Nutrition and Behavior

Effects of Kappa Opiate Agonists, Cholecystokinin andBombesin on Intake of Diets Varying in Carbohydrate-to-Fat Ratio in Rats1

:,DALE R. ROMSOS,2 BLAKE A. GOSNELL,3 JOHN E. MORLEY* ANDALLEN S. LEVINE

Heuroendocrine Research Laboratory, Veterans Administration Medical Center, Minneapolis, Al/V 55417

ABSTRACT Effects of the dietary carbohydrate-to-fat ratio on opiate-stimulated eating and on naloxone-, Cholecystokinin- andbombesin-suppressed eating were examined. Rats were fed either a high carbohydrate (cornstarch) diet (68% of energy fromcarbohydrate and 12% from fat), an intermediate diet (40% carbohydrate and 40% fat) or a high fat (corn oil and lard) diet (3%carbohydrate and 77% fat). Other rats self-selected from the high carbohydrate and high fat diets. Subcutaneous administrationof naloxone, an opiate antagonist, generally suppressed intake of the high fat diet to a greater extent than intake of the highcarbohydrate diet. Neither Cholecystokinin octapeptide nor bombesin (administered intraperitoneally) exerted preferential suppression of fat intake. The opiate agonists ketocyclazocine and butorphanol tartrate administered subcutaneously at 1000 h preferentially, although not exclusively, stimulated intake of the high fat diet in a dose-dependent manner during the 6-h feeding trial.Repeated daily subcutaneous injections of butorphanol tartrate caused rats to consume more than 50% of their daily intakeduring the 6-h period postinjection; intake during the normal night feeding period was suppressed to maintain total daily intakeequal to that of vehicle-injected rats. We conclude that stimulation of the opioid feeding system contributes to the overeatingoften associated with consumption of a high fat diet. J. Nutr. 117: 976-985, 1987.

INDEXING KEY WORDS:

•food intake regulation •diet selection•dietary carbohydrate and fat

opioids •naloxone •Cholecystokinin •bombesin

The possible role of opioid peptides in the regulationof food intake has been actively investigated during thepast 10 yr (see refs. 1-6 for reviews). One hypothesisis that opioid peptides enhance the desire for palatablefoods and thus promote overeating of these foods(7-11). Much of the evidence in support of this hypothesis has been obtained by use of the opiate antagonists naloxone and naltrexone. Administration of theseantagonists at doses that exert minimal effects on foodintake in rats fed a nonpurified diet suppressed the overeating that occurs in rats fed a nonpurified diet plus achoice of several palatable foods (7, 8). To determinewhether opiate antagonists preferentially suppress intake of a specific nutrient, Marks-Kaufman and Kana-rek (12) permitted rats to self-select from food cupscontaining carbohydrate, fat or protein. Naloxone suppressed intake of both fat and carbohydrate, althoughintake of fat was suppressed at lower doses of naloxoneand for a longer time than intake of carbohydrate.

Certain types of stress (such a mild tail pinch) induceovereating in rats, perhaps due to activation of an en

dogenous opioid system (2). One report demonstratedthat stress-induced eating (cold swim) leads to elevatedintake of a high fat diet (when rats were fed single dietsvarying in carbohydrate-to-fat ratio), which was inhibited by naloxne (13). But another report suggested thatstress-induced eating (tail pinch) results in preferentialsucrose hyperphagia (when rats were fed a nonpurifieddiet plus liquid diets high in sucrose or fat), which wasalso inhibited by naloxone (10). Thus, it is unclear fromthese studies with opiate antagonists whether the opioidsystem preferentially influences intake of a specificenergy-yielding nutrient.

Only limited attempts have been made to determine

'Supported by the Veterans Administration.2On leave from Michigan State University when this research was

conducted.3Present address: Department of Physiology, University of Texas

Health Science Center, Dallas, TX 75235."Present address: Geriatric Research, Education and Clinical Cen

ter, Sepulveda VA Medical Center, Sepulveda, CA 91343.

0022-3166/87 $3.00 ©1987 American Institute of Nutrition. Received: 7 October 1986. Accepted: 23 December 1986.

976

by guest on July 11, 2011jn.nutrition.org

Dow

nloaded from

EFFECTS OF PEPTIDES ON FOOD INTAKE 977

whether opiate agonists (as opposed to the studies withantagonists mentioned above) selectively increase intake of palatable foods. Marks-Kaufman and Kanarek(14) examined self-selected intake of carbohydrate, fatand protein in rats injected with morphine. Rats injected with morphine selected a higher intake of fatand a lower intake of carbohydrate and protein over a6-h period than saline-injected rats. A confounding factor in interpretation of these data is that morphine failedto stimulate total energy intake above the intake ofcontrol saline-injected rats. The failure may be relatedto the specific opioid receptors with which morphineinteracts. There are at least three types of opioid receptors [mu, kappa and delta (15)] and highly selectiveagonists at these three receptors have been shown tostimulate food intake (16). Therefore, data on nutrientselection obtained in rats injected with morphine (apreferential mu receptor agonist) may not completelyreflect the opioid-linked feeding system.

Opiate agonists with selectivity for the kappa receptor stimulate intake of nonpurified diet as well as intake of a palatable diet composed of powdered non-purified diet, sweetened condensed milk and water (11,16-18). It appears that kappa agonists stimulate intakeof the palatable diet to a greater extent than intake ofthe nonpurified diet, although direct comparisons havenot been made (11). Effects of kappa receptor agonistson nutrient selection have not yet been reported.

Opioid peptides and the peptide cholecystokinin,which have opposite effects on food intake, may interact to modify feeding behavior because cholecystokininantagonizes certain opioid-mediated responses (6, 19).

Dietary fat, which in some situations appears to stimulate opioid-mediated eating, is also a potent stimu

lator of cholecystokinin release (20). The extent to whichdiet composition influences the ability of cholecystokinin to suppress intake has received limited attention(21).

The purpose of the present study was to examineeffects of the dietary carbohydrate-to-fat ratio on opiate-stimulated eating and on naloxone-, cholecystokinin-and bombesin-suppressed eating in rats. Bombesin wasincluded in comparisons with cholecystokinin as another example of a peptide that inhibits eating wheninjected peripherally (4, 22). Intake of liquid diets mightbe stimulated by activation of eating or drinking centers. Consequently, only dry diets were utilized in thepresent study. The carbohydrate and fat sources wereprimarily in the forms of starch and triglycérides, respectively, both of which required digestion prior toabsorption. Rats were fed the diets for at least 1 wkprior to testing. In some experiments only a single dietwas fed to a rat; a change in carbohydrate or fat intakein these rats caused a concomitant change in total energy intake. In other experiments rats self-selected between two diets varying in carbohydrate-to-fat ratio;these rats were able to change carbohydrate and fatintake independent of total energy intake.

MATERIALS AND METHODS

Male Sprague-Dawley rats (Biolab, St. Paul, MN)weighing 70-90 g were individually housed in metalcages with wire-mesh floors. Lights were on from 0700

to 1900 h daily and room temperature was maintainedbetween 22 and 26°C.Rats had ad libitum access to a

nonpurified diet (Purina Rat Chow, Ralston Purina Co.,St. Louis, MO) for a 2-d adaptation period and then wereassigned to the experimental diet groups. Three purifieddiets were formulated on an equal energy basis; onediet was high in carbohydrate (CHO diet), the secondcontained intermediate concentrations of carbohydrateand fat (INT diet) and the third was high in fat (FATdiet) (Table 1). During the first week of each experimentrats were weighed and gently handled four times tominimize stress during the subsequent feeding tests.

Experiment 1. Rats (40/diet) were fed one of the threeexperimental diets (Table 1). Effects of naloxone (naloxone hydrochloride, E. I. DuPont de Nemours and Co.,Wilmington, DE) on nocturnal eating were examinedat the beginning of the dark cycle (1900 h) on d 7. Eightrats fed each diet were injected subcutaneously with 0,0.01, 0.1, 1.0 or 10.0 mg naloxone hydrochloride (in0.9% NaCl) per kilogram of body weight. They wereimmediately returned to their home cages and givenpreweighed jars of diet. Jars were briefly removed 1, 2and 4 h later to measure nocturnal food intake. Foodhad been removed from the cages 2 h prior to injectionof naloxone to preclude spontaneous meals immedi-

TABLE 1

Diet formulations'

Amount in diet

Ingredient CHO INT FAT

CaseinMethionineCholine

chlorideAINvitaminmix2AINmineralmix2CelluloseCorn

oilLardComstarchTotalEnergy

density,kcal/gEnergy|%)fromProteinFatCarbohydrate21.20.30.21.03.55.05.0—63.8100.03.9201268g/390

kcal21.20.30.21.03.55.05.012.335.984.44.620404021.20.30.21.03.55.05.028.3—64.56.120773

'Assumed that casein contained 90% protein and that the apparent

metabolizable energy values of protein, fat and carbohydrate were 4,9 and 4 kcal/g. Diets were formulated on an equal energy basis.Consequently, the INT and FAT diets contained fewer grams perunit energy than the CHO diet. All diets contained equal proportionsof protein, vitamins and minerals per unit energy.

2See réf.23 for composition.

by guest on July 11, 2011jn.nutrition.org

Dow

nloaded from

978 ROMSOS ET AL.

ately before testing. To measure effects of naloxone oneating induced by fasting (20 h), food was removed at1400 h on d 8 and naloxone was injected at 1000 h ond 9. Food intake was measured 1, 2 and 4 h later.

Effects of the kappa opioid receptor agonist ketocy-clazocine (Sterling-Winthrop Research Institute, Rens-selaer, NY) on eating were tested at 1000 h on d 12.Ketocyclazocine (0, 0.1, 0.5, 1.0 or 10.0 mg/kg body wt)was injected subcutaneously in a 1:1 solution of meth-anol and 0.1 N HC1 (8 rats/group). Food intake wasmeasured 1, 2, 4 and 6 h later. Food had been removedfrom the cages l h prior to injection to preclude spontaneous meals immediately before testing. Rats injected with 0 or 10 mg ketocyclazocine/kg body weightcontinued to receive daily injections at 1000 h on d 13,14, 15 and 16 to test effects of chronic agonist administration on eating. Food intake was measured at 1, 2,4, 6 and 24 h after each injection.

Experiment 2. The protocol for this experiment wasidentical to that in experiment 1 except that the kappaopioid agonist butorphanol tartrate (Bristol Laboratories, Syracuse, NY) replaced ketocyclazocine. Rats fedthe three diets (40 rats/diet) received naloxone injections on d 7 (nocturnal eating) and 9 (fasting-inducedeating) as in experiment 1. They then received subcutaneous injections of butorphanol tartrate (0.0, 0.1, 0.5,1.0 or 10.0 mg/kg body wt) at 1000 h on d 12. Butorphanol tartrate was dissolved in 0.64% NaCl with citricacid (3.3 mg/mL) and sodium citrate (6.4 mg/mL) added.Food intake was measured 1, 2, 4 and 6 h postinjection.Effects of chronic butorphanol tartrate administration(10 mg/kg body weight) on food intake were assessedon d 13-16 as in experiment 1.

Experiment 3. Rats (n - 60) in this experiment weregiven access to two food cups rather than the singlechoice presented in experiments 1 and 2. Each rat received the CHO diet and the FAT diet (Table 1). Theposition of the two food cups within the cages wasrotated daily. Effects of naloxone on nocturnal eatingand on fasting-induced eating from the two food cupswere examined on d 7 and 9, respectively, as describedin experiment 1. On d 12 rats were injected with butorphanol tartrate as described in experiment 2. Foodintake from the two cups was measured 1, 2, 4 and 6h after injection.

Experiment 4. Effects of cholecystokinin octapep-tide and bombesin on nocturnal eating and on fasting-

induced eating were explored. Subjects for this experiment were the rats in experiments 1-3 that had received 0.1, 0.5, or 1.0 mg opiate agonist/kg body weighton d 12. All rats continued to be fed the same dietsthey had consumed during experiments 1-3.

Rats from experiment 1 were injected in traperi to-neally with 0, 5, 10, 20 or 40 jig cholecystokinin oc-tapeptide/kg body weight (cholecystokinin fragment 26-

33 octapeptide, Boehringer Mannheim Biochemicals,Indianapolis, IN). The peptide was dissolved in 0.9%NaCl (which contained 0.1 mg NaHCO3 and 0.1 mg L-

cysteine/mL of 0.9% NaCl). Injections were at 1900 hon d 16 and 21 to measure nocturnal eating responsesand at 1000 h (after a 20-h fast) on d 18 and 23 tomeasure the fasting-induced eating response. Food in

takes were recorded 0.5, 1 and 2 h after injection. Withthe replication of each feeding condition a total of 8-10 rats fed each diet received each dose of cholecystokinin octapeptide. Each dose of cholecystokinin octapeptide was represented each day and no rat receivedthe same dose of peptide more than once.

Effects of bombesin on nocturnal and fasting-induced

eating were examined in rats from experiment 2. Bombesin (Sigma Chemical Co., St. Louis, MO) dissolvedin 0.9% NaCl (pH 3.5) was injected intraperitoneallyat 0, 5, 10, 20 or 40 n-g/kg body weight. The protocolfor the injection schedule and food intake measurements was identical to that used when cholecystokininoctapeptide was injected.

Cholecystokinin octapeptide was also injected intorats permitted to self-select from the CHO and FAT

diets. These rats were from experiment 3. Injections ofcholecystokinin octapeptide followed the protocol usedto examine effects of the peptide on intake of rats fedsingle diets.

Statistics. Data in each experiment were analyzedwith an analysis of variance (ANOVA) at each timepoint (24). Drug dose, diet and day (for the chronic studies) were main factors. Diet was a repeated-measures

factor in the diet choice studies; in the chronic studies,day was treated as a repeated-measures factor. In allother cases, drug dose and diet were analyzed as be-tween-group factors. After these two- and three-factor

ANOVAs were performed to determine main and interaction effects, separate one-factor ANOVAs were

performed for each diet group or diet condition. Basedon the error terms from these one-factor ANOVAs,Dunnett's test (one-tailed, P < 0.05) was used to com

pare the groups receiving each drug dose to the vehiclecontrol group. It should be pointed out that these wereplanned comparisons and were therefore tested regardless of the outcome of the one-factor ANOVAs.

RESULTS

As expected, rats fed the higher fat diets (INT andFAT) consumed more energy and gained more weightduring the first week than rats fed the CHO diet (Table2). Only in experiment 2 did rats fed the FAT diet consume more than rats fed the INT diet. Cumulative energy intake was not monitored beyond the first weekbecause the feeding trials interrupted ad libitum intake.After 2 wk rats fed the CHO, INT and FAT diets weighed194 ±4, 216 ±5 and 223 ±5 g (mean ±SEMfor 80rats/group,- combined results from experiments 1 and 2).

Effects of naloxone, an opiate antagonist, on energyintake in rats fed single diets are presented in Fig. 1.Results of experiments 1 and 2 were similar; data were

by guest on July 11, 2011jn.nutrition.org

Dow

nloaded from

EFFECTS OF PEPTIDES ON FOOD INTAKE 979

TABLE2Energy intake and body weight gain1

Experiment CHO diet INT diet FAT diet

Energy intake, kcal/wk382 ±9" 446 ±7b 447 ±6b399 ±6' 435 ±10" 469 ±6C

Body weight gain, g/wk41 ±2- 51 ±lb 52 ±1"45 ±1' 52 ±lb 54 ±lb

'Mean ±SEMfor 40 rats fed the diets for 1 wk, except n = 38 for

the FAT diet in experiment 1. Initial weights of rats in experiments1 and 2 averaged 93 ±1 and 99 ±1 g, respectively. Means withdifferent superscript letters are significantly different (P<0.05) asdetermined by LSD (least significant difference).

thus combined and presented in one figure for ease ofcomparison. Energy intake measured at the start of thedark period was unaffected by diet composition and wasonly modestly suppressed by naloxone at 1 and 2 h.Rats first subjected to a 20-h fast consumed more energy than they did during nocturnal feeding (Fig. 1).Under these conditions of fasting-induced eating, a significant naloxone x diet interaction was observed. Rats

naloxone (0.1 mg/kg body wt) suppressed intake in ratsfed the FAT diet than in rats fed the CHO diet (1.0 mg/kg body wt).

Injection of ketocyclazocine, an opiate agonist, initially sedated the rats; consequently at l h energy intake was low (data not presented). By 2 h, intake of ratsinjected with the highest dose was still low, but thosethat received intermediate doses started to eat (Fig. 2).Stimulatory effects of ketocyclazocine on eating wereclearly evident at 4 h, with only modest eating occurring between 4 and 6 h. Ketocyclazocine increased eating at lower doses and to a greater extent in rats fedthe FAT diet than in rats fed the CHO diet. This resulted in a significant ketocyclazocine x diet interaction at each time point. All vehicle-injected rats consumed minimal amounts of food during the 6-h feedingtrial and at no time did vehicle-injected rats fed the

FAT diet consume more than rats fed the CHO diet(Fig- 2).

Butorphanol tartrate, another opiate agonist, was amore effective stimulator of eating than ketocyclazocine (Fig. 2). As observed with ketocyclazocine, butor-phanol tartrate stimulated intake of the FAT diet more

CHO DIET INT DIET FAT DIET

more energy than rats fed the CHO diet. Naloxone suppressed intake more effectively in rats fed the FATdietthan

in rats fed the CHO diet; a 10-fold lower doseofCHO

DIET INT DIET FATDIET24«

16w

8<zo

24CE

LU16zLU81

HOUR0--

FASTED*•-NOCTURNAL-''.v

*•*~~~*~~K-*^Ì•k+ t•2

HOURS-

^"-o\

*^^SI*^N

*T1^.-*

*0

0.01 0.1 1.0 10.01

HOUR¿•"¿•-.I\'•*

-,*-Nt-*s|~-.o.*•2

HOURSk—-o.I\V

\t »'.y^'JL--£+

+ ^>» '«

•0

0.01 0.1 1.0 10.0If—It

1HOUR\0- N,DJJ-D;

*\;\*•r-'*'"~*\^o•r

7\ 2HOURS\/*<N»

\^^^""r'At"

^^4'O.-NAN-D

J-.»-N0

0.01 0.1 1.0 10.024m

16ma8

-<zH3524

ü1JTn

16£.8168_(0oW-u

245z

16i

8ELUZLU241682

HOURSBUTORPHANOLO--

TARTRATE*KETOCYCLAZOCINE•-

t__¿.-Õ..,_j

.f-^^r^v4

HOURS•

*•.

£.^

*:é¿—

^*—r~^T''^.ie

HOURS*I

'y•*v-^:2

HOURS•*h'-y'

Ë;jni^^_4

HOURS*

.i"T/

'i'./T/

.•

*^^_»-^*^'f

7 T it6

HOURS ,/.*

/'I/

•r'•

/ f-2

HOURS0--

B,D•-K,K-D+'

J^'tr^f^'

'.\4

HOURS *l\

0-B•-K.D.K-O /'PJS/\m-f-^ik*^

.e

HOURS»p0--

B, B-D/•-

K, D, K-D/i*'.•^¿r"-241682416824168

mm

NALOXONE: MG/KG BODY WEIGHTFIGURE 1 Cumulative energy intake for 1 and 2 h in rats

injected with naloxone hydrochloride at the start of the darkperiod (nocturnal) or after a 20-h fast (fasted). Each point represents the mean ±SEMfor 15 or 16 rats (data combined fromexperiments 1 and 2). *-t Significant differences from thecorresponding vehicle group (P < 0.05) as determined by Dun-nett's test. N, D and N-D in the far-right panels indicatesignificant naloxone, diet and naloxone x diet interactioneffects (P < 0.05) within a time point. Data obtained 4 h afternaloxone injection are not presented; naloxone continued tohave a significant although diminishing effect on intake, butthe diet and naloxone x diet effects in fasting-induced eatingwere no longer significant.

0 0.1 0.5 1.0 1O.O 0 0.1 0.6 1.0 10.0 0 0.1 0.5 1.0 10.0

BUTORPHANOL TARTRATE or KETOCYCLAZOCINE:

MG/KG BODY WEIGHT

FIGURE 2 Cumulative energy intake for 2, 4 and 6 h inrats injected with ketocyclazocine (experiment 1) or butor-phanol tartrate (experiment 2) at 1000 h. Each point represents the mean ±SEMfor seven or eight rats. *'t Significant

differences from the corresponding vehicle group (P < 0.05)as determined by Dunnett's test. B, K, D, B-D and K-D in

dicate significant butorphanol tartrate, ketocyclazocine, diet,butorphanol tartrate x diet and ketocyclazocine x diet interaction effects [P < 0.05) within a time point. Data obtained1 h after drug injections are not presented.

by guest on July 11, 2011jn.nutrition.org

Dow

nloaded from

980 ROMSOS ET AL.

than intake of the CHO diet, although the butorphanoltartrate x diet interaction was significant only at 6 h.Rats fed the CHO diet and injected with 10 mg butorphanol tartrate/kg body weight consumed 16 ±2 kcalin 4 h and 18 ±2 kcal in 6 h whereas those fed theFAT diet consumed 25 ±4 kcal in 4 h and 30 ±4 kcalin 6 h.

Administration of ketocyclazocine and butorphanoltartrate was continued daily for 4 d in rats given thehighest dose (10 mg/kg body wt). Energy intake of ratsgiven butorphanol tartrate chronically is presented inFig. 3; data obtained with ketocyclazocine followed thesame trend as observed with butorphanol tartrate andthus are not presented. Butorphanol tartrate stimulatedenergy intake during each 6-h injection period, and, asobserved following acute administration of butorphanol tartrate, intake was stimulated to a greater extent in rats fed the FAT diet than in rats fed the CHOdiet (significant butorphanol tartrate x diet effect). Energy intake during the 6-h postinjection period also in-

CHO DIET INT DIET FAT DIET

48s

32^ÃœJ16SIz2

72oc1g

4824^g

sg¡60%X0«°

40%CD«S

20%ZZ^r

m-J0«

ujOOLÜ8

HOURSO-.

BUTORPHANOL.TARTRATE•VEHICLE*

**'

* • .-*

~*^9^rTr—?24

HOURS.

>^¿_-4."^S>-«'"'"*:•T

..,-•'-*

*'**i

,,1234e

HOURS>—

6—¿[.o'*

* •.*•

—•—•—«24

HOURS.

.,*—*_.i^Y--v:^.°

'.•Jt>-

—-0* —"^•.&-'*

* *•*»

—( —«—t•1234DAYe

HOURS,

..A•,6"'*

•*Li

—ΗΗ i—I24

HOURSè--*^è—

*-*,&~~^i«•''•l>-^

•*»

**•—

•—r-f12344832

mm16

S^Z172

ü>48

S24O

IM«

moniis60%

egm~*40%

CgZHl20%

»§(—

\rn»§•<

FIGURE 3 Effect of chronic butorphanol tartrate administration (10 mg/kg body wt daily at 1000 h) on energy intakeduring the 6-h period immediately postinjection (intake wasmeasured at 1, 2, 4 and 6 h postinjection, but only data forthe 6-h period are presented for ease of comparison) and during the entire day (experiment 2). Each point represents themean ±SEMfor eight rats. *Significant difference betweenvehicle- and butorphanol tartrate-injected rats as determinedby Student's t-test (two-tailed, P < 0.05). The following significant effects (P < 0.05) were observed when intake per 6h and percentage of energy consumed in 6 h were analyzed:butorphanol tartrate, butorphanol tartrate x diet, day andbutorphanol tartrate x day interaction. During the 24-h period, butorphanol tartrate, day and butorphanol tartrate xday effects were significant (P < 0.05).

creased with successive daily injections of butorphanoltartrate, with the most pronounced increase betweend 1 and d 2 (Fig. 3).

Rats chronically administered butorphanol tartratecompensated for the stimulatory effect on intake during the 6 h immediately postinjection by depressingintake during the remaining 18 h of the day. Consequently, daily energy intake was not increased in butorphanol tartrate-injected rats (Fig.3). Rats fed the FATdiet demonstrated the greatest degree of compensation;they consumed 30-48 kcal during the 6-h period post-injection and then limited their intake during the normal night feeding period to approximately 20-25 kcalto maintain total daily intake essentially comparableto that of vehicle-injected rats (Fig.3). Rats fed the CHOdiet and injected with butorphanol tartrate consumedapproximately 45% of their daily intake during the6-h postinjection period. Vehicle-injected rats consumed less than 10% of their daily intake between 1000and 1600 h. Body weight gains were unaffected by butorphanol tartrate; control and drug-treated rats fed theCHO diet gained 34 ±3 and 35 ±1g/5 d, respectively;corresponding values for rats fed the INT diet were 35±2 and 39 ±1 g/5 d, and for rats fed the FAT dietvalues were 41 ±3 and 39 ±3 g/5 d, respectively.

When fed single diets, rats have only two choices: toeat or not eat. By permitting rats to select from twodiets differing in composition it should be possible todetermine whether opiate antagonists and agonists exertpreference for carbohydrate or fat independent of totalenergy intake. Thus, rats were adapted to select foodfrom two cups differing in carbohydrate (CHO diet) andfat (FAT diet) content for 1 wk. Ad libitum intake was427 ±7 kcal during the week and body weight gainwas 47 ±l g (Fig. 4). They weighed 212 ±5 g after

X(0•o(DU^ili<

601--

40ENERGYMOD

FATDIETB

CHODIET45

6160±3%±3%±4%-t-\»l*!604020

1-2 3-6

DAYFIGURE 4 Energy intake in rats permitted to self-select

for 1 wk from two food cups, one contained the FAT diet andthe other the CHO diet (see Table 1 for composition) (experiment 3). Bars represent mean ±SEMenergy intake for 60rats weighing 99 ±l g initially. Percentage of energy consumed from the CHO diet during the indicated days is givenabove each bar.

by guest on July 11, 2011jn.nutrition.org

Dow

nloaded from

EFFECTS OF PEPTIDES ON FOOD INTAKE 981

being fed the diets for 2 wk. These values approximatethose observed in experiments 1 and 2, where rats fedthe single INT diet consumed 440 kcal and gained 52g in 1 wk (Table 2) and weighed 216 ±5 g after 2 wk.Rats permitted to select consumed approximately 60%of energy from the CHO diet and 40% from the FATdiet (Fig. 4). The proximate composition of the combined diets (20% of energy from protein, 39% from fatand 41% from carbohydrate) equaled that of the INTdiet (Table 1).

As observed when vehicle-injected rats were fed single diets, fasting-induced eating in rats permitted toselect from two food cups was initially more robustthan nocturnal eating (Fig. 5). By 4 h, however, cumulative intake of nocturnally fed rats essentiallyequaled that of fasted-refed rats; this response paralleled that observed when single diets were fed. Nalox-one depressed energy intake in fasted-refed rats to a

FASTEDNOCTURNAL241681

HOURND

FATDIETB

CHODIETilhIC¡a*I1

HOURm

f,.,BÕJUJ24168

m0ZitLU<

24HZ

16^O

8OCLU2

HOURSN,DfiIIfiI.n*ÃŒl

*i i •2

HOURSDj^n

ñ11Ãi i -m3DO24

"*216

HZ8

my?Z

0

241684

HOURSDtT¡il1i1(T,1A*! 4 HOURS N,Dr]•fifiIfiIi24168

0 0.01 0.1 1.0 1O.O 0 0.01 0.1 1.0 10.0

NALOXONE: MG/KG BODY WEIGHT

FIGURE 5 Cumulative energy intake for 1, 2 and 4 h inrats permitted to self-select from two diets and injected withnaloxone hydrochloride after a 20-h fast (fasted) or at the startof the dark period (nocturnal) (experiment 3). Each bar represents the mean ±SEMfor 12 rats. The clear bar signifiesintake of the FAT diet and the hatched bar intake of the CHOdiet; total intake is represented by the sum of the two bars.Asterisks above a bar indicate significant reduction in totalenergy intake compared to vehicle-injected rats and asteriskswithin a bar indicate significant reduction in intake of thatdiet compared to vehicle-injected rats (P < 0.05) as determined by Dunnett's test. N and D indicate significant effects

(P < 0.05) of naloxone and diet, respectively, within a timepoint.

greater extent than in nocturnally fed rats. The depression in total intake resulted primarily from a reductionin intake of the FAT diet with only minimal reductionin intake of the CHO diet. The same relative discrimination against consumption of the FAT diet occurredin both fasted-refed and nocturnally fed rats, althoughonly in fasted-refed rats did the results attain statistical

significance (Fig. 5). All doses of butorphanol tartrateinjected, except the lowest, stimulated total energy intake in rats permitted to select from two food cups (Fig.6). Intakes of both the FAT and CHO diets were increased, although the relative increase in intake of theFAT diet generally exceeded that of the CHO diet (Fig.6).

Because opioid peptides and cholecystokinin may interact to modify feeding behavior (6) it was of interestto examine food intake responses of rats to cholecystokinin under the dietary conditions used when opiateagonists were tested. Injection of cholecystokinin oc-tapeptide markedly suppressed energy intake at 0.5 h

(0O

Ooctuiu

7.55.02.52

HOURSQ

FATDIETEg

CHO DIET*

*BTJ*rI

«i:t7.55.02.5

1510520151054

HOURS*.

*B,D¿i

[gf.rrr.ir1*ia6

HOURS«B,

D*!è

È\iA1pt*t>î151052015105O 0.1 O.S 1.0 10.0

BUTORPHANOL TARTRATE:MG/KG BODY WEIGHT

FIGURE 6 Cumulative energy intake for 2, 4 and 6 h inrats permitted to self-select from two diets and injected withbutorphanol tartrate at 1000 h (experiment 3). Each bar represents the mean ±SEMfor 12 rats. The clear bar signifiesintake of the FAT diet and the hatched bar intake of the CHOdiet; total intake is represented by the sum of the two bars.Asterisks above a bar indicate significant increase in totalenergy intake compared to vehicle-injected rats and asteriskswithin a bar indicate significant increase in intake of thatdiet compared to vehicle-injected rats (P < 0.05} as determined by Dunnett's test. B and D indicate significant (P <

0.05) effects of butorphanol tartrate and diet, respectively,within a time point.

by guest on July 11, 2011jn.nutrition.org

Dow

nloaded from

982 ROMSOS ET AL.

in all nocturnally fed rats; it was less effective at 1and2h (Fig. 7). Although intake of the FAT dietexceededthat

of the CHO diet, no cholecystokinin x diet interaction was evident. Cholecystokininoctapeptidessuppressed

energy intake in fasted-refed rats aswell.Underintakein

ratsthese

conditions, the peptidesuppressedenergytoa greater extent in rats fed the FAT dietthanfed

the CHOdiet at 0.5 h (significantcholecys-tokininx diet effect) but not at the latertimeCholecystokinin

octapeptide alsosuppressedinratspoints.intakeselectingfrom two food cups (Fig. 8). Atleastinthe fasted condition, it appears that the reductionintotal

intake by cholecystokinin octapeptide was primarily due to a reduction in intake of thecarbohydratediet.

Bombesin, a satiety-producing peptide,suppressedintakeas

hadinall rats ina dose-dependent manner (Fig.9),cholecystokinin

octapeptide. No bombesinxdietinteractions wereevident.DISCUSSIONTo

demonstrate reliably stimulation offoodintakeinresponse to opiate agonists, rats were examinedinthedaytime, when baseline intake was low. Undertheseconditions

rats eat 1-4 g nonpurified diet withinsev-30201Oo¿j

30<£

20~0

10OC111zLU302010CHO

DIET0.5HOUR0-FASTED»•-

NOCTURNAL-^ti±j-1

HOUR'1•fr~a-\

¿NÕ--fOfrt...r"*2

HOURS.4

T --& T'TX

•••t

+•'0

5 10 20 40INT

DIET0.5

HOUR15.X•N^jf'f-*-1

HOURá--

.f—

. 4-"5—°•\^j*•yy**^^2

HOURSi-—

4.,K_.?^r<S';>^S:•*0

5 10 20 40FAT

DIET0.5

HOUR.C.D.CD$

•-C,D\*&

.' X .i*\t\£_.I'j

1HOURX^""i'"i—

i•*.*"~T\^»

9 "f.O.-C.D1 ii'

»-C.DF--Ì-Ì.

T.K

S-Vis.-.-I

•0-D.

•-C,D2

HOURS0

5 10 20 4030201Omm3D30

5520

H>10

m0m302010CHOLECYSTOKININ-OCTAPEPTIDE:

¿IGKG BODYWEIGHTFIGURE

7 Cumulative energy intake for 0.5, 1 and 2 hinratsinjected with cholecystokinin octapeptide at thestartofthedark period (nocturnal) or after a 20-h fast (fasted) (ex

periment 4). Each point represents the mean ±SEMfor8-10rats.*-t Significant differences fromthe correspondingve-hielegroup (P < 0.05) as determined by Dunnett's test. C,Dand

C-D in the far-right panels indicate significant cholecystokininoctapeptide, diet and cholecystokinin octapeptidexdiet

interaction effects (P < 0.05) within a time point.rayLU<h->.Ote.LULUC

FUratspcholectartoidi

Lrepresigni(P<omitright

to acicantindiet~^TìprLeJJJCJ.eral

I1takeswerejectetheirrats

(thispoteiperhírecep(25).Ra1thanif

theingoopiatif

r\aree aj.diet.incre

FASTED NOCTURNAL

241680.5

HOURD

«T DIETCO

CHODIETiI*

**ym.•110.5

HOURCU¿i¿i24168

32241681HOUR¿11[giiI1I•r'/tt1

11

HOURCS-È

S Q¿r322416B4032

24168•;.'

•

•h2

HOURS

C*1

?i*¡*i11t

2 HOURSDiíi\Trl-t-l4,lTl403224168

mmsoO

xmjrAO)

0.5 10 2O 40 0.5 10 20 4O

CHOLECYSTOKININ-OCTAPEPTIDE: JJG/KG BODY WEIGHT

FIGURE 8 Cumulative energy intake for 0.5, 1 and 2 h inrats permitted to self-select from two diets and injected withcholecystokinin octapeptide after a 20-h fast (fasted] or at thestart of the dark period (nocturnal) (experiment 4). Each barrepresents the mean ±SEMfor 12-15 rats. Asterisks indicatesignificant differences from the corresponding vehicle group(P < 0.05) as determined by Dunnett's test. Asterisks were

omitted from the individual diet comparisons in the upperright panel (nocturnal, 0.5 h) because the bars were too smallto accommodate an asterisk; intake of each diet was significantly suppressed at each dose of cholecystokinin. C and Dindicate significant effects (P < 0.05) of cholecystokinin octapeptide and diet, respectively, within a time point.

eral hours after injection of opiate agonists (5, 17); intakes of the purified diets utilized in the present studywere of the same general magnitude (Fig. 2). Rats injected with butorphanol tartrate consumed 30-60% oftheir daily intake within 6 h, whereas vehicle-injectedrats consumed less than 10% of their daily intake inthis time period (Fig. 3). Ketocyclazocine was a lesspotent stimulator of eating than butorphanol tartrate,perhaps due in part to different actions at other opioidreceptors in addition to their kappa agonist activities

5).Rats fed high fat diets usually consume more energy

than rats fed high carbohydrate diets. We reasoned thatif the opioid system is involved primarily in stimulating overconsumption of palatable foods, then effects ofopiate agonists and antagonists should be greater in ratsfed a high fat diet than in rats fed a high carbohydratediet. In agreement, the opiate agonists produced a greaterincrease in intake of the FAT diet than of the CHO diet

by guest on July 11, 2011jn.nutrition.org

Dow

nloaded from

EFFECTS OF PEPTIDES ON FOOD INTAKE 983

CHO DIET INT DIET FAT DIET

o*

30

20

10

30

20

10

30

20

1O

0.5 HOUR

o- FASTED«•—NOCTURNAL

1 HOUR

O 5 10 20 40

0.5 HOURo- B,D

2 HOURS

5 10 20 40 Os 10 20 40

30

20

10

30

20

10

30

20

10

zm0

o•

BOMBESIN : JJG/KG BODY WEIGHT

FIGURE 9 Cumulative energy intake for 0.5, 1 and 2 h inrats injected with bombesin at the start of the dark period(nocturnal) or after a 20-h fast (fasted) (experiment 4). Eachpoint represents the mean ±SEMfor 8-10 rats. *-t Significant

differences from the corresponding vehicle group (P < 0.05)as determined by Dunnett's test. B and D in the far-rightpanels indicate significant bombesin and diet effects (P <0.05) within a time point.

(Fig. 6). This occurred when naive rats fed single dietswere injected with the kappa agonists (Fig. 2). The greaterintake of FAT diet than of CHO diet in response to theagonists also persisted in rats given repeated daily injections (Fig. 3). Intake of the FAT diet on d 4 in ratsinjected daily with butorphanol tartrate was 70% greaterthan intake of the CHO diet in the 6-h postinjection

period.Further evidence for opiate-mediated stimulation of

fat intake is provided by the self-selection feeding trial,where the agonists also increased intake of the FATdiet more than intake of the CHO diet (Fig. 6). Theseresults are in agreement with recent reports that adlibitum-fed rats self-selecting from three cups con

taining fat, carbohydrate or protein generally increasedintake of fat more than intake of carbohydrate wheninjected with the opiate agonist morphine (26, 27). Theserats, however, showed the greatest preferential increasein consumption of protein. But when rats were fed foronly 6 h daily, rather than ad libitum, morphine administration failed to stimulate intake of protein (27). Further, the opiate antagonist naloxone failed to suppressintake of protein in rats self-selecting from three cupscontaining fat, carbohydrate or protein, whereas intakeof fat was suppressed (12). Thus, the relationship between protein intake and the opioid feeding systemremains unclear. It would be of interest to examine

effects of opiate agonists other than morphine on protein intake regulation and to focus more specificallyon protein intake by limiting the dietary variables toprotein versus nonprotein energy supplied in a fixedratio of fat to carbohydrate.

If fat intake is linked to the opioid feeding system,antagonists of the system should have differential effects on intake of the diets employed in the presentstudy. First, it should be noted that our results are compatible with the suggestion that the opioid system isinvolved in hyperphagia rather than control of maintenance levels of energy intake (7-10). Regardless ofthe level of intake by vehicle-injected rats (high in fasting-induced FAT-fed rats and relatively low in all noc-

turnally fed rats, Fig. 1), the highest dose of naloxoneinjected suppressed intake to comparable base levels.All groups of rats injected with 10 mg naloxone/kg bodyweight consumed approximately 6-8 kcal after l h and6-10 kcal after 2 h. Thus, basal energy intake in thepresence of high doses of naloxone was independent ofdiet composition. Naloxone exerted differential effectson intake of the diets fed when intake of vehicle-injected rats was high, as occurred in the 20-h fasted-refed rats, but not when intake of the vehicle-injectedrats was low, as occurred in the nocturnally fed rats(Figs. 1 and 5). In agreement with the differential effectsof the opiate agonists on intake of the diets naloxonewas more effective in suppressing intake of rats fed theFAT diet than intake of rats fed the CHO diet whentested after a 20-h fast. Likewise, when rats were offered

a choice of the FAT and CHO diets, naloxone decreasedtotal energy intake primarily by suppressing intake ofthe FAT diet (Fig. 5).

The mechanism whereby the opioid feeding systemstimulated preference for the FAT diet is unknown. Apossibility is that the results are a consequence of the56% higher energy density of the FAT diet than theCHO diet (Table 1). Clearly, ad libitum-fed rats wereable to compensate for the high energy density of theFAT diet by consuming fewer grams of the FAT dietthan of the CHO diet. In response to the opiate agonists,rats fed the FAT diet tended to consume fewer gramsat 2 h, comparable grams at 4 h and more grams at 6h than rats fed the CHO diet (Fig. 2). This suggests thatgrams consumed is not the only factor regulating intakeof the FAT and CHO diets after injection of opiateagonists. Data obtained with the opiate antagonist naloxone also support the suggestion that the results cannot be explained merely on the basis of differences inenergy density of the diets. In rats permitted to selectbetween the FAT and CHO diets naloxone suppressedgrams of the FAT diet consumed without affecting gramsof the CHO diet consumed. In rats fed single diets aftera 20-h fast naloxone suppressed grams of the FAT dietconsumed more than grams of the CHO diet consumed(Fig. 1).

Two lines of evidence support the conclusion thatopiates are capable of exerting selective regulation on

by guest on July 11, 2011jn.nutrition.org

Dow

nloaded from

984 ROMSOS ET AL.

fat intake. First, under conditions where naloxone wasshown to preferentially suppress intake of the FAT diet,neither cholecystokinin nor bombesin exerted preferential suppression of fat intake. Second, norepineph-rine, peptide YY and neuropeptide Y selectively stimulate intake of carbohydrate rather than fat (28, 29).Thus, not all pharmacological manipulations that suppress or enhance eating do so by selectively altering fatintake. Starch and lard were the primary sources ofcarbohydrate and fat, respectively, for the diets in thepresent study. Preference for carbohydrate in responseto the opiate agonists may have been enhanced by inclusion of glucose or sucrose in the CHO diet. Bertiereet al. (10) found that sucrose intake, rather than fatintake, was responsive to opiate agonists and antagonists when the test diets were presented in liquid form(sucrose added to skimmed milk or water, and wholemilk as the fat source) and were available only duringthe test sessions. Systematic studies are needed to resolve the influence of type of carbohydrate, form of diet(liquid or solid) and previous diet history on nutrientselection in response to opiates and to determine thespecific factor(s) associated with the diets (i.e., taste,texture, smell or postingestional factors) that mediateresponses to opiates.

Repeated daily injections of opiate agonists enhancedenergy intake during the 6-h postinjection period (Fig.3). This probably occurred because tolerance developedto the sedative effects of the agonists with repeatedinjections and rats thus initiated eating sooner (datanot presented; see réf.17 for similar data for rats fed anonpurified diet). Based on energy intake of rats fed theFAT diet (nearly 50 kcal in the light phase) after fourinjections of butorphanol tartrate, one would expecttotal daily energy intake to be elevated. But the ratsexhibited a remarkable degree of compensation; thosethat consumed the greatest amount of energy in response to the agonists (i.e., those fed the FAT diet)reduced their dark phase intake to the greatest extent,whereas those that responded less to the agonists (i.e.,those fed the CHO diet) also reduced their dark phaseintake less. As a result of this compensation total dailyintake was not increased in rats injected with kappaagonists. Whether multiple daily injections of agonistwould have increased total daily intake, as has beenobserved with multiple daily injections of norepineph-rine into the hypothalamic paraventricular nucleus ofrats (30), is unknown.

Because cholecystokinin antagonizes some opioid-mediated actions (6, 19) we reasoned that administration of cholecystokinin to rats might mimic the diet-specific responses observed when naloxone was injected. The other possibility was that consumption offat, a potent stimulator of endogenous cholecystokininrelease (20), would cause adaptations that might leadto differential effects after cholecystokinin administration to rats fed high carbohydrate or high fat diets. Noevidence to support these suggestions was obtained.

Intake of all diets fed was essentially equally suppressed in rats injected with cholecystokinin octapep-tide. Likewise, bombesin was equally effective in limiting intake of each diet fed.

In summary, opiate agonists stimulated and naloxone suppressed intake of all diets fed. The magnitudeof response was greater in rats fed the FAT diet thanin rats fed the CHO diet, and when both diets wereavailable simultaneously intake of the FAT diet wasaffected to a greater extent than intake of the CHOdiet. Neither cholecystokinin nor bombesin exhibiteddifferential effects on intake of diets varying in car-bohydrate-to-fat ratio.

ACKNOWLEDGMENTS

We thank Martha Grace and Julie Kniep for assistance with the feeding trials, Chris Oberg for preparation of the figures and Barbara Anderson for typing themanuscript.

LITERATURE CITED

1. MORLEY,J. E., LEVINE,A. S., YIM, G. K. & LOWY,M. T. (1983)Opioid modulation of appetite. Neurosa'. Biobehav. Rev. 7: 281 -

305.2. MORLEY,}. E., LEVINE,A. S. & ROWLAND,N. E. (1983] Stress

induced eating. Life Sci. 32: 2169-2182.3. MORLEY,J. E., LEVINE,A. S., GOSNELL,B. A. & BILLINGTON,C. J.

(1984) Which opioid receptor mechanism modulates feeding?Appetite 5: 61-68.

4. MORLEY,J. E., LEVINE,A. S., GOSNELL,B. A., MITCHELL,J. E., KRAHN,D. D. &. NIZIELSKI,S. E. (1985) Peptides and feeding. PeptidesfNYJ 6 (Suppl. 2): 181-192.

5. COOPER,S. J., JACKSON,A. & KIRKHAM,T. C. 11985) Endorphinsand food intake: kappa opioid receptor agonists and hyperphagia.Phannacol. Biochem. Behav. 23: 889-901.

6. BAILE,C., MCLAUGHLIN,C. L. & DELLA-FERA,M. A. (1986) Roleof cholecystokinin and opioid peptides in control of food intake.Physiol. Rev. 66: 172-234.

7. APFELBAUM,M. & MANDENOFF,A. (1981) Naltrexone suppresses hyperphagia induced in the rat by a highly palatable diet.Phannacol. Biochem. Behav. 15: 89-91.

8. MANDENOFF,A., FUMERON,F., APFELBAUM,M. & MARGULES,D. L.(1982) Endogenous opiates and energy balance. Science (Washington, DC) 215: 1536-1538.

9. DUM, ]., GRAMSCH,CH. & HERZ, A. (1983) Activation of hypothalamic B-endorphin pools by reward induced by highly palatable food. Phannacol. Biochem. Behav. 18: 443-447.

10. BERTIERE,M. C., SY, T. M., BAIGTS,F., MANDENOFF,A. &. APFEL-BAUM,M. (1984) Stress and sucrose hyperphagia: role of endogenous opiates. Phannacol. Biochem. Behav. 20: 675-679.

11. JACKSON,A. & COOPER,S. J. (1985) Effects of kappa opiate agonists on palatable food consumption in non-deprived rats, withand without food preloads. Brain Res. Bull. 15: 391-396.

12. MARKS-KAUFMAN,R. & KANAREK,R. B. (1981) Modificationsof nutrient selection induced by naloxone in rats. Psychophar-macology 74: 321-324.

13. VASWANI,K., TEIWANI,G. A. &MousA,S. (1983) Stress induceddifferential intake of various diets and water by rats: the role ofthe opiate system. Life Sci. 32, 1983-1996.

14. MARKS-KAUFMAN,R. & KANAREK,R. B. (1980) Morphine se-

by guest on July 11, 2011jn.nutrition.org

Dow

nloaded from

EFFECTS OF PEPTIDES ON FOOD INTAKE 985

lectively influences macronutrient intake in the rat. Pharmacol.Biochem. Behav. 12: 427-430.

15. Am, H., WATSON,S. ]., YOUNGE., LEWIS,M. E., KHACHATURIAN,H. & WALKER,J. M. (1984| Endogenous opioids: biology andfunction. Anna. Rev. Newosci. 7: 223-255.

16. GOSNELL,B. A., LEVINE,A. S. & MORLEY,J. E. (1986) The stimulation of food intake by selective agonists of mu, kappa anddelta opioid receptors. Life Sci. 38: 1081-1088.

17. MORLEY,J. E., LEVINE,A. S., GRACE,M. & KNEIP,J. (1982| Aninvestigation of the role of kappa opiate receptor agonists in theinitiation of feeding. Life Sci. 21: 2617-2626.

18. MORLEY,J. E. & LEVINE,A. S. (1983) Involvement of dynorphinand the kappa opioid receptor in feeding. Peptides (NY) 4: 797-800.

19. PARIS,P. L., KOMISARUK,B. R., NATKINS,L. R. & MAYER,D. J.(1983) Evidence for the neuropeptide cholecystokinin as an antagonist of opiate analgesia. Science (Washington, DC) 219:310-312.

20. FORGACS,I. C., ISASCS,P. E. T., TADAS,S., MURPHY,G. M. &DOWLING,R. H. (1983) Meal composition influences plasmaimmuno-relative cholecystokinin (5R-CCK) levels and gallbladder (GB) contraction. Gastroenterology 84: 1157.

21. VANDERWEELE,D., DEEMS,D. A. & GIBBS,J. (1984) Cholecystokinin, lithium, and diet self-selection in the rat: lithiumchloride decreases protein, while cholecystokinin lowers fat andcarbohydrate ingestion. Nutr. Behav. 1: 127-135.

22. GIBBS,j. (1985) Effect of bombesin on feeding behavior. LifeSci. 37: 147-153.

23. BIERI,J. G., STOEWSAND,G. S. & BRIGGS,G. M. (1977) Reportof the American Institute of Nutrition Ad Hoc Committee onstandards for nutritional studies. /. Nutr. 107: 1340-1348.

24. KIRK,R. E. (1982) Experimental Design, Procedures for theBehavioral Sciences, 2nd ed., Brooks/Cole, Monterey, CA.

25. MARTIN,W. R. (1984) Pharmacology of opioids. Pharmacol.Rev. 35: 283-323.

26. BHAKTHARATSALAM,P. & LEIBOWITZ,S. F. (1986) Morphine-elicited feeding: diurnal rhythm, circulating corticosterone andmacronutrient selection. Pharmacol. Biochem. Behav. 24: 911-917.

27. SHOR-POSNER,G., AYAR,A. P., FILART,R., TEMPEL,D. & LEIBOWITZ,S. F. (1986) Morphine-stimulated feeding: analysis of macro-nutrient selection and paraventricular nucleus lesions. Pharmacol. Biochem. Behav. 24: 931-939.

28. LEIBOWITZ,S. F., WEISS,G. F., YEE,F. &. TRETTER,J. B. (1985)Noradrenergic innervation of the paraventricular nucleus: specific role in control of carbohydrate ingestion. Brain Res. Bull.14: 561-567.

29. STANLEY,B. G., DANIEL,D. R., CHIN, A. S. & LEIBOWITZ,S. F.(1985) Paraventricular nucleus injections of peptide YY andneuropeptide Y preferentially enhance carbohydrate ingestion.Peptides (NY) 6: 1205-1211.

30. LEIBOWITZ,S. F., ROOSSIN,P. & ROSSENN,M. (1984) Chronicnorepinephrine injection into the hypothalamic paraventricularnucleus produces hyperphagia and increased body weight in therat. Pharmacol. Biochem. Behav. 21: 801-808.

by guest on July 11, 2011jn.nutrition.org

Dow

nloaded from