The role of a noxious taste in determining food intake in the rat

Physiology & Behavior, Vol. 24, pp. 1027-1030. Pergamon Press and Brain Research Publ., 1980. Printed in the U.S.A.

The Role of a Noxious Taste in Determining Food Intake in the Rat C O N S T A N C E M. K R A T Z 3 A N D D A V I D A. L E V I T S K Y

Division of Nutritional Sciences, and Department of Psychology, Cornell University, Ithaca, N Y 14853 R e c e i v e d 28 July 1978 KRATZ, C. M. AND D. A. LEVITSKY. The role o f a noxious taste in determiningfood intake in the rat. PHYSIOL. BEHAV.

24(6) 1027-1030, 1980.--Adult female rats were given access to cornstarch, fat, and 2 cups of casein ad lib. Sucrose octaacetate was added to one or both casein cups in concentrations of 0, 0. l, or 2.5%. The lower concentrations of SOA had no effect on total caloric intake or food choices, although a taste aversion experiment indicated that the rats could detect SOA at this level. The 2.5% concentration had no effect on total caloric intake. If only one casein cup was treated with SOA, the animals consumed 70% of their protein from the unadulterated casein supply, but total protein intake was unchanged. If both cups were treated with 2.5% SOA, protein intake decreased, but carbohydrate consumption showed a compensatory increase. Fat intake was unchanged. Taste

Palatability

Sucrose octaacetate

Food intake

STUDIES evaluating the importance of the taste characteristics of the diet in feeding have generally shown that taste plays only a minor role in determining the quantity of food consumed. This conclusion is primarily based on the results of studies demonstrating that the addition of unpalatable additives to the food supply may initially produce a decrement in food intake, but if the study is c6ntinued for longer intervals of time the anorexia proves to be transient, and food intake will return to control levels. This effect has been found with a variety of aversive dietary adulterants [ 1, 7, 11]. Further evidence supporting the hypothesis that taste has little effect on total caloric intake comes from studies in which oral sensations are bypassed, and feeding is accomplished by the subject depressing a bar to obtain a direct intragastric injection of food. Rats eating in this manner consume a normal caloric intake, and are capable of responding in a compensatory fashion when the caloric density of the diet is manipulated [3]. Even though these experiments provide evidence showing that chronically manipulating the taste of an animal's food supply or depriving the animal of the taste sensations of food will not affect total caloric intake, other researchers have emphasized the importance of taste both as a factor in the motivation to eat and as a cue in the identification and selection of food. Epstein concludes from early studies of P. T. Young and others that palatable foods have motivational characteristics, since brief exposure to sugar solutions can serve as an effective reinforcer on learning tasks; he suggests that the significance of this role can predominate when the animal experiences little hunger, as in VMH lesioned

Self-selection

Protein

animals [2]. The use of taste to identify and select foods has been demonstrated in studies of amino acid imbalanced diets [10], in taste aversion studies [4], and in studies involving recovery from a specific nutrient deficiency [5]. However, almost all of the work on the effects of taste on the feeding process has examined its effects on the consumption of a single food. A better method to examine this question is to use a self-selection feeding paradigm. Its advantage lies in the fact that it allows the investigator to evaluate within a single experiment the importance of taste in food selection, motivation to eat, and caloric regulation. This experiment extends the investigation of the importance of taste in food selection by examining the effect of adulteration of a protein source with a bitter substance on food selection, nutrient balance, and total calorie intake. EXPERIMENT 1 This experiment was conducted to determine the effect of an aversive food additive on the rat's feeding behavior. Sucrose octaacetate was chosen as the additive because it has a prominent bitter taste and, unlike quinine, which is used more frequently in taste experiments, SOA is physiologically benign. METHOD

Subjects The subjects were eighteen adult female Sprague Dawley rats obtained from Blue Spruce Farms, Inc. The animals were housed in standard wire mesh cages. Overhead lights

1Supported in part by grants from the National Science Foundation (GB-43732), Weight Watchers Foundation, and research funds from the

Division of Nutritional Sciences, Cornell University. 2The authors would like to thank Roger Epstein for his assistance in conducting the second experiment. 3Current address: Monell Chemical Senses Center, 3500 Market Street, Philadelphia, PA 19104.

C o p y r i g h t © 1980 B r a i n R e s e a r c h P u b l i c a t i o n s Inc.--0031-9384/80/061027-04502.00/0

1028

KRATZ AND LEVITSKY

°ii

40

CONTROLsoA

I ~ CHOICE = 30

"6 0

~

0.1

Y

,o k.-.-.,.--

2.5

0

SOA concentration (%) FIG. 1. Effect of SOA adulteration of protein on protein preference.

0

0.1

2.5

SOA concentration in diet

80

~ ~

Procedure

R E S U L T S

The addition of 0.1% SOA to one or both casein cups had no effect on intake of any diet component or on total caloric intake. This concentration did not affect casein preference either; the animals selected as much protein from the SOAtrea_ted casein as they did from the unadulterated cup (Fig. 1). With the 2.5% concentration of SOA, a marked preference for the untreated casein became apparent; of the total protein consumed, only 28% was selected from the cup wit~ bitter tasting casein when an alternative source was present. Total protein intake remained unchanged, however, provided that a source of unadulterated casein was available. If there were no unadulterated casein present, then protein consumption decreased (t=2.69, p<0.025), so that only 22% of total caloric intake was consumed as protein (Fig. 2). (The

(%)

FIG. 2. Effect of SOA adulteration of protein on protein consumption.

were placed on a 12 hour light, 12 hour dark cycle. Room temperature was maintained at 22°C.

The animals were divided into three groups of six rats each. All groups had ad lib access to the three macronutrients: protein in the form of casein, carbohydrate in the form of cornstarch, and a fat mixture consisting of 24% Primex, 36% corn oil and 40% alphacel; these were offered in separate food cups. All foods were supplemented with vitamins (ICN) and minerals (Hegsted salt mix). All animals were adapted to these foods for several months prior to t h e beginning of this study. At the beginning of the testing phase, all groups were provided with two casein-containing cups, as well as cups containing cornstarch and the fat mixture. For Group 1, both casein cups contained unadulterated casein. Group 2 received a 0.1% concentration of sucrose octaacetate in o n e casein source, while for Group 3 both casein sources were made bitter by the addition of 0.1% SOA. This level of diet adulteration continued for 7 days. For the first 4 days each of the two casein cups of Group 2 was placed in the same fixed location. After this time the positions of the two cups were switched for 3 days. Next, the concentration of SOA was increased to 2.5% in the casein cups of Groups 2 and 3 for 17 days. Food intake and food spillage were measured daily. Body weight was measured on alternate days. The data were analyzed using one-way analysis of variance for each treatment period; t-tests were used in post hoc analyses.

--

iii ....

"

CONTROL SOA CHOICE

- -

Eiiii

3

-

0

:0:':

0.1

SOA concentration in diet

2.5

(%)

FIG. 3. Effect of SOA adulteration of protein on total caloric intake.

overall ANOVA approached significance at the 0.05 level, F(2,15)=3.57.) However, despite the decrease in protein intake, total caloric intake (Fig. 3) remained unchanged in these animals at a mean daily value of 68 kcals, since there was a rather precise compensatory increase in carbohydrate intake from a level of 11 kcals during the control period to a value of 27 kcals with the 2.5% concentration of SOA in both casein sources (Fig. 4), t=2.01, p<0.05. (The overall ANOVA approached significance at the 0.05 level, F(2,15)=3.07.) This shift in nutrient balancing was a selective one; as protein intake decreased, carbohydrate consumption increased, but fat intake was unaffected (Fig. 5). These changes in intake patterns did not occur in animals with an untreated source of casein. In this group, consumption of all nutrients remained at control levels, as did total caloric intake. EXPERIMENT 2 Experiment 1 demonstrated that high concentrations of a bitter adulterant can produce marked changes in the rat's pattern of nutrient intake. The lack of effect of the 0.1% concentration was unexpected, however. One possible explanation is that this dose was too low to be discriminable by the animals. In Experiment 2 a taste aversion paradigm was

NOXIOUS TASTE AND FOOD SELECTION

1029

tram CONTROL SOA ~ CHOICE

30

40

mm CONTROL D SOA E ~ CHOICE

30

0

10

o

.

2.5

0

i 0

SOA concentration in diet (%)

0.1

2.5

$OA concentration in diet (%)

FIG. 4. Effect of SOA adulteration of protein on carbohydrate consumption.

FIG. 5. Effect of SOA adulteration of protein on fat consumption.

used to determine whether the animals were capable of discriminating a 0.1% concentration of SOA in protein.

TABLE 1 EFFECT OF LiCIPOISONINGON FOODPREFERENCE(MEAN -+ SD)

METHOD

Subjects Ten naive adult female Sprague Dawley rats from Blue Spruce Farms, Inc. were used as subjects. Lighting was on a 12 hour light, 12 hour dark cycle, and room temperature was maintained at 22°C.

Procedure All animals were adapted to a two hour per day feeding regimen for two weeks. The diet used during this period was Agway ground rat chow. Food intake and body weight were measured daily. Once the rats were adapted to the feeding schedule, as demonstrated by stable food intake and body weight measures, taste aversion testing was begun. The animals were divided into two groups. Group 1 was given access to unadulterated casein for two hours instead of chow on treatment day. Group 2 received casein with a 0.1% concentration of SOA added. At the conclusion of the two hour feeding period, food cups were removed from the cages. Both groups were then given 6 cc of 0.12 M LiCI by intragastric intubation. Both groups were then returned to the two hour per day ground chow feeding schedule. Aversion testing was conducted two days later. On testing day both groups were given access for two hours to two food cups, one containing SOAadulterated casein, the other with pure casein, and two hour food intake was measured, t-tests were used to analyze the data. RESULTS The results of this study are presented in Table 1. On first exposure, rats consumed 6.3 g of pure casein and 5.3 g of SOA-treated casein; this difference was not significant. When tested after lithium chloride treatment, both groups consumed a larger amount of food from the unadulterated source of protein than from the untreated cup, although this preference for the untreated protein was stronger in those animals which had been made ill after consuming SOAadulterated casein. This group of animals chose 97% of their food from the cup containing pure casein, while the preference in the group in which illness had been paired with pure casein consumed only 70% of their total intake from this cup;

Food Pretreatment paired intake (g) with LiCI

Casein Casein + SOA

Post LiCI intake (g) Casein

Casein + SOA

Amount of casein consumed as a percent of total intake

6.3 _+ 1.0

3.9 _+ 0.9

1.6 - 0.6

70.5

5.3 _+ 2.1

4.7 _ 2.9

0.2 _+ 0.3

97.0

this difference was statistically significant (t=3.25; p<0.01). There was no difference between the two groups in total food consumption on test day. These data suggest, then, that the animals could taste the 0.1% concentration of SOA in casein. DISCUSSION Previous studies have shown that the addition of unpalatable substance has no long-term effect on the total consumption of a single composite diet [1, 7, 11]. The results of these studies indicate that bitter tastes in food do not affect total caloric intake when the bitter additive is placed in one of several foods available to the rat. Consumption of a particular nutrient, on the other hand, proved malleable to experimental manipulation. Alteration of the taste of the protein supply can affect the rat's choice of foods provided that the taste cue is sufficiently aversive. When the rat was offered two sources of protein and one was adulterated with 2.5% SOA, the animal significantly depressed its intake of the bitter casein while increasing its consumption of the nonadulterated protein by an equivalent amount. This fairly precise compensation in protein intake suggests the existence of a motivational system operating to maintain a constant level of protein intake. However, when both sources of casein were adulterated with 2.5% SOA, total protein intake was suppressed, although caloric homeostasis was maintained by a selective increase in carbohydrate consumption to balance the reduced casein intake. Thus, this protein regulating system appears to be only weakly motivating since the total protein intake was depressed by the addition of bitter tastes.

1030

KRATZ AND LEVITSKY

A further indication of the existence of protein motivation is that the effect of SOA on food selection was apparent only with the higher concentration. The adulteration of casein with 0.1% SOA produced no changes in feeding behavior, although the results of Experiment 2 indicate clearly that the rats could detect the presence of 0.1% SOA in casein. Thus, even though rats can taste 0.1% SOA, it is not sufficiently aversive to affect choice protein. It is interesting to note that when 2.5% SOA was added to only one source of casein the consumption of that particular diet was not totally eliminated, but continues to contribute about 22% of total casein consumed. This observation demonstrates that either the effect of taste on protein intake is fairly weak or that motivation to consume a variety of foodstuffs is very strong. The protein component of the diet was chosen for manipulation because, of the three macronutrients presented separately to the animals, it is the only one for which there is a substantial dietary requirement. The adult rat requires protein in the amount of 4% of its total caloric intake; this value was selected on the basis of nitrogen balance studies [9]. However, under conditions where there was at least one source of unadulterated casein, these adult rats chose approximately 37% of their total caloric intake as protein. Even when protein intake dropped to a lower level as a consequence of adulteration of all available protein, protein consumption still accounted for 22% of total caloric intake. These findings suggest that behavioral regulation of protein intake exists and the level at which intake is regulated is much higher than the established protein requirement. This raises the possibility that the nitrogen balance technique may

be inadequate for determining protein requirements, and that the use of this technique may give values that are actually below an organism's physiological needs. This idea is supported by work demonstrating that deposition of nitrogen in tissue is maximal when dietary protein intake approaches 45% of total caloric intake [6]. Others have also reported protein intakes in the range of 35 to 45% in self-selecting rats, and have speculated that factors such as total carcass nitrogen may be an important factor in determining protein intake [8]. Finally, since protein is unique among the macronutrients presented separately in this self-selection regimen, the generality of the effects observed here remains to be tested. It may be that the effects of manipulating the taste of one food would differ had we chosen to add the SOA to the cornstarch or the fat mixture, since both of these represent simply nonspecific sources of energy in the rat's diet. The self-selection technique provides a useful tool for evaluating the organism's priorities in feeding. In these studies, by altering the taste of a single food, it was possible not only to examine the role of taste in feeding behavior, but also the relative importance of various nutritional factors in determining the rat's food intake. In all situations studied, normal daily caloric intake was always defended. In some situations this required an adjustment in the consumption of other nutrients. Total protein intake was defended in situations that required a shift from one protein source to another. However, when both protein sources were adulterated, total protein intake was not well-defended, but was depressed by the bitter-tasting SOA. Yet, even in this condition, total daily protein intake remained well in excess of dietary requirements, suggesting that the regulation of protein still played some role in the control of feeding.

REFERENCES 1. Bauer, E. R. Schedule and quinine-induced deprivation in feeding and refeeding. Physiol. Behav. 6: 87-90, 1971. 2. Epstein, A. N. Oropharyngeal factors in feeding and drinking. In: Handbook o f Physiology, Section 6: Alimentary Canal, Vol. 7. edited by C. F. Code. Baltimore: Williams and Wilkins Company, 1967. 3. Epstein, A. N. and P. Teitelbaum. Regulation of food intake in the absence of taste, smell, and other oropharyngeal sensations. J. comp. physiol. Psychol. 55: 753-759, 1962. 4. Garcia, J. and W. G. Hankins. On the origin of food aversion paradigms. In: Learning Mechanisms in Food Selection, edited by L. M. Barker, M. R. Best and M. Domjan. Waco, Texas: Baylor University Press, 1977. 5. Harris, L., J. Clay, F. Hargreaves and A. Ward. Appetite and choice of diet. The ability of vitamin B deficient rats to discriminate between diets containing and lacking the vitamins. Proe. R. Soc. (Sect. B) 113: 161-190, 1933.

6. Hartsook, E. W., T. V. Hershberger and J. C. M. Nee. Effects of dietary protein content and ratio of fat to carbohydrate calories on energy metabolism and body composition of growing rats. J. Nutr. 103: 167-178, 1973. 7. Kratz, C. M., D. A. Levitsky and S. L. Lustick. Differential effects of quinine and sucrose octaacetate on food intake in the rat. Physiol. Behav. 20: 665-667, 1978. 8. Musten, B., D. Peace and G. H. Anderson. Food intake regulation in the weanling rat: Self-selection of protein and energy. J. Nutr. 104: 563-572, 1974. 9. NAS-NRC (USA) Nutrient Requirements o f Domestic Animals. No. 10. Nutrient Requirements o f Laboratory Animals. Washington, DC: National Academy of Sciences, 1972. 10. Rogers, Q. R. and P. M. B. Leung. The control of food intake: When and how are amino acids involved? In: The Chemical Senses and Nutrition, edited by M. R. Kare and O. Mallet. New York: Academic Press, 1977. 11. Scott, E. and E. Quint. Self-selection of diet. If. The effect of flavor. J. Natr. 32: 113-119, 1946.