Behavioral and physiological consequences of intragastric oil feeding in rats

Physiology & Behavior, Vol. 33, pp. 421-426. Copyright©Pergamon Press Ltd., 1984. Printed in the U.S.A.

0031-9384/84 $3.00 + .00

Behavioral and Physiological Consequences of Intragastric Oil Feeding in Rats ISRAEL RAMIREZ

Monell Chemical S e n s e s Center, 3500 Market Street, Philadelphia, PA 19104 R e c e i v e d 10 J a n u a r y 1984 RAMIREZ, I. Behavioral and physiological conseq#ences of intragastric oil feeding in rats. PHYSIOL BEHAV 33(3) 421-426, 1984.--Four experiments were conducted to examine the appropriateness of intragastric feeding of vegetable oil. The first three experiments demonstrated that pairing intragastric feeding with a taste of saccharin, reduced subsequent saccharin preference slightly. A dose of lithium chloride which did not reduce food intake, produced a very strong conditioned aversion. It is therefore difficult to argue on the basis of taste aversions, that any reduction in food intake resulting from intragastric fat feeding is due to malaise. Intragastric fat feeding did not always reduce subsequent food intake; a large reduction in food intake was observed only when non-starved animals were given at least two previous spaced exposures to fat. The effects of oral and intragastric oil feeding on blood levels of triglycerides and free glycerol were examined. Blood triglycerides and glycerol rose sooner and fell sooner following intragastric than after oral oil feeding. Emulsifying the oil did not correct the abnormality; indeed it exaggerated the early rise in blood triglycerides and glycerol. These results indicate that interpretation of studies involving intragastric fat feeding is more complicated than generally recognized. Conditioned taste aversion

Food intake

Fat tolerance

I N T R A G A S T R I C feeding is a commonly used manipulation in behavioral-physiological research (see [2,7] for reviews). Many researchers seem to consider it a benign procedure which permits one to analyze physiological response to nutrients while bypassing oral chemoreceptors. Deutsch [2,3] has challenged this idea because intragastric feeding can produce a learned taste aversion. In other words, repeated pairing of a novel taste with intragastric injections of oil decreased subsequent preference for solutions containing the novel taste. Deutsch assumes that this reduced preference is attributable to malaise induced by intragastric feeding. On the other hand, Maggio and Koopmans [9] reported that intragastric loads reduced subsequent food intake but did not produce a conditioned taste aversion. Holman found that pairing a taste with intragastric feeding actually increased preference for that taste [6]. My interest in these issues was aroused by Maggio and K o o p m a n ' s [9] report that intragastric loads of fat reduced food intake within a few minutes while other research indicated that orally ingested fat did not decrease food intake until two hours following ingestion [5]. Furthermore, comparison o f two experiments on the effects o f fat feeding suggested that intragastric fat produced a greater decrease in subsequent food intake than did orally ingested fat [12]. The f'wst three experiments-reported below examine the ability of intragastric fat to produce taste aversions and reduce food intake. Several improvements over previous studies were incorporated. Rats were fed an oil-water emulsion rather than plain oil because emulsification improves digestion of intragastric loads of fat [13]. Intragastric loads and taste stimuli were given via catheters in order to provide good control over stimulus presentation without undue trauma. Rats in Experiments 2 and 3 were neither food nor water deprived, in order to avoid complications resulting

Fat digestion

from these treatments. The efficacy of intragastric fat as an aversive stimulus was compared to a dose of lithium chloride, which does not reduce food intake. This comparison allows one to evaluate the claim that intragastric fat is an unusually potent stimulus for producing taste aversions [2]. EXPERIMENT 1 METHOD

Subjects Twenty four Charles River (Wilmington, MA) CD rats, weighing 322_+27 g (mean_+S.D.) at the time of surgery, were used. In this and the following experiments, rats were maintained on a 12/12 hr light/dark cycle at an ambient temperature of approximately 24°C. All procedures were performed in the light part of the day/night cycle. In all experiments, Purina Laboratory Chow (3.7 kcal/g) and tap water were available ad lib except as noted.

Procedure Rats were fitted with silastic gastric catheters according to the method of Young and Deutsch [14]. Prior to surgery the animals were anesthetized with ketamine (90 mg/kg) and acepromazine (1 mg/kg) occasionally supplemented with ether. After a two week recovery period, the rats were food deprived for 24 hours and then permitted to eat wet mash diet (50% water and 50% powdered Purina Laboratory Chow) for 3 hours. The next day, the rats were again permitted to consume the mash for three hours. On the experimental day, the rats were allowed to eat three grams of the mash diet flavored with 0.1% sodium saccharin. One-half

421

--

RAMIRt'.Z 0.7

20

0) O

A

t-

0.5

F~ !ii;i;iii!iiii!!}ili

e-f-

LL

0.3

o

:::::::::::::::::::::::: .:..:.:~.:,:,:.:,:~.~ :::::::::::::::::::::

~:~:!:i:i /:!?!:~~.~:~:i:i:i:i:i:i:?3 :::::::::::::::::::::::

10

~:i!~i~;iiiii:;ili!!~!: ::::::::::::::::::::::

0.1 ii!iiii!??;!i i;: !iiii:!:

Control

Lithium

16.7% oil

Control

50% oil

Itlh311~

16.7% oil

FIG. 1. The effects of intragastric administration of 3 ml water (control), 0.1 molar lithium chloride, 16.7% oil-water emulsion, or 50% oil-water emulsion on subsequent food intake. The star indicates a statistically significant difference between the control and 513%oil group (p<0.01). Vertical bars are standard errors.

FIG. 2. The effects of intragastric administration of water (control), lithium chloride, or oil-water emulsion on preference for saccharinflavored food. The star indicates that the lithium treatment significantly reduced saccharin preference compared to control rats (p<0.01). Vertical bars are standard errors.

hour later, the rats were given intragastric injections of 3 ml of either plain water (control), 0.1 M lithium chloride, 50% oil-water emulsion (stabilized with 1% phosphatidyl choline 12.2 kcal/3 ml) or 16.7% oil-water emulsion (stabilized with 0.3% phosphatidyl choline 4 kcal/3 ml). One half-hour later they were permitted to eat mash without saccharin for three hours. The next day, they were given a choice of saccharinflavored and plainfood for four hours.

intake than lithium on the day of infusion. Lithium, however, is more effective in modifying food preference. Thus, there is no direct correspondence between reduction of food intake and conditioned aversions.

Statistical Analysis Food intake on the day of intragastric infusion was analyzed by analysis of covariance using the previous day's intake as the covariate. All other analyses were based on analysis of variance followed by comparisons with controls using Dunnet's test [4]. RESULTS

Administration of 50% oil-water emulsion reduced food intake on the same day, while the other treatments did not, F(3,20)=13.6, p<0.001, Dunnet's test for controls versus 50% oil-water p<0.01, see Fig. 1. Total caloric intake (including oil) did not decrease, F(3,20)=2.3, p>0.1. The only treatment which significantly reduced preference for saccharin-flavored food the next day was lithium, F(3,20)=3.9, p<0.025, Dunnet's test for controls versus lithium p<0.01, see Fig. 2. The two oil groups did show a trend for reduced saccharin preference, which was not significant even when the two oil groups were pooled (p>0.05). DISCUSSION

Intragastric oil is more effective in reducing total food

EXPERIMENT 2 It is possible that the failure to obtain a statistically significant conditioned aversion is due to the use of food deprived rats or to the use of flavored food or to the use of only one training trial. The next experiment used nondeprived rats, tested with flavored solutions, and given three training trials. METHOD

Procedure Rats were fitted with silastic gastric catheters according to the method of Young and Deutsch [ 14] and simultaneously with oral-cheek catheters (one per rat) according to a modification of the method of Berridge and Grill [ 1]. Both catheters were glued with cyanoacrylate to each other at the top of the head and sewn to the skin and underlying muscles. Catheters were not anchored to the skull as described by Berridge et al. [1]. Rats were anesthetized with ketamine (90 mg/kg) and acrepromazine (1 mg/kg), occasionally supplemented with ether, and given ampiciUin (10 rag/rat) to control infections. After a three week recovery period, the rats were given daily intraoral injections (0.3 ml) of 0.1% sodium saccharin for three days, each injection immediately followed by one of three intragastric injections: 3 ml deionized water, 3 ml of 50% corn oil in water (stabilized with 1% phosphatidyl choline), or 3 ml 0.15 M lithium chloride. Each rat received

INTRAGASTRIC OIL F E E D I N G

423 TABLE 1 PERCENT SACCHARINPREFERENCEAND INTAKE* Experiment 2

Intragastric Treatment Saccharin Preference Saccharin Intake (ml) Water Intake (ml) N

Oil

Lithium

Pt

88 ± 4% 111 ± 17 11 ± 3 10

7 ± 4% 5± 3 45 ± 1 10

0.001 0.001 0.001

Water

Nothing

Saccharin

Pt

95 ± 2% 102 ± ll 9± 3 10

86 ± 6% 75 ± 10 4± 2 12

61 -+ 11% 54 ± 15 20 ± 5 10

0.02 0.05 0.025

Water 98 ± 1% 128 ± 12 2± 1 10 Experiment 3

Oral Treatment Saccharin Preference Saccharin Intake (ml) Water Intake (ml) N

*Means S.E. intake/rat. ~According to analysis of variance.

the same substance on all three days. Three days after the last intragastric oil treatment, the rats were offered a choice of 0.1% sodium saccharin and deionized water to drink for 24 hours.

Statistics Statistical significance was estimated with analyses of variance or, when appropriate, by individual t-tests. Results are presented as means___s.e. RESULTS Saccharin preference differed markedly among the three treatment groups (Table 1). Rats given intragastric water drank 98% of their fluid from the saccharin bottle; rats given intragastric oil drank 88% of their fluid from the saccharin bottle; and rats given intragastric lithium drank only 7% of their fluid from the saccharin bottle, F(2,27)= 2.51, p <0.001. Individual t-tests revealed that the three groups all differed from each other at the 2.5% level or better. Surprisingly, the groups did not differ much with respect to 24 hour food intake on the days they were given intragastric injections: water rats ate 24.6_-.0.6 g/day, oil rats ate 24.0_+0.5 g/day, and lithium rats ate 26.3_+0.7 g/day (means_+s.e. F(2,27)=3.7, p>0.05). An analysis of covariance on food intake using body weight as a covariate resulted in a statistically significant group difference in food intake (p <0.05), but the only effect was that lithium treated rats ate slightly more than the other two groups. The results were the same on each of the three infusion days (trials and conditions interaction, p>0.2). However, total caloric intake (including oil) was higher in rats given intragastric oil (100.0 kcal/day) than in rats given intragastric water (90.2 kcal/day, F(2,27)=4.4, p<0.025 Dunnet's test p<0.01). DISCUSSION Intragastric oil produced only a weak conditioned taste aversion. Lithium chloride produced a very strong taste aversion. Neither treatment reduced food intake. It is clear that malaise (as assayed by tests of conditioned aversion) does not necessarily result in reduced food intake.

EXPERIMENT 3 The failure of intragastric oil to reduce 24 hour food itake was surprising. In the second experiment, the rats were given intragastric oil on three consecutive days and none of the rats had any previous experience with vegetable oil. In other studies conducted in this laboratory [5,12], in which intragastric fat did reduce 24 hour food intake, rats were given many exposures to (~il in several spaced trials. The second experiment also differs from previous studies in that saccharin was presented via mouth catheter immediately prior to intragastric infusion. The present experiment was designed to examine these factors. METHOD

Subjects Thirty-two male Charles River CD rats, weighing 337+ 15 g ( m e a n - s . d . ) at the time of surgery were used.

Procedure Rats were fitted with gastric and oral catheters using the same procedures as in the previous experiment. After surgery they were given penicillin (60,000 U/rat) to control infections. After a two week recovery period, the rats were given an intra-oral injection of 0.3 ml water, 0.3 ml 0.1% sodium saccharin, or nothing immediately followed by an intragastric injection of 3 ml of 50% corn oil in water (the same mixture used in the previous experiment). Each rat received the same treatment combination three days later and again three days after the second time. Food intake in the 24 hours immediately following intragastric oil injection was compared to the mean of the day before and the day after the test. Three days after the last intragastric oil treatment, the rats were offered a choice of 0.1% sodium saccharin and deionized water to drink for 24 hours. RESULTS Intragastric oil feeding had very little effect on food intake in the first two trials but substantially decreased food intake

424

RAMIREZ

,~m :3 on" Ow g_u.

2

u_

1

N W S TRIAL 1

N W S TRIAL 2

N W S TRIAL 3

FIG. 3. The effects of intragastric oil on food intake in Experiment 3. N is for rats that were not given intraoral infusions before intragastric oil; W and S are for rats that were given water or saccharin respectively before intragastric oil. Food intake is presented as the difference between baseline and the day ofoil feeding. Accurate regulation of total calorie intake would require a difference score of 3.3 g. Vertical bars are standard errors.

4

E

EMULSION

03 Itl

==3

T

o >. J L9 hI..-

j:/l

2

"

X

1¢ O.AL on the third trial (Fig. 3, F(2,58)=4.0, p<0.025). Oral administration of water or saccharin did not affect the magnitude of the food intake reduction; there were no significant differences among the three different oral treatment groups (all p>0.2). Once again, intragastric feeding appeared to be mildly aversive (Table 1). When the rats were tested for saccharin preference, rats given intra-oral water before intragastric oil drank 95% of their fluid from the saccharin bottle, rats given oral saccharin drank only 61% of their fluid from the saccharin bottle, and rats given no intra-oral fluid drank 86% of their fluid from the saccharin bottle, F(2,29)=5.3, p<0.02.

0 0

I 1

I 2

I 1 3 4 HOURS

I 5

I 6

FIG. 4. The effects of three different methods of oil feeding on plasma triglycerides in Experiment 4. Vertical bars are standard errors.

EXPERIMENT 4 The previous experiments demonstrate that intragastric oil administration is at least mildly aversive. The next experiment examined the physiological effects of oral and intragastric oil administration in order to provide some insight into the factors contributing to the aversiveness of intragastric oil feeding. The effects of emulsification were also studied. The basic approach used is analogous to the widely used carbohydrate tolerance test; a small meal of fat was given and changes in blood levels of triglycerides were measured. Blood levels of free glycerol are also reported because they provide some' insight into the physiological processes involved. METHOD

Subjects Ten male Charles River CD rats, weighing 338-+8 (mean_+s.d.) at the time of surgery, were used. As in the previous experiments, food and water were available ad lib but procedures were conducted in the daytime when consumption is minimal.

Procedure Rats were fitted only with gastric catheters, using the same procedures as in the previous experiments. One week after surgery, rats were trained to consume 1.5 ml of corn oil from a small inverted bottle cap attached to the bottom of the cage (see [5,11]). The training consisted of simply placing oil in the cup three times a week. Oil that was not consumed within 24 hours was replaced. By the end of the two week training period, all rats consumed all of the oil placed in the cup, usually within 10 rain. During the following week, the rats were given an intragastric injection of 1.5 ml plain oil and, 3 days later, an intragastric injection of 3 ml of the oil-water emulsion used previously. After this training period, the rats were given a series of three tests each 48 hours apart. Each test consisted of an oral presentation of 1.5 ml corn oil, intragastric injection of 1.5 ml corn oil, or intragastric injection of 3 ml oil-water emulsion. Order of treatments was randomized; each rat was assigned to one of six possible treatment sequences at random. Tail

INTRAGASTRIC OIL FEEDING

425 tion F(6,108)=6.1, p<0.001), suggesting that the observed effects on triglycerides were not due to defects in hydrolysis of blood triglycerides.

0.3

GENERAL DISCUSSION

A

0.2

...I

O tr iii

o

0.1 -

l

_r'.\.\ N

OAAL

\'\.\ 0 0

I 1

I 2

i 3

I 4

I 5

I i 6

HOURS

FIG. 5. The effects of three different methods of oil feeding on plasma levels of free glycerol in Experiment 4. Vertical bars are standard errors.

blood samples (120/~1) were taken 1, 2, 4 and 6 hours after oil feeding and assayed for plasma triglycerides and free glycerol using enzymatic methods described in previous publications [11,12]. RESULTS Rats allowed to consume oil orally showed an inverted U-shaped triglyceride tolerance curve with blood triglycerides peaking at 2-4 hours (see Fig. 4). When the rats were given intragastric straight oil, plasma triglycerides were higher than with oral oil at 1 hour, peaked at 2 hours and by 6 hours were below those observed in the oral group 09<0.05). When the rats were given emulsified oil intragastrically, the same pattern was seen as with intragastric straight oil except that plasma triglycerides reached much higher levels (at 2 hours, p<0.01). Overall treatment effects were statistically significant (interaction of treatment by time F(6,108)=9.1, p<0.001). The general pattern of results for glycerol was similar to that observed for triglycerides (see Fig. 5, interac-

The results confirm the reports by Deutsch et al. [2] that intragastric fat feeding reduces preference for flavors paired with it. However, the observed alterations in preference were small; no aversion to the taste of saccharin was produced. In contrast, a dose of lithium chloride which does not reduce food intake, produces a strong conditioned aversion to saccharin. Thus, the results are also generally consistent with those of Maggio and Koopmans [9] who found that lithium chloride was more potent in modifying taste preferences than intragastric feeding. Their failure to detect any aversive effects of intragastric triolein feeding might have been due to their use of hungry rats. Hungry rats might find intragastric feeding simultaneously rewarding [6] and aversive. The conditioned aversion experiments are ambiguous; one may conclude that they either do or do not indicate that intragastric feeding may affect feeding by abnormal mechanisms. The other findings of the present investigation are much more difficult to dismiss. The degree of suppression of food intake resulting from intragastric feeding is highly dependent on the animal's previous experience with oil. Nondeprived rats with no previous experience with oil showed only small reductions in food intake while rats with at least two or more previous spaced trials with oil showed larger reductions in food intake (Experiment 3, [5,11]). The lack of learning in Experiment 2 indicates that oil feeding trials must be spaced. The importance of learning with respect to intragastric feeding has so far gone unrecognized. It is now clear that the effects of intragastric oil feeding cannot be fully evaluated without knowing the animal's previous experience. Intragastric oil feeding produces a highly abnormal tolerance curve. Emulsifying the oil in water makes the abnormality worse rather than better. The abnormal fat tolerance curves seen with intragastric feeding may be due to the more rapid stomach emptying associated with intragastric feeding [8,10]. It is possible that the rapid reduction in food intake following intragastric oil emulsion [9] might be attributable to such metabolic abnormalities. ACKNOWLEDGEMENTS This work was supported by National Institutes of Health grant AM 31612. R. Threatte showed me how to implant oral catheters and commented on the manuscript. M. Friedman provided valuable advice.

REFERENCES 1. Berridge, K. and H. J. Grill. Relation of consummatory responses and preabsorptive insulin release to palatability and learned taste aversions. J Comp Physiol Psychol 95: 363-382, 1981. 2. Deutsch, J. A. The stomach in food satiation and the regulation of appetite. Prog Neurobiol 10" 135-153, 1978. 3. Deutsch, J. A., F. Molina and A. Puerto. Conditioned taste aversion caused by palatable nontoxic nutrients. Behav Biol 16: 161-174, 1976.

4. Dunnett, C. W. A multiple comparison procedure for comparing several treatments with a control. J Am Stat Assoc 50: 10961121, 1955. 5. Friedman, M. I., N. K. Edens and I. Ramirez. Differential effects of medium- and long-chain triglycerides on food intake of normal and diabetic rats. Physiol Behav 31: 851-855, 1983. 6. Holman, G. L. Intragastric reinforcment effect. J Comp Physiol Psychol 69: 432--441, 1968.

426

7. Houpt, K. Gastrointestinal factors in hunger and satiety. Neurosci Biobehat, Rev 6: 145-164, 1982. 8. Hunt, J. N. Some properties of an alimentary o s m o r e c e p t o r m e c h a n i s m . J Physiol (Lond) 132: 267-288. 1956. 9. Maggio, C. A. and H. S. K o o p m a n s . Food intake after intragastric meals o f short-, medium- or long-chain triglyceride. Physiol Behav 28: 921-926, 1982. 10. Molina, F., T. Thiel, J. A. D e u t s c h and A. Puerto. C o m p a r i s o n between s o m e digestive p r o c e s s e s after eating and gastric loading in rats. Pharmacol Biochem Behav 7: 347-350, 1977. I1. Ramirez, I. and M. 1. Friedman. Metabolic c o n c o m i t a n t s of hypophagia during recovery from insulin-induced obesity in rats. Am J Physiol 245:E211-E219. 1983.

RAMIREZ

12. Ramirez, 1. and M. 1. Friedman. Food intake and blood fucl~, after oil consumption: Differential effects in normal and diabetic rats. Phv~iol Bchm' 3 h 847-850. 1983. 13. Roy, C. C., M. Roulet, D. Lefeburc, L. Chantrand, (3. Lepagc and L. A. Fournier. The role of gastric lipolysis on fat absorption and bile acid metabolism in the rat. lipid~ 14: 811-81~, 1979. 14. Young, W. G. and J. A. Deutsch. The construction, surgical implantation and use of gastric catheters anti a pyloric cuff. ,I .,Veto'osci Methods 3: 377-384, 1981