Appetite and taste preference in growing rats given various levels of protein nutrition

Brain Research Bulletin,

Vol. 27, pp. 417422. 0 PergamonPress pk. 1991.Printedin the U.S.A

0361-9230193 $3.00 + .OO

Appetite and Taste Preference in Growing Rats Given Various Levels of Protein Nutrition MASATO MORI, TERUh41 KAWADA AND KUNIO TORI? Ajinomoto Co., Inc., Life Science Laboratories, Central Research Laboratories 214 Maeda-cho Tots&a-ku, Yokohama, 244, Japan

MOIU, M., T. KAWADA AND K. TORII. Appetite and taste preference in growing rats given various Ievefs of prorein nutrition. BRAIN RES BULL 27(3/4) 417-422, 1991.--The cephalic gustatory stimuli during a meal yield nu~tional info~ation and aid in the effkient control of homeostasis. This study was focused on either appetite for flavored food or feeding behavior in growing male Sprague-Dawley rats under various states of protein nutrition. In fasted rats, endogenous protein degradation was suppressed by ingestion of glucose that was sufficient to meet energy needs. The decrease in the amount of diet intake was compensated by sugar ingestion, except for sucrose. Rats that ingested sucrose exceeded 115%of total energy intake, compared to ingestion of saccharin as a control. Appetite and meal size are primarily dependent upon the dietary protein level, whether it was beyond normal requirements or not and, thus, flavoring by certain taste material is effective for a diet sufficient in protein, but not for a deficient one. In addition, rats fed a diet containing amino acids preferred saccharin and monosodium L-glutamate (MSG) and grew normally. But, when L-tryptophan-deficient diet was offered, the preference for tryptophan was elicited, and then MSG intake was elevated and their growth became normal. However, preference for saccharin never occurred because the level of ~toph~ in blood fluctuated and was not main~n~ within normal limits. The strong preference for sweetness that is evoked by starvation is directly regulated by the negative energy balance. The animals’ primary concern was energy intake and their second concern was protein nutrition, regardless of flavoring. Appetite

Energy intake

Protein nutrition

Taste-flavoring

The acidic L-amino acids, L-glutamic and L-aspartic acid, which are major components of proteins, taste sour as well as savory (umami taste). Yet the sodium salt of both defines the umami taste. Monosodium L-glutamate (MSG) is a typical umami taste substance (4). We previously suggested that taste preference may closely reflect the amount of dietary protein, and that intake of protein beyond its required level for normal growth could be coupled with a preference for umami taste substances (16-18). Rats under severe protein deficiency preferred NaCl, and sometimes glycine, to the umami taste. NaCl may be selected to maintain electrolyte and body fluid balance. Preference for glycine may be indicative of its presumed ability to spare nitrogen and thereby minimize the negative nitrogen balance that the nonprotein diet induces. It is also well known that people who become diabetic display a strong preference for sweetness (especially sugar), reflecting a hunger for glucose in body tissues, muscles, brain, etc., arising as a result of hyperglycemia due to hypoinsulinemia. The altered taste preference indicates that a deficiency of some nutrient has occurred. The strong taste preference ensures that organism’s requirement for the deficient nutrient will be satisfied.

THE cephalic gustatory stimuli during a meal yield nutritional information and aid in the efficient digestion of food (15). When animals detect a food with a bitter and/or sour taste, they exercise caution in ingesting something that might be toxic. They may stop eating and even vomit the food (2,lO). But foods having a familiar or pleasant taste can be swallowed without hesitation. In general, foods having a sweet or mildly salty taste are palatable, but those that are sour and bitter are usually aversive (13). In addition, some of the L-amino acid and 5’-ribonucleotides, which are found in such foods as animal and fish meats following autolysis of either protein or polyribonucleotides, are generally attractive, not only to lower ~cr~rganisms, but also to higher primates, including humans (5,7). The taste, produced by the release of both L-amino acids and 5’-ribonucleotides from foods during chewing, may provide a detection signal for foods that contain protein and which are called “umami” in Japanese, meaning delicious or savory. However, it is not clear that the taste of every individual L-amino acid fits into any of the usual categories of taste qualities as sweet, salty, sour or bitter (14,19). All essential L-amino acids except L-Boone are generally bitter, especially L-tryptophan, whereas nonessential ones and L-threonine are sweet. ._

‘Requests for reprints should be. addressed to Kunio Torii, D.V.M., Ph.D., Ajinomoto Co., Inc., Life Science Laboratories, Central Research ~~mto~es 214 Maeda-cho, Tot&a-ku, Yokohama, 244, Japan.

417

MORI, KAWADA AND TORI1

This study was focused not only on the manipulation of appetite for flavored food but also on the biological role of taste preference in rats under various levels of protein nutrition. METHOD

Male Sprague-Dawley rats (Charles River Japan Inc., Japan) were used throughout. Room temperature and relative humidity of animal rooms were maintain at 22 ‘-c1°C and 60 t 5%, respectively. The lights were turned on at 0700 for 12 h. The rats were supplied with a diet containing various levels of purified egg protein (PEP) (Taiyo Kagaku Co. Ltd., Japan) or casein. The ex~~mental diets were fo~ulated using nutrients having a pharmaceutical grade, previously reported (17). PEP was employed as an ideal protein source for growing rats. PEP was acid-hydrolyzed with 6 N hydrochloride for 24 h and assayed for each amino acid using an automatic amino acid analyzer (835. Hitachi, Ltd., Japan). The amino acid mixture, which was prepared to mimic the composition, was also employed as a protein source for comp~ison with PEP, previously reported (17). The statistical analysis was carried using unpaired Student’s r-test. Experiment 1. The Preference for Sweetness in Rats Under Prolonged Starvation

Rats weighing 22924 g (7 weeks of age, N = 6, in each group). were offered a choice of three sweet-tasting solutions, 4 mM sodium saccharin (hereafter referred to as saccharin), and 0.5 M and 1.0 M glucose, and drink deionized-distilled water (hereafter referred to as water) as drinking solutions under conditions of diet or protein restriction. Each group was supplied with the 15% PEP diet for two weeks, and then introduced to protein or diet restriction. During restriction of diet or protein, rats were housed individually in a metabolic cage, and 24-hour urine samples were collected into a 200 ml conical beaker with 10 ml of saturated boric acid (approximately 48, w/v). There were two control groups allowed to drink water. One was fasted, whereas the other was allowed a nonprotein diet ad lib. Body weight was determined every 2 days. The daily solution intake and food intake were recorded. Urinary nitrogen was assayed by micro-Kjeld~l method (I). Energy intake from glucose solution or nonprotein diet was calculated. Experiment 2. Comparison of Anorexigenic Effect of Sugar Solution Intake in Rats

Rats weighing 127 1-6 g (5 weeks of age, N =6, in each group), were housed indiv~du~ly and ingested a 15% PEP diet, and were given access to both water and one of following sweet-tasting solutions: 1.0 M glucose, 0.5 M sucrose, 1.0 M fructose, 0.5 M maltose, 0.5 M lactose, or 4 mM saccharin as drinking water. Body weight was determined every 2 days. The daily solution intake and food intake were recorded. Energy intake from both the diet and the sweet soiution was calculated for each group.

0.2% quinine hydrochloride (hereafter referred to as quinine) (bitterness). Body weight and food intake were determined every day. ~xperimeni 4. The DiurnaE Pattern of Diet Intake and histogram of Meal Size in Rats Fed Nonprotein or 1.5% Casein Diet with Taste Flavoring

Rats weighing 282 ~tr20 g (9 weeks of age, N = 6. occasionally N=5, in each group), were supplied with nonprotein or 15% casein diet containing taste materials, 5.6% MSG (umami), 0.2% MSG+0.5% GMP (umami), 6.3% citric acid (sourness), 20% sucrose (sweetness), 0.2% quinine (bi~emess) for two weeks. Their diet intake every 10 min was monitored using a strain gauge system. The meal size was defined so that the particular meal had an interval of more than 20 min between the previous and next one. Experiment 5. The Taste Preference in Rats Under L-T~ptophan De~e~en~~

Rats weighing 97rt: 8 g (4 weeks of age, N=5, in each group), were housed together in a large transparent poly-resin cage (30 X 30 X 80 cm) to which 15 drinking bottles in maximum were attached on a single face, and were supplied with a diet containing 20% amino acid mixture with or without L-tryptophan, equivalent to PEP (17), and water ad lib for seven days. Then they were ahowed to select four kinds of drinking solution, 0.125 M DL-tryptophan (bitterness), 0.15 M MSG (umami), 4 mM saccharin (sweetness) and water (tastelessness) for 25 days. Each experimental diet was changed to another one for 32 days. Body weight and food intake were determined every 2 days. The daily solution intake was recorded. RESULTS

Experiment I

In every experimental group, the ingestion of each sweet solution (saccharin or glucose) exceeded that of water. A strong preference for both sweet-tasting compounds, i.e., glucose and saccharin, was observed immediately after food deprivation. Moreover, these preferences were sustained in both groups of rats ingesting one of the glucose solutions. The absolute amount of glucose intake in the two glucose groups was not significantly different from each other. The intake of the saccharin solution declined during the period of starvation, as it did in fasted controls. When the 15% PEP diet was returned, the water and solution intake in the experimental groups and in both control groups returned to the levels reached before restriction (Fig. 1). During restriction of diet or protein, urinary nitrogen excretion in fasted rats that drank saccharin solution or water was essentially the same and much greater than that in fasted rats ingesting a glucose solution (0.5 or 1.O M) or given a nonprotein diet. The energy intakes among these latter three groups were quite comparable to one another (Fig. 2). Experiment 2

Experiment 3. Effect of Flavoring to Nonprotein or 1.5% Casein Diet on the Appetite and Growth of Rats

Rats weighing 142*7 g (6 weeks of age, N =5, in each group), were housed in a cage together, and supplied with nonprotein, 15% casein diet and water ad lib for three weeks. Each cage contained a diet, each with or without (control) a distinctive flavoring: 5.6% (w/w) MSG (umami), 0.2% MSG + 0.5% guanosine 5’-monophosphate (GMP) (umami), 6.3% citric acid (sourness), 20% sucrose (sweetness), 1.8% NaCl (saltiness) and

Rats displayed a strong preference for the sugars, glucose, fructose, maltose, sucrose and lactose, compared with saccharin as a control in the choice paradigm of two bottles, filled with an aqueous solution of sugar or water. The total energy intake of each group ingested sugars, except sucrose, was unchanged. They decreased the consumption of a diet containing 15% PEP, close to the same amount of energy from these solutions of sugar, except sucrose. In the case of sucrose, the degree of decreased consumption of diet was at half, and the total energy

APPETITE,

TASTE PREFERENCE

150

m5awm ‘E 5

(Cal/rat)

Starvation

Starvation

E

419

AND PROTEIN NUTRITION

4mM

100

Saccharin

r Solution

i

n1.OM So,ution

o-

DAY

AFTER

OSM

Glucose Sucrose Water

Water

l.OM

OSM

0.m

Fructose Water

Maltose Water

Lactose Water

4mM

Saccharin Water

OFFERED

SOLUTION

FIG. 1. The pattern of sweet solution intake during starvation in growing rats. The mean intake values for each solution (noted in this figure) arc shown.

FIG. 3. Energy intake composition in rats consumed diet and drinking water containing sweet taste materials. The concentration of each sweet taste material was noted at the bottom of this figure. Both energy intake from diet and sugar solution in each group were calculated and shown as mean value with a standard deviation.

intake was significantly

QKO.01) exceeded at 115% of control group. The order of preference for sugar solution was as follows: 1) sucrose, 2) glucose, fructose, and maltose, 3) lactose (see Fig. 3).

SOLUTION INTAKE 150

Experiment 3 75 The growth and consumption of 15% casein diet containing MSG, MSG+ GMP, or sucrose, was quite comparable to both in controls fed the diet without taste materials. But when rats were supplied with a diet with NaCl, quinine or citric acid, their diet intakes and growth were retarded in this order, throughout the 3 weeks of the experimental period. In the case of the nonprotein diet with the same flavoring offered to rats, the diet consumption and growth in any group were completely retarded, regardless of flavoring. The body weight gain for 3 weeks in groups of rats fed a 15% casein diet with quinine and citric acid

*~~* 0 ENERGY

INTAKE

50

25

0 URINARY

NITROGEN

EXCRETION

200

was suppressed significantly, but that in any other group was unchanged (Fig. 4). In contrast, the degree of the body weight loss under protein restriction was quite comparable to each other, regardless of any flavoring (Fig. 4). Experiment 4

100

The diurnal pattern of feeding in rats fed 15% casein or nonprotein diet with flavoring was recorded using strain gauge system, indicating that nocturnal rhythm was sustained regardless of flavoring, as well as protein restriction (Fig. 5). But meal size of each group was affected by flavoring. When palatable taste materials were added to 15% casein diet, such as MSG, MSG+

0 4mM Solution

Diet

Water

Saccharin

alone

Water

Fast

Fast

0.5M

1M

GlUCOS0

Water Fast

Water

alone Non-protein

FIG. 2. Energy intake and urinary nitrogen excretion during starvation in rats. The treatment of each experimental group was the same as described in Fig. 1, and is also indicated at the bottom of this figure. Mean values and standard deviations are shown for solution intake, energy intake, and urinary nitrogen excretion.

GMP or sucrose (Fig. 5A), their meal size was exceeded, rather than the cases of aversive compound, quinine or citric acid (Fig. 5A). This phenomenon was not so obvious whenever rats were fed a nonprotein diet with any flavoring (Fig. 5B). Experiment 5

Rats fed a diet containing 20% amino acid mixture, equivalent to PEP, preferred saccharin, as well as MSG, and grew

420

MORI, KAWADA

Taste

Maternal

I

Added

normally. But animals supplied with L-tryptophan-deficient diet preferred tryptophan, and then MSG intake began to elevate, along with their growth, to become normal (Fig. 6). However, preference for saccharin never occurred during this period. Once they were allowed to ingest the L-tryptophan-sufficient diet in replacement of the deficient one, high preference for saccharin occurred (Fig. 6). They still sustained the ingestion of the MSG solution, but stopped preference for tryptophan. This phenomenon was also observed in the case of the reversed sequence of the experimental diet offered to rats.

I 5 6%MSG

0.2%MSG+O

5%GMP

20% sucrose I .8%

6.3%

N&l

Cltrlc

AND TORI1

Acid DISCUSSION

None I

I

I

(Control)

-100

1

0

NON

PROTEIN

0

lOO(g’

DIET

100 15% CASEIN

OFFERED

DIET

200(s) OFFERED

FIG. 4. The comparison of body weight gain for three weeks in rats a nonprotein or 15% casein diet containing taste materials. Data were expressed as mean values with one standard deviation. The statistical significance between control and experiment groups noted in this figure as follows *p
A

19

Dark

Ltghf 7

19

7

Animals in the wild are often compelled to adapt to a state of prolonged starvation. They are able to do this by catabolizing nutrients stored in their bodies and by enhancing the efficiency nutrient retention. Also, animals are able to monitor dietary nutrients during ingestion, digestion, absorption, and metabolism and, thus, function so as to maintain extracellular fluid levels of each nutrient within normal limits and to correct certain nutritional imbalances when they sense some deficiency (8,12). Although experimental animals such as rodents are domesticated and are usually free from starvation (as are humans, generally), they retain this capacity to adapt under the conditions of

FREQUENCY (2 weeks)

FREQUENCY (2 weeksi 5.6’0

MSG

0.2Oo Quinine

32 N-6

J.yl(lL4h1,.

I

O.ZOoMSG and 0.5% GMP

6.3Oo Cltrx ,-

N=6

32F

No Taste

. 0 I

N-

a _. 2

MEAL

3

0

Dark 19

1

PATTERN,

L&t

,9

Matenal

Added

‘$ ?4J4L.; : . /

4

FREQUENCY (2 weeks)

1

2

MEAL

lomin

3

4

SIZE.

9

FREQUENCY

1

_ (2 weeks) 321‘ 0.2%

5.6% MSG

u++.+

5

SIZE, g FEEDING

Acid

Quinine

_,.F.

1

0.2% MSG and 0.5% GMP

6.3”& Citric

No Taste

Acid

Material

Added

*,...,r.= 0 FEEDING

PATTERN,

1Dmin

2 3 4 1 MEAL’ SIZE. g

A” 0 FEEDING

PATTERN,

MEAL

lomin

FIG. 5. The diurnal feeding pattern of 15% casein (A) or nonprotein (B) diet containing with a taste material. B. The histogram of meal size was summarized data for two weeks from each group using a microcomputer.

12

Each

taste

material

was

_s*..

34 SIZE.

noted

9

in A and

APPETITE, TASTE PREFERENCE AND PROTEIN NUTRITION

20%

J 5 3

Amino

Acid

Diet

(ml/rat/day)

40

Z

l

Trvptophan

A

Saccharin

0

MSG

(Bitter) (Sweet)

(Urnam!

taste)

Y -t 9

I

-7

1

1

0

DAY AFTER

TASTE

I

,

20

40

SOLUTION

OFFERED

I

60

TO RATS

FIG. 6. The preference in rats fed a diet with L-tryptophan (Trp) deficiency when taste solutions were available in choice.

nutritional deprivation that have been described (11). Moreover, results of studies on taste preferences and appetites for specific nutrients in which the animal is deficient may be extrapolated to humans as well. In previous reports, endogenous protein was spared when energy intake was sufficient for the spontaneous motor activity as well as for the maintenance of body temperature, whereas degradation of endogenous protein was evoked under conditions of starvation. (18). The cephalic stimuli of sweetness (i.e., from glucose or saccharin) causes insulin to be secreted into circulating blood, but the plasma insulin that is released by saccharin declines rapidly to basal levels because blood glucose remains unchanged (3,15). This study clearly indicates that at some point animals are able to discriminate which of various sweet-tasting substances are an energy source or not. When the animals were forced to restrict their diet supply for one week, they displayed a high preference for sweetness, especially for glucose, not saccharin (Fig. 1). It is apparent that the animals’ primary concern was energy intake. Their strong preference for glucose solutions was coupled with a negative energy balance. Absolute glucose intake was strictly regulated to accord with the levels physiologically required. Thus, in fasted rats, endogenous protein degradation was suppressed by the ingestion of a level of glucose that was sufficient to meet energy needs (Fig. 2). In contrast, because saccharin cannot be utilized as an energy source in the living body, the preference for saccharin actually disappeared (Figs. 1 and 2). Glucogenic L-amino acids such as L&urine, L-serine, and glycine are also used to spare endogenous protein under conditions of starvation (18). The data shown in Fig. 3 support the concept that the degree of decreased diet intake was paralleled with sugar ingestion, except in the case of sucrose. Sucrose, which is a disaccharide and composed of a glucose and fructose moiety, may not yield enough nutritional information as an energy source and affect the control of total energy intake in rats. It was previously reported (9) that the energy expenditure in humans who consumed sucrose or both glucose and fructose was twice as high as that in those who consumed monosaccharide, e.g., glucose, maltose or galactose. This fact suggested that the recognized energy intake of sucrose in animals should be like

421

monosaccharide (9). This exceeded energy intake beyond required level could be expectedly wasted as energy expenditure. But, unfortunately, some portion might be. accumulated in the adipose tissue as a body fat and attributed to fall in the obesity. It is vitally important that the taste solutions be at the most preferable level of concentration when taste preference in experimental animals is being measured (17). The most preferred and the mean concentration of each L-amino acid were reported previously (17). These data indicate that (a) those L-amino acids, to be palatable in humans due to their sweetness and/or umami taste character, were also found to be preferred by rats; likewise those imparting a bitter and/or sour quality were aversive; and (b) in the rat, gustatory perceptions (including umami taste) of each these L-amino acid solutions were not very different from those found in humans (2, 10, 13). A number of other investigators have also reported that taste perceptions of L-amino acids in rats and in humans are generally quite similar to each other (6,19). Mammals, including humans, can easily detect the amount and quality of macronutrients ingested during a meal, such as protein, energy and mineral sources, and use this information via cephalic relays to initiate digestion in the alimentary tract (17,18). They must ultimately have a means to determine whether the amount of dietary nutrients consumed are sufficient for their body needs. When the animal realizes that protein intake is deficient or imbalanced, the mechanisms of protein sparing, as well as those that are needed to motivate a search for the deficient materials, should occur as predicted by the homeostatic model. When a nonprotein diet with either palatable or aversive flavorings was offered to rats, the feeding pattern and meal size was quite comparable in each group (Fig. 5B). In contrast, the larger meal size was recorded in the case of the addition of palatable taste materials to 15% casein diet, but both daily diet intake and growth remained unchanged (Figs. 4 and 5A). Nonetheless, the addition of quinine or citric acid strongly suppressed the growth, reflecting the anorexia for a diet with aversive flavoring (Fig. 4). These facts indicated that the flavored diet had sufficient dietary protein to fit the physiologically required need. In addition, rats fed a diet containing a mixture of crystalline L-amino acids with the same composition as PEP (17), preferred saccharin and MSG and grew normally (Fig. 6). But when animals were supplied with L-tryptophan-deficient diet, the preference for tryptophan was elicited, and then MSG intake was elevated so that their growth could become normal (Fig. 6). However, preference for saccharin never occurred, because tryptophan levels in the blood fluctuated greatly and were hardly maintained within the normal narrow rage. Once a L-tryptophan-sufficient diet was offered to rats, remarkable saccharin ingestion occurred, suggesting that ingestion of a nonnutritive sweet compound, saccharin, concurrently with water beyond physiological needs, should be allowed under normal status of protein nutrition. These results suggest that sweettasting stimuli can serve for markers as energy sources during a meal, and are highly preferred by any species of animal that needs to satisfy a demand for energy. Thus, the strong preference for sweetness that is evoked by starvation, which we might call the “appetite” for energy, is directly regulated by the negative energy balance. It is also apparent that the animal’s primary concern was energy intake, while its second concern was for the state of protein nutrition; strong preference for sugars and glucogenie L-amino acid were coupled with the homeostasis among body temperature, metabolism or motor activity, and the disappearance of preference for noncaloric sweetener, e.g., saccharin, under the disorder of protein and/or energy intake, was quite reasonable.

MORI, KAWADA AND TORI1

422

REFERENCES 1. Association of Official Analytical Chemists. Horwitz, W., ed. Official method of analysis, 13th ed. Washington, DC: AOAC; 1980. 2. Beauchamp, G. K.; Maller, 0. The development of flavor preferences in humans. In: Kare, M. R.; Maller, O., eds. The chemical senses and nutrition. New York: Academic Press; 1977:291-310. 3. Berthoud, H. R.; Trimbel, E. R.; Siegel, E. G.; Bereiter, D. A.; Jeanrenaud, B. Cephalic-phase insulin secretion in normal and pancreatic islet-transplanted rats. Am. J. Physiol. 238:E336-E340; 1980. 4. Cagan, R. H.; Torii, K.; Kare, M. R. Biochemical studies of glutamate taste receptors: The synergistic taste effect of L-glutamate and 5’ribonucleotides. In: Filer, L. J., Jr.; Garattini, S. J.; Kare, M. R.; Reynolds, W. A.; Wurtman, R. J., eds. Glutamic acid: Advances in biochemistry and physiology. New York: Raven Press; 1979:1-9. 5. Fermandez-Flares, E. Kline, D. A.; Johnson, A. R.; Leber, B. L. Qualitative GLC analysis of free amino acids in fruits and juices. J. Am. Oil. Chem. Sot. 536:1203-1208; 1970. 6. Kirimura, J.; Shimizu, A.; Kimizuka, A.; Ninomiya, T.; Kastuya, N. The contribution of peptides and amino acids to the taste of foodstuffs. J. Agric. Food Chem. 17:689-695; 1969. 7. Kuninaka, A.; Kibi, M. Sakaguchi, K. History and development of flavor nucleotides. Food Technol. 18:287-293; 1964. 8. Leung, P. M. B.; Rogers, Q. R.; Harper, A. E. Effect of amino acid imbalance on dietary choice in the rats. J. Nun. 95:4831192; 1968. 9. Macdonald, I. Differences in dietary-induced thermogenesis following the ingestion of various carbohydrates. Ann. Nutr. Metab. 28: 226-230; 1984. IO. Pfaffmann, C. Physiological and behavioral processes of the sense of taste. In: Wolstenholme, G. E. W.; Knight, J.. eds. Taste and

smell in vertebrates.

London: Churchill; 1970:31-50. produced behavior reactions to food in wild domestic rats. Ann. NY Acad. Sci. 56:225-239;

Il. Richter, C. P. Experimentally 12.

13.

14. 15. 16.

17.

18.

19.

poisoning 1953. Rogers, Q. R.; Leung, P. M. B. The control of food intake: When and how are amino acids involved? In: Kare, M. R.; Mailer, 0.. eds. The chemical sense and nutrition. New York: Academic Press;: 1977:213-248. Rozin, P.; Kalat, J. W. Specific hungers and poison avoidance as adaptive specializations of learning. Psychol. Rev. 78:459485: 1971. Schiffman, S. S.; Engelhard, H. H., III. Taste of dipeptides. Physiol. Behav. 17:523-535; 1976. Steffens, A. B. Influence of the oral cavity on insulin release in the rat. Am. J. Physiol. 230:1411-1415; 1976. Torii, K.; Mimura, T.; Yugari, Y. Preference for umami, sweet and salty taste in rats fed diets containing various amounts and quality of protein. Proc. Jpn. Symp. Taste Smell 17th 89-92; 1983. Torii, K.; Mimura, T.; Yugari. Y. Effects of dietary protein on the taste preference for amino acids in rats. In: Kare, M. R.; Brand, J. G., eds. Interaction of the chemical senses with nutrition. New York: Academic Press; 1986:45%69. Torii, K.; Mimura, T.; Yugari, Y. Biochemical mechanism of umami taste perception and effect of dietary protein on the taste preference for amino acids and sodium chloride in rats. In: Kawamura. Y.; Kare, M. R., eds. Umami: A basic taste. New York: Marcel Dekker; 1987:513-563. Yoshida, M.; Saito, S. Multidimensional scaling of the taste of amino acids. J. Psychol. Res. 11:149-166; 1969.