Physiology&Behavior,Vol. 52, pp. 541-546, 1992
0031-9384/92 $5.00 + .00 Copyright © 1992 Pergamon Press Ltd.
Printed in the USA.
Consumption of Salty Food by Rats: Regulation of Sodium Intake? GEORGE
COLLIER 1 AND
DEANNE
F. J O H N S O N
Department of Psychology, Rutgers University, New Brunswick, N J 08903 R e c e i v e d 25 O c t o b e r 1991 COLLIER, G. AND D. F. JOHNSON. Consumption of saltyfood by rats:Regulation of sodium intake?PHYSIOLBEHAV 52(3) 541-546, 1992.--The extent to which sodium levels may be regulated by consumption was examined in two experiments that offered rats foods varying in sodium chloride (NaC1) content. In the first, rats received single purified diets containing from 0% to 3% NaCI. There were no effects of NaC1 level on the amount or pattern of daily food intake; water intake, however, increased with salt content. In the second study, rats had choices between a NaCl-free food and a food containing either 1, 2, or 3% NaC1 for 1 week each. Total food intake was unaffected. Proportional intake of the salt-free option increased with the salt content of the alternate food, but not sufficiently to maintain a constant NaC1 intake. After 8 weeks of exposure to a single food, intake of the salty option increased in the choice tests, but the level of NaC1 (from 0.5 to 3.0%) in the exposure-phase food did not affect the subsequent choice. We conclude that when only one food is available, salt intake is governed by caloric requirements and sodium levels are regulated by excretion. When foods differing in NaCI Content are available, consumption does contribute to the regulation of sodium balance, but the amount consumed is not tightly controlled. Rats' salt preference appears to increase with age or with experience eating the purified foods offered here, but experience eating salty food does not affect the preferred level of salt. Salt
Food intake
Meal patterns
Sodium regulation
Salt preference
Rats
There are a number of reasons why the intake of salt mixed into food (one way in which animals c o m m o n l y encounter salt) may differ from the intake of saline solutions. In mixed food, the a m o u n t consumed will be controlled, at least in part, by the caloric and/or macronutrient content of the food in addition to, or rather than, the salt content. Also, the tastes of the other nutrients may mask or interact with the taste of salt. Thus, in food acceptance studies, the a m o u n t of salt consumed is likely to be primarily a function of the a m o u n t of food consumed. When several foods differing in salt content are available, an animal can adjust the a m o u n t of salt consumed without altering food intake by changing the relative intake of each food. For example, Bertino and Tordoff (4) allowed rats simultaneous access to unsalted and salted food and found that the intake of each food depended on the concentration of salt in the salty option. The extent to which these adjustments are made by sodium-replete animals is not known. Another question concerns the factors that can alter the level of salt consumed by the sodium-replete animal. Rats that have experienced an experimental sodium depletion consume more of a sodium solution in a subsequent need-free acceptance test than do rats with no such experience (7); however, a similar persistent preference for salted food was not induced in rats by sodium depletion (4). A different factor, recent dietary experi-
R E G U L A T I O N of electrolyte balance occurs by a complex interaction of consumption, conservation, and excretion (6,12). Sodium levels, for example, are tightly controlled by physiological mechanisms that conserve sodium in times of deficiency and that excrete sodium in times of excess. On the other hand, the extent to which consumption of sodium is regulated is unclear. Although salt consumption is clearly stimulated by sodium deficiency (6), consumption is not necessarily inhibited by sodium repletion (2,6). It may be that the evolution of efficient mechanisms to excrete sodium eliminated the need for strict limits on overconsumption. Most laboratory studies of salt intake have shed little light on the determinants of normal intake. Rather, thirsty or sodiumdeplete animals are offered saline solutions in brief tests of either acceptance (one solution available) or preference (more than one solution available). Water and chow are offered following the session. Rats typically prefer saltier solutions with m a x i m u m volume intake at isotonicity (9,12,14). Only a few laboratory studies have investigated ad lib intake of salted food. The results for rats in these tests differed from those in tests using salt solutions in that salted food was not preferred (3). One study suggests that the intake of sodium in food is quite independent of the intake of sodium in solution, and that only the latter is regulated (8).
Requests for reprints should be addressed to George Collier, Department of Psychology, Busch Campus, Rutgers University, New Brunswick, NJ 08903.
541
542
COLLIER AND JOHNSON
ence, may alter humans' preference for salt such that, for example, an enforced low level of salt intake can decrease the preferred amount of salt in food (2). Animals' salt appetite may not be altered in the same way by dietary experience. Rats offered a salted-food o n i o n in addition to rat chow did not increase consumption of the option over time (3); and baboons fed a high-salt diet did not prefer a higher level of salt than those fed a lower-salt diet (1). The present experiments address three questions: first, will the concentration of salt in food affect the meal patterns and total intake of rats offered ad lib access to a single diet? Although there is evidence (8) that total food intake is unaffected by NaC1 levels up to 6%, meal patterns may change to reflect any palatability differences introduced by different salt levels. Second, will rats eat a constant amount of salt when offered ad lib access to pairs of foods differing in salt content? And, third, what effect will extended experience with salty food have on the diet composed?
TABLE 1 THE COMPOSITION OF THE DIETS (g]kg) Diet Ingredient
0.0%
0.5%
1,0%
2,0%
3.0%
Casein DL-Methionine Cornstarch Sucrose Corn oil NaCl-free mineral mix* Vitamin mix, ICN 76 Choline bitartrate Celluflour NaCI
200 3 150 500 50 35 10 2 50 0
200 3 150 500 50 35 10 2 45 5
200 3 150 500 50 35 10 2 40 10
200 3 150 500 50 35 10 2 30 20
200 3 150 500 50 35 10 2 20 30
* ICN 76, NaCI replaced with sucrose.
EXPERIMENT I METHOD
Subjects and Apparatus Four male Sprague-Dawley rats (Camm Research Inst., Wayne, N J) were individually housed in double-sized cages (41 × 23 × 19 cm; Hoeltge, Cincinnati, OH) equipped with a feeder tunnel giving access to a retractable food Cup, a T-shaped response bar mounted 12 cm above the floor and 7.5 cm from the feeder tunnel and requiring 0.35 N to depress, and a graduated water bottle. A photocell monitored the presence of the rat in the feeder tunnel. Access to food was contingent upon completion of 10 bar presses, which caused the food cup to move up to the feeder tunnel. A meal of any size could be consumed; after 10 consecutive minutes elapsed without feeder entry, the food cup was withdrawn. Another meal could be initiated at any time. The operation of the equipment was controlled and the rats' responses were recorded by microprocessors (Commodore, Pet 4032). Collier (5) discusses this paradigm in detail. Two additional control rats were individually housed in singlesized cages (24 × 20 × 20 cm; Hoeltge) equipped with a feeder tunnel giving access to a food jar and a graduated water bottle.
Diets The experimental foods were composed of casein, sucrose, vegetable shortening, cellulose, vitamins, and minerals, and they were nutritionally identical except that the amounts of NaC1 and cellulose were adjusted to provide either 0, 0.5, 1.0, 2.0, or 3.0% NaC1 by weight (Table 1). The foods will be referred to by their NaC1 concentration.
Procedure All rats were housed in the same room at 23-25°C with the lights on from 0800 to 2000 h. The room was entered at 1030 h each day for a maintenance period of about 30 min during which the rats were weighed, food and water intakes were measured, food and water were replenished, and the equipment was tested. All rats were fed Purina rat chow for a 2-week period during which the experimental rats were adapted to their cages and trained to use the apparatus. Meal patterns had stabilized by the end of this period. Each rat then received purified diets containing either 0, 0.5, 1,0, or 2.0% NaC1 for 10 days each in a random order. Because little effect on food intake was seen with this
range of dietary NaC1, in a replication the rats were fed these four foods, plus the 3% NaC1 food in a random order, for 10 days each. One month elapsed between the first and second replication; during this time the rats were fed Purina chow.
Data Analysis Mean daily total food intake, meal size (g/meal), meal frequency (meals/day), and water intake (ml water/g of food) were determined for each rat for each diet condition. Because there were no differences between the first and second exposure to the 0.0, 0.5, 1.0, and 2.0% diets, the data from both exposures were combined in the analysis presented here. Differences among diets were evaluated with repeated measures ANOVAs with an alpha level of 0.05. RESULTS AND DISCUSSION All rats grew normally, with body weight increasing from 486 + 10 to 571 ± 11 g over approximately 4 months. Growth rates did not differ between the experimental and control rats nor between the periods when the rats ate Purina chow and the experimental foods. There were no statistically significant effects of dietary salt content on daily caloric intake or the pattern of meals (Fig. 1). A slight decline in average total food intake as a function of increasing salt concentration was not significant, F(3, 4) = 0.82, p > 0.05. There was no change in meal size, F(3, 4) = 1.69, p > 0.05, or meal frequency, F(3, 4) = 1.23, p > 0.05, with dietary salt. In contrast, water intake did increase significantly with dietary salt content, F(3, 4) = 22.22, p < 0.002. It, thus, appears that food intake in this no-choice situation is determined by the caloric requirement, and, if excess salt is consumed, then water intake increases to allow the elimination of the excess. If gastric NaC1 loads limited intake within a single meal or if the salty taste were aversive, then one might have expected meals of saltier foods to be smaller [cf. (10)]. This effect was not seen. However, a larger baseline meal size, which could be produced by imposing a higher procurement cost, and/or a wider range of salt levels, need to be examined before dismissing the possibility that meal size may be influenced by dietary NaC1. EXPERIMENT II This study examined, first, the diet composed by rats given concurrent access to two foods differing in NaC1 content, and,
SALT CONSUMPTION BY RATS
,-'--
25 24
"~
22 21
543 second, whether prolonged intake of salty food would alter the composition of the diet. METHOD
Subjects, Apparatus and Diets Thirty-three naive male Sprague-Dawley rats (Carom Research Inst., Wayne, N J), 90 days old at the start of the experiment, were housed individually in double cages (41 × 23 N 19 cm, Hoeltge, Cincinnati, OH) fitted at the front with two feeder tunnels, each providing access to ajar containing approximately 60 g of food. The foods were identical to those in Experiment I. Tap water was available from an inverted graduated cylinder attached to the cage between the feeders.
20 O
~, ~'~ 121
a0 18-
17. 16. 15
Procedure
10"
~' "0
9" 8" 7" 6"
u~
5-
0
4" 3"
Throughout the study, the rats were undisturbed except for a daily maintenance period when each animal was weighed, food and water intakes were recorded, and the food(s) and water were replenished. When two foods were offered, one was placed in each food cup and the position of the cups was alternated daily. When one food was offered, only one cup contained food and its location was alternated daily. Adaptation phase. For 2 weeks all rats were fed the 0.5% food. Two groups then were formed, a naive group (n = 8) and an exposure group (n = 25), matched for body weight. First choice test. Each of the Exposure rats was offered three pairs of foods for 1 week each. The pairs were the 0 and 1.0%, 0 and 2.0%, and 0 and 3.0% foods, and the order of pair presentation was a modified Latin square. The naive group was fed the 0.5% food during this time. Salt exposure phase. The exposure group of 25 was subdivided into three groups matched for body weight and average NaC1 intake during the first choice test. One group (n = 8) was fed the 0.5% food, another (n = 8) the 2.0% food, and the third (n = 9) the 3.0% food, for 8 weeks. The naive group continued to eat the 0.5% food. One of the naive rats died of a virus during this time. Second choice test. All rats (exposure and naive) were offered the three pairs of diets, 0 and 1.0%~0 and 2.0%, and 0 and 3.0%, for 1 week each, as in the first choice test.
2-
m
o
E E O
RESULTS
First Choice Test ~'"0
0 ~0
E 0
3.0
We compared the pattern of intake in each of the three dietpair conditions using one-factor ANOVA with repeated measures. Total food intake was unaffected by the foods available; the rats ate about 20 g per day and grew normally. The 0% food was preferred to the alternate food in all three pairs and made up an increasing proportion of the diet: 0.66, 0.81, and 0.85, as the NaC1 concentration of the alternate food increased from 1 to 2 to 3%, respectively, F(2, 48) = 49.95, p < 0.001. Despite this change, the NaC1 consumed by the rats increased significantly over the three pairings from 0.34% to 0.37% to 0.44% of the diet, F(2, 48) = 4.40, p < 0.05. There was a small, nonsignificant increase in mean daily water consumption as NaC1 intake increased.
2.5 2.0 1.5 Z3; 1.0
$
*.,
0.5 0.0
,
,
0%
0.5%
;
1. %
,
2.0%
3. %
Dietary Salt Content FIG. 1. Mean (+SE) food intake, meal frequency, meal size and water intake of four rats eating purified foods containing different levels of NaC1 in Experiment I.
Exposure Phase Effects of the food consumed during the exposure phase were analyzed with one-factor ANOVA. The NaC1 concentration of the food available did not alter daily caloric intake. The rats ate about 20 g per day and grew normally. Water intake was sig-
544
COLLIER AND JOHNSON
nificantly greater for the rats eating the 3.0% food (41.6 ml/day) than the rats in the other groups, F(2, 31) = 21.5, p < 0.001, but there was no difference in water intake between the 0.5% and the 2.0% groups (25.0 ml/day and 26.2 ml/day, respectively).
Second Choice Test
"0 0 0
I,
0
,8=
E 0
The data for the second choice test were first analyzed by two-factor (four groups X three food pairs) ANOVA with repeated measures on one factor. The intake patterns of the four groups (three salt exposure groups plus the naive group) were not significantly different. That is, choices between the foods were similar regardless of experience with salty food; and the naive rats, that had never been exposed to food higher in NaC1 than 0.5%, patterned their intake between the food pairs no differently than the other rats. The effects of food pair were similar to those in the first choice test. Total daily food intake and body weight were unaffected by the pair of foods presented. The proportion of intake from the 0% food (Fig. 2) increased as the NaC1 concentration of the alternate food increased, F(2, 56) = 80.71, p < 0.001; only for the naive group did the dietary NaC1 (Fig. 2) increase with the NaC1 concentration in the alternate food, F(2, 56) = 4.01, p < 0.05. The intake of the rats that participated in both choice tests was examined with a three-factor (three groups X three food pairs X two tests) ANOVA with repeated measures on the two latter factors. Total daily intake was not different between the tests. For the amount of NaC1 consumed, however, there was a significant main effect of test (lst vs. 2nd), F(I, 22) = 4.95, p < 0.05, and a significant group X food pair interaction, F(4, 44) = 2.65, p < 0.05. Post hoc comparisons showed that rats in all three groups consumed more salt during the second than the first test for the 0 vs. 1% and 0 vs. 2% food pairs. Only the 0.5% exposure rats ate more NaCI during the second test for the 0 vs. 3% pair (Fig. 3). GENERAL DISCUSSION The tendency seen in Experiment I for food intake to decline with salt concentration was not seen in Experiment II. Daily food (calorie) intake was constant across all conditions; and even during the salt exposure phase, when some animals ale food containing 3% NaC1, body weight was not different among the groups. These results are consistent with those of Fregly et al. (8) who found that up to 6% NaCI in food failed to alter daily food intake. The increased water intake by rats consuming highsalt diets suggests that in a no-choice situation, sodium balance is maintained by excretion. It is clear that when otherwise equivalent foods vary in salt content, intake defends calories. The fact that the rats do not reduce their caloric intake as NaC1 content increased indicates that the rats could have eaten randomly from the available foods during the choice tests without deleterious effects. However, they did not. Although the amount of NaC1 consumed in the choice tests increased across tests with increasing NaC1 concentration &the alternate food, the increases were small and much less than the changes in average NaC1 concentration in the two available foods. Sodium balance was maintained in part by consumption. These data support those of the control rats in Bertino and Tordoff (4) that, over a 4-day period, appeared to eat less of a salty food option and more of the unsalted food as the concentration of salt in the salty food increased. When there is a choice between otherwise equivalent foods containing different amounts of salt, intake contributes to sodium balance.
°7=
.o_ 0 .5 o
a. I
.9o
E cO 0 03 nO e--
(,3 O~ Z
.8-
•
!
I
Naive (0.5 %)
o 0.5% [] 2.0%
T ]
,7" ,6" ,5" ,4" ,3" .2
1%
2%
3%
NaCI Content of Alternate Food FIG. 2. Proportion of total intake from the NaCl-free food (top panel) and resulting dietary NaCI content (bottom panel) as a function of the NaC1 content of the alternativelyavailablefood during the second choice test in Experiment II. Means (_+SE)are shown for four groups of rats, each of which had 8-weeksexposure to a diet containing a fixed amount of NaCI. All except the naive group had participated in the first choice test.
The mechanism of this diet selection cannot be determined from our data. It may be the result of a postingestive regulatory process for sodium or of palatability differences among the foods. The former explanation seems unlikelY in light of data indicating that in rats with a choice of water, saline solution, and salted food, intake of neither food nor the saline solution was affected by the level (0-6%) of NaC1 in the food, although intake of the saline solution was a function of its concentration (8). A postingestive mechanism should be insensitive to the source of sodium. The palatability hypothesis is difficult to test because the dominance of caloric regulation precludes a determination of the acceptance function over the range of salt concentrations in our foods. Experience eating a high-salt diet for 8 weeks had no effect on the level of NaC1 consumed by these rats, a finding which differs from that for humans (2) but replicates that for baboons (1). We were surprised that rats in all groups ate more salt in the second than the first choice test, indicating that age may affect the preferred level of salt intake in rats. Alternatively, extended consumption of the purified diets offered here may alter sodium balance in some unknown manner.
0
2
~ o
;~.~
~z
0-.~, pm-.-n
e. ~)
I
,-I
~ ~
~
l
|
~
l
u
"~
~
n
u
n
m
u
a~ e~ o o e,~
~
| ,I
Ne
~t
|
,
I ~
~.~
0
~
•-
. . . . . . ,,.~. ~.
,:
.
.
.
©.
.
.
~
.
~
m
.~ .~
~ , . . . ~ . ~ .
m~
peuJnsuoo ~,e!p u! IOeN %
0'~
~
•
pool % 0
IllOJt
uo!lJodoJd
545
~
546
COLLIER AND JOHNSON ACKNOWLEDGEMENTS
This research was supported by grant DK31016 from the National Institutes of Health and by a grant from Campbell's Soup Company.
Portions of the data from Experiment I were reported to the Eastern Psychological Association in 1986 by Karen CyBulski, whom we thank for her participation. Jose Naveira provided excellent technical assistance in Experiment II.
REFERENCES 1. Barnwell, G. M.; Dollahite, J.; Mitchell, D. S. Salt taste preference in baboons. Physiol. Behav. 37:279-284; 1986. 2. Beauchamp, G. K. The human preference for excess salt. Am. Sci. 75:27-33; 1987. 3. Beauchamp, G. K.; Bertino, M. Rats (Rattus norvegicus) do not prefer salted solid food. J. Comp. Psychol. 99:240-247; 1985. 4. Bertino, M.; Tordoff, M. G. Sodium depletion increases rats' preferences for salted food. Behav. Neurosci. 102:65-73; 1988. 5. Collier, G. H. Operant methodologies for studying feeding and drinking. In: Toates, F. M.; Rowland, N. E., eds. Techniques in the behavioral and neural sciences, vol. 1, Feeding and drinking. Amsterdam: Elsevier; 1987:37-76. 6. Denton, D. The hunger for salt. Berlin: Springer-Verlag; 1982. 7. Epstein, A.; Sakai, R. Angiotensin-aldosterone synergy and salt intake. In: Buckley, J. P.; Ferrario, C. M., eds. Brain peptides and catecholamines in cardiovascular regulation. New York: Raven; 1987:337-345.
8. Fregly, M. J.; Harper, J. M.; Radford, E. P., Jr. Regulation of sodium chloride intake by rats. Am. J. Physiol. 209:287-292; 1965. 9. Forman, S.; Falk, J. L. NaC1 solution ingestion in genetic (SHR) and aortic-ligation hypertension. Physiol. Behav. 22:371-377; 1979. I0. Levitsky, D. Feeding patterns in rats in response to fasts and changes in environmental conditions. Physiol. Behav. 5:291-300; 1970. 11. Richter, C. P. Total self regulatory functions in animals and human beings. Harvey Lect. 38:63-103; 1942-1943. 12. Stellar, E.; Hyman, R.; Samet, S. Gastric factors controlling water- and salt-solution-drinking. J. Comp. Physiol. Psychol. 47:220-226; 1954. 13. Stricker, E. M.; Verbalis, J. G. Hormones and behavior: The biology of thirst and sodium appetite. Am. Sci. 76:261-267; 1988. 14. Weiner, I. H.; Stellar, E. Salt preference of the rat determined by a single-stimulus method. J. Comp. Physiol. Psychol. 44:391-401; 1951. 15. Wolf, G.; Shulkin, J.; Simson, P. E. Multiple factors in the satiation of salt appetite. Behav. Neurosci. 98:651-673; 1984.
0031-9384/92 $5.00 + .00 Copyright © 1992 Pergamon Press Ltd.
Printed in the USA.
Consumption of Salty Food by Rats: Regulation of Sodium Intake? GEORGE
COLLIER 1 AND
DEANNE
F. J O H N S O N
Department of Psychology, Rutgers University, New Brunswick, N J 08903 R e c e i v e d 25 O c t o b e r 1991 COLLIER, G. AND D. F. JOHNSON. Consumption of saltyfood by rats:Regulation of sodium intake?PHYSIOLBEHAV 52(3) 541-546, 1992.--The extent to which sodium levels may be regulated by consumption was examined in two experiments that offered rats foods varying in sodium chloride (NaC1) content. In the first, rats received single purified diets containing from 0% to 3% NaCI. There were no effects of NaC1 level on the amount or pattern of daily food intake; water intake, however, increased with salt content. In the second study, rats had choices between a NaCl-free food and a food containing either 1, 2, or 3% NaC1 for 1 week each. Total food intake was unaffected. Proportional intake of the salt-free option increased with the salt content of the alternate food, but not sufficiently to maintain a constant NaC1 intake. After 8 weeks of exposure to a single food, intake of the salty option increased in the choice tests, but the level of NaC1 (from 0.5 to 3.0%) in the exposure-phase food did not affect the subsequent choice. We conclude that when only one food is available, salt intake is governed by caloric requirements and sodium levels are regulated by excretion. When foods differing in NaCI Content are available, consumption does contribute to the regulation of sodium balance, but the amount consumed is not tightly controlled. Rats' salt preference appears to increase with age or with experience eating the purified foods offered here, but experience eating salty food does not affect the preferred level of salt. Salt
Food intake
Meal patterns
Sodium regulation
Salt preference
Rats
There are a number of reasons why the intake of salt mixed into food (one way in which animals c o m m o n l y encounter salt) may differ from the intake of saline solutions. In mixed food, the a m o u n t consumed will be controlled, at least in part, by the caloric and/or macronutrient content of the food in addition to, or rather than, the salt content. Also, the tastes of the other nutrients may mask or interact with the taste of salt. Thus, in food acceptance studies, the a m o u n t of salt consumed is likely to be primarily a function of the a m o u n t of food consumed. When several foods differing in salt content are available, an animal can adjust the a m o u n t of salt consumed without altering food intake by changing the relative intake of each food. For example, Bertino and Tordoff (4) allowed rats simultaneous access to unsalted and salted food and found that the intake of each food depended on the concentration of salt in the salty option. The extent to which these adjustments are made by sodium-replete animals is not known. Another question concerns the factors that can alter the level of salt consumed by the sodium-replete animal. Rats that have experienced an experimental sodium depletion consume more of a sodium solution in a subsequent need-free acceptance test than do rats with no such experience (7); however, a similar persistent preference for salted food was not induced in rats by sodium depletion (4). A different factor, recent dietary experi-
R E G U L A T I O N of electrolyte balance occurs by a complex interaction of consumption, conservation, and excretion (6,12). Sodium levels, for example, are tightly controlled by physiological mechanisms that conserve sodium in times of deficiency and that excrete sodium in times of excess. On the other hand, the extent to which consumption of sodium is regulated is unclear. Although salt consumption is clearly stimulated by sodium deficiency (6), consumption is not necessarily inhibited by sodium repletion (2,6). It may be that the evolution of efficient mechanisms to excrete sodium eliminated the need for strict limits on overconsumption. Most laboratory studies of salt intake have shed little light on the determinants of normal intake. Rather, thirsty or sodiumdeplete animals are offered saline solutions in brief tests of either acceptance (one solution available) or preference (more than one solution available). Water and chow are offered following the session. Rats typically prefer saltier solutions with m a x i m u m volume intake at isotonicity (9,12,14). Only a few laboratory studies have investigated ad lib intake of salted food. The results for rats in these tests differed from those in tests using salt solutions in that salted food was not preferred (3). One study suggests that the intake of sodium in food is quite independent of the intake of sodium in solution, and that only the latter is regulated (8).
Requests for reprints should be addressed to George Collier, Department of Psychology, Busch Campus, Rutgers University, New Brunswick, NJ 08903.
541
542
COLLIER AND JOHNSON
ence, may alter humans' preference for salt such that, for example, an enforced low level of salt intake can decrease the preferred amount of salt in food (2). Animals' salt appetite may not be altered in the same way by dietary experience. Rats offered a salted-food o n i o n in addition to rat chow did not increase consumption of the option over time (3); and baboons fed a high-salt diet did not prefer a higher level of salt than those fed a lower-salt diet (1). The present experiments address three questions: first, will the concentration of salt in food affect the meal patterns and total intake of rats offered ad lib access to a single diet? Although there is evidence (8) that total food intake is unaffected by NaC1 levels up to 6%, meal patterns may change to reflect any palatability differences introduced by different salt levels. Second, will rats eat a constant amount of salt when offered ad lib access to pairs of foods differing in salt content? And, third, what effect will extended experience with salty food have on the diet composed?
TABLE 1 THE COMPOSITION OF THE DIETS (g]kg) Diet Ingredient
0.0%
0.5%
1,0%
2,0%
3.0%
Casein DL-Methionine Cornstarch Sucrose Corn oil NaCl-free mineral mix* Vitamin mix, ICN 76 Choline bitartrate Celluflour NaCI
200 3 150 500 50 35 10 2 50 0
200 3 150 500 50 35 10 2 45 5
200 3 150 500 50 35 10 2 40 10
200 3 150 500 50 35 10 2 30 20
200 3 150 500 50 35 10 2 20 30
* ICN 76, NaCI replaced with sucrose.
EXPERIMENT I METHOD
Subjects and Apparatus Four male Sprague-Dawley rats (Camm Research Inst., Wayne, N J) were individually housed in double-sized cages (41 × 23 × 19 cm; Hoeltge, Cincinnati, OH) equipped with a feeder tunnel giving access to a retractable food Cup, a T-shaped response bar mounted 12 cm above the floor and 7.5 cm from the feeder tunnel and requiring 0.35 N to depress, and a graduated water bottle. A photocell monitored the presence of the rat in the feeder tunnel. Access to food was contingent upon completion of 10 bar presses, which caused the food cup to move up to the feeder tunnel. A meal of any size could be consumed; after 10 consecutive minutes elapsed without feeder entry, the food cup was withdrawn. Another meal could be initiated at any time. The operation of the equipment was controlled and the rats' responses were recorded by microprocessors (Commodore, Pet 4032). Collier (5) discusses this paradigm in detail. Two additional control rats were individually housed in singlesized cages (24 × 20 × 20 cm; Hoeltge) equipped with a feeder tunnel giving access to a food jar and a graduated water bottle.
Diets The experimental foods were composed of casein, sucrose, vegetable shortening, cellulose, vitamins, and minerals, and they were nutritionally identical except that the amounts of NaC1 and cellulose were adjusted to provide either 0, 0.5, 1.0, 2.0, or 3.0% NaC1 by weight (Table 1). The foods will be referred to by their NaC1 concentration.
Procedure All rats were housed in the same room at 23-25°C with the lights on from 0800 to 2000 h. The room was entered at 1030 h each day for a maintenance period of about 30 min during which the rats were weighed, food and water intakes were measured, food and water were replenished, and the equipment was tested. All rats were fed Purina rat chow for a 2-week period during which the experimental rats were adapted to their cages and trained to use the apparatus. Meal patterns had stabilized by the end of this period. Each rat then received purified diets containing either 0, 0.5, 1,0, or 2.0% NaC1 for 10 days each in a random order. Because little effect on food intake was seen with this
range of dietary NaC1, in a replication the rats were fed these four foods, plus the 3% NaC1 food in a random order, for 10 days each. One month elapsed between the first and second replication; during this time the rats were fed Purina chow.
Data Analysis Mean daily total food intake, meal size (g/meal), meal frequency (meals/day), and water intake (ml water/g of food) were determined for each rat for each diet condition. Because there were no differences between the first and second exposure to the 0.0, 0.5, 1.0, and 2.0% diets, the data from both exposures were combined in the analysis presented here. Differences among diets were evaluated with repeated measures ANOVAs with an alpha level of 0.05. RESULTS AND DISCUSSION All rats grew normally, with body weight increasing from 486 + 10 to 571 ± 11 g over approximately 4 months. Growth rates did not differ between the experimental and control rats nor between the periods when the rats ate Purina chow and the experimental foods. There were no statistically significant effects of dietary salt content on daily caloric intake or the pattern of meals (Fig. 1). A slight decline in average total food intake as a function of increasing salt concentration was not significant, F(3, 4) = 0.82, p > 0.05. There was no change in meal size, F(3, 4) = 1.69, p > 0.05, or meal frequency, F(3, 4) = 1.23, p > 0.05, with dietary salt. In contrast, water intake did increase significantly with dietary salt content, F(3, 4) = 22.22, p < 0.002. It, thus, appears that food intake in this no-choice situation is determined by the caloric requirement, and, if excess salt is consumed, then water intake increases to allow the elimination of the excess. If gastric NaC1 loads limited intake within a single meal or if the salty taste were aversive, then one might have expected meals of saltier foods to be smaller [cf. (10)]. This effect was not seen. However, a larger baseline meal size, which could be produced by imposing a higher procurement cost, and/or a wider range of salt levels, need to be examined before dismissing the possibility that meal size may be influenced by dietary NaC1. EXPERIMENT II This study examined, first, the diet composed by rats given concurrent access to two foods differing in NaC1 content, and,
SALT CONSUMPTION BY RATS
,-'--
25 24
"~
22 21
543 second, whether prolonged intake of salty food would alter the composition of the diet. METHOD
Subjects, Apparatus and Diets Thirty-three naive male Sprague-Dawley rats (Carom Research Inst., Wayne, N J), 90 days old at the start of the experiment, were housed individually in double cages (41 × 23 N 19 cm, Hoeltge, Cincinnati, OH) fitted at the front with two feeder tunnels, each providing access to ajar containing approximately 60 g of food. The foods were identical to those in Experiment I. Tap water was available from an inverted graduated cylinder attached to the cage between the feeders.
20 O
~, ~'~ 121
a0 18-
17. 16. 15
Procedure
10"
~' "0
9" 8" 7" 6"
u~
5-
0
4" 3"
Throughout the study, the rats were undisturbed except for a daily maintenance period when each animal was weighed, food and water intakes were recorded, and the food(s) and water were replenished. When two foods were offered, one was placed in each food cup and the position of the cups was alternated daily. When one food was offered, only one cup contained food and its location was alternated daily. Adaptation phase. For 2 weeks all rats were fed the 0.5% food. Two groups then were formed, a naive group (n = 8) and an exposure group (n = 25), matched for body weight. First choice test. Each of the Exposure rats was offered three pairs of foods for 1 week each. The pairs were the 0 and 1.0%, 0 and 2.0%, and 0 and 3.0% foods, and the order of pair presentation was a modified Latin square. The naive group was fed the 0.5% food during this time. Salt exposure phase. The exposure group of 25 was subdivided into three groups matched for body weight and average NaC1 intake during the first choice test. One group (n = 8) was fed the 0.5% food, another (n = 8) the 2.0% food, and the third (n = 9) the 3.0% food, for 8 weeks. The naive group continued to eat the 0.5% food. One of the naive rats died of a virus during this time. Second choice test. All rats (exposure and naive) were offered the three pairs of diets, 0 and 1.0%~0 and 2.0%, and 0 and 3.0%, for 1 week each, as in the first choice test.
2-
m
o
E E O
RESULTS
First Choice Test ~'"0
0 ~0
E 0
3.0
We compared the pattern of intake in each of the three dietpair conditions using one-factor ANOVA with repeated measures. Total food intake was unaffected by the foods available; the rats ate about 20 g per day and grew normally. The 0% food was preferred to the alternate food in all three pairs and made up an increasing proportion of the diet: 0.66, 0.81, and 0.85, as the NaC1 concentration of the alternate food increased from 1 to 2 to 3%, respectively, F(2, 48) = 49.95, p < 0.001. Despite this change, the NaC1 consumed by the rats increased significantly over the three pairings from 0.34% to 0.37% to 0.44% of the diet, F(2, 48) = 4.40, p < 0.05. There was a small, nonsignificant increase in mean daily water consumption as NaC1 intake increased.
2.5 2.0 1.5 Z3; 1.0
$
*.,
0.5 0.0
,
,
0%
0.5%
;
1. %
,
2.0%
3. %
Dietary Salt Content FIG. 1. Mean (+SE) food intake, meal frequency, meal size and water intake of four rats eating purified foods containing different levels of NaC1 in Experiment I.
Exposure Phase Effects of the food consumed during the exposure phase were analyzed with one-factor ANOVA. The NaC1 concentration of the food available did not alter daily caloric intake. The rats ate about 20 g per day and grew normally. Water intake was sig-
544
COLLIER AND JOHNSON
nificantly greater for the rats eating the 3.0% food (41.6 ml/day) than the rats in the other groups, F(2, 31) = 21.5, p < 0.001, but there was no difference in water intake between the 0.5% and the 2.0% groups (25.0 ml/day and 26.2 ml/day, respectively).
Second Choice Test
"0 0 0
I,
0
,8=
E 0
The data for the second choice test were first analyzed by two-factor (four groups X three food pairs) ANOVA with repeated measures on one factor. The intake patterns of the four groups (three salt exposure groups plus the naive group) were not significantly different. That is, choices between the foods were similar regardless of experience with salty food; and the naive rats, that had never been exposed to food higher in NaC1 than 0.5%, patterned their intake between the food pairs no differently than the other rats. The effects of food pair were similar to those in the first choice test. Total daily food intake and body weight were unaffected by the pair of foods presented. The proportion of intake from the 0% food (Fig. 2) increased as the NaC1 concentration of the alternate food increased, F(2, 56) = 80.71, p < 0.001; only for the naive group did the dietary NaC1 (Fig. 2) increase with the NaC1 concentration in the alternate food, F(2, 56) = 4.01, p < 0.05. The intake of the rats that participated in both choice tests was examined with a three-factor (three groups X three food pairs X two tests) ANOVA with repeated measures on the two latter factors. Total daily intake was not different between the tests. For the amount of NaC1 consumed, however, there was a significant main effect of test (lst vs. 2nd), F(I, 22) = 4.95, p < 0.05, and a significant group X food pair interaction, F(4, 44) = 2.65, p < 0.05. Post hoc comparisons showed that rats in all three groups consumed more salt during the second than the first test for the 0 vs. 1% and 0 vs. 2% food pairs. Only the 0.5% exposure rats ate more NaCI during the second test for the 0 vs. 3% pair (Fig. 3). GENERAL DISCUSSION The tendency seen in Experiment I for food intake to decline with salt concentration was not seen in Experiment II. Daily food (calorie) intake was constant across all conditions; and even during the salt exposure phase, when some animals ale food containing 3% NaC1, body weight was not different among the groups. These results are consistent with those of Fregly et al. (8) who found that up to 6% NaCI in food failed to alter daily food intake. The increased water intake by rats consuming highsalt diets suggests that in a no-choice situation, sodium balance is maintained by excretion. It is clear that when otherwise equivalent foods vary in salt content, intake defends calories. The fact that the rats do not reduce their caloric intake as NaC1 content increased indicates that the rats could have eaten randomly from the available foods during the choice tests without deleterious effects. However, they did not. Although the amount of NaC1 consumed in the choice tests increased across tests with increasing NaC1 concentration &the alternate food, the increases were small and much less than the changes in average NaC1 concentration in the two available foods. Sodium balance was maintained in part by consumption. These data support those of the control rats in Bertino and Tordoff (4) that, over a 4-day period, appeared to eat less of a salty food option and more of the unsalted food as the concentration of salt in the salty food increased. When there is a choice between otherwise equivalent foods containing different amounts of salt, intake contributes to sodium balance.
°7=
.o_ 0 .5 o
a. I
.9o
E cO 0 03 nO e--
(,3 O~ Z
.8-
•
!
I
Naive (0.5 %)
o 0.5% [] 2.0%
T ]
,7" ,6" ,5" ,4" ,3" .2
1%
2%
3%
NaCI Content of Alternate Food FIG. 2. Proportion of total intake from the NaCl-free food (top panel) and resulting dietary NaCI content (bottom panel) as a function of the NaC1 content of the alternativelyavailablefood during the second choice test in Experiment II. Means (_+SE)are shown for four groups of rats, each of which had 8-weeksexposure to a diet containing a fixed amount of NaCI. All except the naive group had participated in the first choice test.
The mechanism of this diet selection cannot be determined from our data. It may be the result of a postingestive regulatory process for sodium or of palatability differences among the foods. The former explanation seems unlikelY in light of data indicating that in rats with a choice of water, saline solution, and salted food, intake of neither food nor the saline solution was affected by the level (0-6%) of NaC1 in the food, although intake of the saline solution was a function of its concentration (8). A postingestive mechanism should be insensitive to the source of sodium. The palatability hypothesis is difficult to test because the dominance of caloric regulation precludes a determination of the acceptance function over the range of salt concentrations in our foods. Experience eating a high-salt diet for 8 weeks had no effect on the level of NaC1 consumed by these rats, a finding which differs from that for humans (2) but replicates that for baboons (1). We were surprised that rats in all groups ate more salt in the second than the first choice test, indicating that age may affect the preferred level of salt intake in rats. Alternatively, extended consumption of the purified diets offered here may alter sodium balance in some unknown manner.
0
2
~ o
;~.~
~z
0-.~, pm-.-n
e. ~)
I
,-I
~ ~
~
l
|
~
l
u
"~
~
n
u
n
m
u
a~ e~ o o e,~
~
| ,I
Ne
~t
|
,
I ~
~.~
0
~
•-
. . . . . . ,,.~. ~.
,:
.
.
.
©.
.
.
~
.
~
m
.~ .~
~ , . . . ~ . ~ .
m~
peuJnsuoo ~,e!p u! IOeN %
0'~
~
•
pool % 0
IllOJt
uo!lJodoJd
545
~
546
COLLIER AND JOHNSON ACKNOWLEDGEMENTS
This research was supported by grant DK31016 from the National Institutes of Health and by a grant from Campbell's Soup Company.
Portions of the data from Experiment I were reported to the Eastern Psychological Association in 1986 by Karen CyBulski, whom we thank for her participation. Jose Naveira provided excellent technical assistance in Experiment II.
REFERENCES 1. Barnwell, G. M.; Dollahite, J.; Mitchell, D. S. Salt taste preference in baboons. Physiol. Behav. 37:279-284; 1986. 2. Beauchamp, G. K. The human preference for excess salt. Am. Sci. 75:27-33; 1987. 3. Beauchamp, G. K.; Bertino, M. Rats (Rattus norvegicus) do not prefer salted solid food. J. Comp. Psychol. 99:240-247; 1985. 4. Bertino, M.; Tordoff, M. G. Sodium depletion increases rats' preferences for salted food. Behav. Neurosci. 102:65-73; 1988. 5. Collier, G. H. Operant methodologies for studying feeding and drinking. In: Toates, F. M.; Rowland, N. E., eds. Techniques in the behavioral and neural sciences, vol. 1, Feeding and drinking. Amsterdam: Elsevier; 1987:37-76. 6. Denton, D. The hunger for salt. Berlin: Springer-Verlag; 1982. 7. Epstein, A.; Sakai, R. Angiotensin-aldosterone synergy and salt intake. In: Buckley, J. P.; Ferrario, C. M., eds. Brain peptides and catecholamines in cardiovascular regulation. New York: Raven; 1987:337-345.
8. Fregly, M. J.; Harper, J. M.; Radford, E. P., Jr. Regulation of sodium chloride intake by rats. Am. J. Physiol. 209:287-292; 1965. 9. Forman, S.; Falk, J. L. NaC1 solution ingestion in genetic (SHR) and aortic-ligation hypertension. Physiol. Behav. 22:371-377; 1979. I0. Levitsky, D. Feeding patterns in rats in response to fasts and changes in environmental conditions. Physiol. Behav. 5:291-300; 1970. 11. Richter, C. P. Total self regulatory functions in animals and human beings. Harvey Lect. 38:63-103; 1942-1943. 12. Stellar, E.; Hyman, R.; Samet, S. Gastric factors controlling water- and salt-solution-drinking. J. Comp. Physiol. Psychol. 47:220-226; 1954. 13. Stricker, E. M.; Verbalis, J. G. Hormones and behavior: The biology of thirst and sodium appetite. Am. Sci. 76:261-267; 1988. 14. Weiner, I. H.; Stellar, E. Salt preference of the rat determined by a single-stimulus method. J. Comp. Physiol. Psychol. 44:391-401; 1951. 15. Wolf, G.; Shulkin, J.; Simson, P. E. Multiple factors in the satiation of salt appetite. Behav. Neurosci. 98:651-673; 1984.
