Bombesin suppresses feeding in rats. Nature 282: 208-210, 1979. 25. Gibbs, J. and G. P. Smith. Bombesin--satiety or malaise? reply to Deutsch. Nature 285: 592, 1980. 26. Gibbs, J., P. J. Kulkowsky and G. P. Smith. Effects of peripheral and central bombesin on feeding behavior of rats. Peptides 2: 179-183, 1981. 27. Gillette, K., G. M. Martin and W. P. Bellingham. Differential use of food and water cues in the formation of conditioned aversions by domestic chicks (GaUus gallus). J Exp Psychol [Anita Behav] 6: 99--111, 1980. 28. Hightower, N. C. Movements of the alimentary canal. In: Physiological Basis o f Medical Practice, edited by C. H. Best and N. B. Taylor. Baltimore, MD: Williams and Wilkins, 1966, pp. 1196-1241. 29. Holt, J., J. Antin, J. Gibbs, R. C. Young and G. P. Smith. Cholecystokinin does not produce bait shyness in rats. Physiol Behav 12: 497-498, 1974. 30. Houpt, T. R., S. M. Anika and N. C. Wolff. Satiety effects of cholecystokinin and caerulein in rabbits. Am J Physiol 235: R23-R28, 1978. 31. Hughes, B. O. Headshaking in fowls: the effect of environmental stimuli. Appl Anita Ethol ll: 45-53, 1983. 32. Ivy, A. C. and E. Oldberg. Hormone mechanisation for gallbladder contraction and evacuation. Am J Physiol 86: 599-613, 1928. 33. Jones, R. B., I. J. H. Duncan and B. O. Hughes. The assessment of fear in domestic hens exposed to a looming human stimulus. Behav Proc 6: 121-133, 1981. 34. Keiper, R. R. Drug effects on canary stereotypies. Paper read at Animal Behavior Society meeting in Columbus, Ohio, 1968. 35. Kraly, F. S. Vagotomy does not alter cholecystokinin's inhibition of sham feeding. Am J Physiol 246: R829-R831, 1984. 36. Kulkowsky, P. J. Reduction of drinking-associated feeding by C-terminal octapeptide of cholecystokinin-pancreozymin. Behav Neural Biol 29:111-116, 1980. 37. Lorenz, D. N. and S. A. Goldman. Vagal mediation of the cholesystokinin satiety effect in rats. Physiol Behav 29: 599604, 1982. 38. McLaughlin, C. L. and C. A. Baile. Cholecystokinin, amphetamine and diazepam and feeding in lean and obese Zucker rats. Pharmacol Biochem Behav 10: 87-93, 1979. 39. Martin, C. F. and J. Gibbs. Bombesin elicits satiety in sham feeding rats. Peptides 1: 131-134, 1980. 40. Mayer, E. A., J. Elashoffand J. H. Walsh. Characterization of bombesin effects on canine gastric muscle. Am J Physiol 243: GI41-G147, 1982. 41. Mineka, S. and C. T. Snowdon. Inconsistency and possible habituation of CCK-induced satiety. Physiol Behav 21: 65-72, 1978. 42. Moore, B. O. and J. A. Deutsch. An antiemetic is antidotal to the satiety effects of cholecystokinin. Nature 315: 321-322, 1985. 43. Moran, T. H. and P. R. McHugh. Cholecystokinin suppresses food intake by inhibiting gastric emptying. Am J Physiol 242: R491-R497, 1982. 44. Morley, J. E., A. S. Levine, S. S. Murray, J. Kneip and M. Grace. Peptidergic regulation of stress-induced eating. Am J Physiol 243: R159-R163, 1982. 45. Morley, J. E., A. S. Levine, J. Kneip and M. Grace. The effect of vagotomy on the satiety effects of neuropeptides and naloxone. Life Sci 30: 1943-1947, 1982.

SAVORY

46. Rawdon, B. B. and A. Andrew. An immunocytochemical survey of endocrine cells in the gastrointetinal tract of chicks at hatching. Cell Tissue Res 220: 279-292, 1981. 47. Savory, C. J. An investigation into the role of the crop in control of feeding in Japanese quail and domestic fowls. Physiol Behav 35: 917-928, 1985. 48. Savory, C. J. and M. J. Gentle. Intravenous injections of cholecystokinin and caerulein suppress food intake in domestic fowls. Experientia 36: 1191-1192, 1980. 49. Savory, C. J., G. E. Duke and R. W. Bertoy. Influence of intravenous injections of cholecystokinin on gastrointestinal motility in turkeys and domestic fowls. Comp Biochem Physiol [A] 70: 179-189, 1981. 50. Savory, C. J. and M. J. Gentle. Effects of food deprivation, strain, diet and age on feeding responses of fowls to intravenous injections of cholecystokinin. Appetite 4: 165-176, 1983. 51. Savory, C. J. and M. J. Gentle. Brain cholecystokinin and satiety in fowls. Appetite 4: 223, 1983. 52. Savory, C. J. and J. P. Hodgkiss. Influence of vagotomy in domestic fowls on feeding activity, food passage, digestibility and satiety effects of two peptides. Physiol Behav 33: 937-944, 1984. 53. Savory, C. J. and C. J. V. Smith. Are there hunger and satiety factors in the blood of domestic fowls? Appetite 8:1987 (in press). 54. Smith, G. P., C. Jerome, B. J. Cushin, R. Eterno and K. J. Simansky. Abdominal vagotomy blocks the satiety effect of cholecystokinin in the rat. Science 213: 1036-1037, 1981. 55. Smith, G. P., C. Jerome and J. Gibbs. Abdominal vagotomy does not block the satiety effect of bombesin in the rat. Peptides 2: 409-411, 1981. 56. Stein, L. J. and S. C. Woods. Cholecystokinin and bombesin act independently to decrease food intake in the rat. Peptides 2: 431-436, 1981. 57. Sturdevant, R. A. L. and H. Goetz. Cholecystokinin both stimulates and inhibits human food intake. Nature 261: 713-715, 1976. 58. Swerdlow, N. R., D. van der Koy, G. F. Koob and J. R. Wenger. Cholecystokinin produces conditioned placeaversions, not place-preferences, in food-deprived rats: evidence against involvement in satiety. Life Sei 32: 2087-2093, 1983. 59. Timson, C. M., J. M. Polak, J. Wharton, M. A. Ghatei, S. R. Bloom, L. Usellini, C. Capella, E. Solcia, M. R. Brown and A. G. E. Pearse. Bombesin-like immunoreactivity in the avian gut and its localisation to a distinct cell type. Histoehemistry 61: 213-221, 1979. 60. Uvnas-Moberg, K. Release of gastrointestinal peptides in response to vagal activation induced by electrical stimulation, feeding and suckling. J Auton Nerv Syst 9: 141-155, 1983. 61. Watkins, L. R., I. B. Kinscheck and D. J. Mayer. Potentiation of opiate analgesia and apparent reversal of morphine tolerance by proglumide. Science 224: 395-396, 1984. 62. West, D. B., D. Fey and S. C. Woods. Cholecystokinin persistently suppresses meal size but not food intake in free-feeding rats. Am J Physiol 246: R776-R787, 1984. 63. Wingate, D. L., E. A. Pearce, M. Hutton, A. Dand, H. H. Thompson and E. Wunsch. Quantitative comparison of the effects of cholecystokinin, secretin, and pentagastrin on gastrointestinal myoelectric activity in the conscious fasted dog. Gut 19: 593-601, 1978. 64. Young, W. G. Alcohol, diazepam, and morphine sulfate reduces bombesin-induced elevation of gastric pressure and motility in the rat. Soc Neurosci Abstr 6: 165, 1980. 65. Young, W. G., J. A. Deutsch and T. D. Tom. Diazepam reverses bombesin-induced reduction of food intake by abolishing intragastric pressure and motility. Fed Proc 40: 941, 1981.