- Email: [email protected]
Food ingestion is more important to plasma corticosterone dynamics than water intake in rats under restricted daily feeding
Physiology & Behavior, Vol. 37, pp. 791-795. Copyright©PergamonPress Ltd., 1986. Printedin the U.S.A.
0031-9384/86$3.00 + .00
Food Ingestion is More Important to Plasma Corticosterone Dynamics Than Water Intake in Rats Under Restricted Daily Feeding KEN-ICHI HONMA, SATO HONMA, TAKAYUKI HIRAI, YUMIKO KATSUNO AND TSUTOMU HIROSHIGE
D e p a r t m e n t o f Physiology, H o k k a i d o University School o f Medicine, Sapporo 060 Japan R e c e i v e d 25 S e p t e m b e r 1985 HONMA, K., S. HONMA, T. HIRAI, Y. KATSUNO AND T. HIROSHIGE. Food ingestion is more important to plasma corticosterone dynamics than water intake in rats under restricted daily feeding. PHYSIOL BEHAV 37(5) 791-795, 1986.--The roles of food and/or water ingestion in the regulation of plasma corticosterone level were examined in rats under restricted daily feeding. When the time of food-pellets and water supply was restricted to 2 hours in the early light period (meal feeding) for 2 weeks, the corticosterone level increased prior to meal (prefeeding peak). A similar prefeeding hormone peak was observed when supply of food-pellets was restricted to 2 hours with free-access to water (food restriction). In contrast, when water supply was restricted to 2 hours with free-access to food-pellets (water restriction), the hormone level before water supply did not increase as much as that under meal feeding or food restriction. Shortening of an available time for water under water restriction or prolongation of the restriction schedule failed to elevate the hormone level furthermore. On the other hand, the high prefeeding corticosterone level before meal decreased subsequently to meal feeding (prandial fall), which was not observed when rats were kept fasting during the meal time. This prandial fall of the hormone level was not observed by water intake alone, and closely related to food-pellets ingestion. It is concluded that food ingestion is more important than water intake to the formation of the prefeeding corticosterone peak and to the prandial fall of the hormone level under restricted daily feeding. Restricted daily feeding
Prefeeding corticosterone peak
Food ingestion
ioral study is confirmed in other variables such as plasma corticosterone, the interpretation would be more plausible that the prefeeding events are specifically related to food ingestion and are not generally associated with any motivationally relevant signal [15]. In the present study, we examined the effects of food and water ingestions separately on plasma corticosterone level in rats in attempting to know which factor contributes mainly to the changes in the corticosterone levels associated with restricted daily feeding.
IT is well established that plasma corticosterone levels in rats increase prior to the time of meal supply (prefeeding peak) and decrease subsequently to meal feeding (prandial fall), when animals are subjected to restricted feeding schedule, in which daily meal is restricted to a fixed time of the day [14,16]. Similar changes in plasma adrenocorticotropic hormone were observed [5,6]. Two different processes seem to be involved in the changes of the plasma corticosterone level under restricted feeding. One is a process related to the prefeeding corticosterone peak which has properties of oscillation [7,12], and the other is responsible for the prandial fall of the hormone level which is a direct result of meal intake [8]. Although precise mechanisms of these hormonal changes are not known, we have previously suggested an involvement of some metabolic process in establishing the prefeeding hormone peak [9]. The prefeeding corticosterone peak was reported to appear not only by daily restriction of food-pellets and water, but also by restriction of either of them [4, 8, 10, 11]. Recently, the restriction of water intake alone was shown to be ineffective in inducing the rat prefeeding activity peak [15] which seems to be regulated by a common mechanism to the prefeeding hormone peak [2,7]. If the finding in the behav-
METHOD
Animals and Housing Male rats of the Wistar strain were bred and reared in our animal quarter where environmental conditions are kept constant (temperature, 22_ + I°C; humidity, 60_+5%; lights-on from 06:00 to 18:00 hr). Light was supplied by fluorescent tubes and the light intensity at the surface of the rat cages was about 200 Ix. Four rats were housed together in a plastic cage (36x30x 17 cm) and fed commercial chow and tap water ad lib, unless otherwise stated. At the beginning of each experiment, rats were about 3 months old and their body weight were 300-400 g.
791
7~2
Q
H(~NM,\ / / Fo*~d
30
30
[
.
C~
..
if
20
20
i.
10 T. pg/dl
8
10
12
14
A
8
10
12
T i m e of day ( hours )
FIG. 1. Effects of meal feeding or starvation on plasma corticosterone levels in rats under restricted daily feeding (meal feeding). The hormone value is expressed with the mean (n = 8) and standard error (vertical bar). The meal was supplied from 10:00to 12:00 hr (rectangle column). Open circles with dotted lines indicate the hormone levels on the day of meal feeding, and closed circles with solid lines indicate those on the day of starvation. Asterisks indicate a statistically significant difference between meal feeding and starvation (*p<0.05, **p <0.01).
Water
30 .
1
20
2 10
t
I
8
I
i
10
i
12
Water
Blood Sampling and Plasma Corticosterone Determination Blood samples were obtained from freely moving rats. A small incision was made with a razor blade at the tip of the tail, and ca 30 tzl of blood were collected in a heparinized capillary tube. Eight rats (in 2 cages) were used to obtain the hormone level at one time point. Sampling from each cage was finished within 2 min from the initial contact with the cage, so that the total sampling time for one point was less than 5 rain. In one experiment (Experiment 1), blood was sampled from the same animal sequentially with intervals at least 4 hr apart from the previous sampling. To obtain the hormonal fluctuation in a shorter period than 4 hr, we subjected several groups of rats (each consists of 8 rats) to an identical experimental design and sampled one by one at the interval designated. Plasma corticosterone was determined by a competitive protein binding assay with 10/xl of specimen [8].
Statistical Analysis One-way or two-way analysis of variance was used for comparison of two mean corticosterone values. EXPERIMENTAL PROCEDURESAND RESULTS
Experiment I Procedures. Eighty rats (20 cages) were subjected to restricted daily feeding for 2 weeks. Rats were allowed to access to food-pellets and water from 10:00 to 12:00 hr each day (meal feeding). On the 14th day of meal feeding, plasma corticosterone levels were determined from 08:00 to 14:00 hr at 15 to 60 rain intervals. The feeding schedule was continued thereafter. Two days later, the hormone levels were determined again without meal supply. The food and water intakes were measured in 8 cages under ad lib feeding and 2 weeks after the beginning of meal feeding. The base-line control of plasma corticosterone was obtained at 10:00 hr from rats (n=8) fed ad lib. Results. Figure 1 illustrates effects of meal feeding and of starvation at the regular meal time on plasma corticosterone levels. Plasma corticosterone level prior to meal increased
30 20 10
I
8
I
I
I
I
tO
|
12
Time of day ( hours )
FIG. 2. Effects of food-pellets (food restriction; upper) or water restriction (2 hr water; middle, 30 min water; lower) on plasma corticosterone levels in rats. The hormone value is expressed with the mean (n=8) and standard error (vertical bar). The time of food or water supply is indicated by a rectangle column. Open circles with dotted lines in each panel indicate the hormone levels under meal feeding (Experiment 1). Asterisks indicate a statistically significant difference between meal feeding and the experimental group (*p<0.05, **p<0.01).
under meal feeding as compared with the base-line control (3.3_+0.6 p,g/dl). When the regular meal was supplied, the high prefeeding hormone level decreased rapidly almost to the basal morning level. In contrast, when the meal was omitted, the high hormone level persisted during the whole meal time. But the hormone level decreased gradually thereafter. The corticosterone levels under meal feeding were used as the meal-feeding control in the following experiments.
Experiment 2 Procedures. Three groups of rats were subjected to feeding schedules of the following designs for 2 weeks. Each was consisted of 48 rats (12 cages). In the first group rats had free-access to food-pellets from 10:00 to 12:00 hr (food restriction). Water was supplied continuously. In the second group rats had free-access to water from 10:00 to 12:00 hr
FOOD AND PLASMA CORTICOSTERONE
793
supply
Water
30
30
i ~
i
I
20
20
.u
""
10
'i
' I m
E
+
i
8
i
I
10
4 I
I
14
12
o'3
n.,
8
10
Food
30
12
i
Time of day ( hours )
FIG. 3. Effects of prolonged meal feeding (open circles with dotted lines), food restriction (closed circles with solid lines), and water restriction (closed circles with broken lines) on plasma corticosterone levels. The time of respective diet supply is indicated by a rectangle column. Asterisks indicate a statistically significant difference between prolonged meal feeding and prolonged water or food restriction (*p<0.05, **p<0.01).
e...
20
o
10
I
o ,,9 i
E
I
8
i
l
i
a" (water restriction, 2 hr). In the third group free-access to water was restricted to 30 min per day from 10:00 to 10:30 hr (water restriction, 30 min). To the latter two groups foodpellets were supplied continuously. Plasma corticosterone levels were determined from 08:00 to 12:00 hr at 30 to 60 min intervals. Food and water intakes were measured in 8 cages for each group. An additional experiment with the same design as mentioned above was performed extending the period of restriction schedule to 4 weeks. Each group was designated by adding a prefix, "prolonged." A group of water restriction (30 min) was excluded. The prolonged meal-feeding control was also obtained. Results. Figure 2 illustrates results of daily food or water restriction on the plasma corticosterone level. When foodpellets supply was restricted with free-access to water (food restriction; upper panel), a prefeeding hormone peak similar to that under meal feeding was established, except for the hormone value at 08:00 hr which was significantly lower under food restriction. The rapid prandial decrease in the hormone level was also detected under food restriction. On the other hand, when water supply was restricted with freeaccess to food-pellets (water restriction, 2 hr; middle panel), the hormone levels prior to water supply were significantly lower than those under meal feeding or food restriction. However, the prefeeding hormone level at 10:00 hr was high under the water restriction as compared with the base-line control under ad lib feeding (7.5_+0.9/~g/dl vs. 3.3_+0.6/zg/dl). Shortening of the access-time to 30 min under water restriction did not elevate the prefeeding hormone level furthermore (7.5_+0.9/xg/dl vs. 8.1_+1.7 /zg/dl at 10:00 hr). A decrease in the plasma hormone level 30 min after water intake was slight but statistically significant in both cases (water restriction, 2 hr, 7.5_+0.9/zg/dl vs. 4.9_+0.7 tzg/dl; water restriction, 30 rain, 8.1-4--1.7/xg/dl vs. 4.6_+0.3/zg/dl). Figure 3 illustrates the effects of prolonged food or water restriction on plasma corticosterone levels. The prefeeding hormone levels under prolonged water restriction were significantly lower at 9:00 and 10:00 hr than those under prolonged meal feeding, and were not different from those ob-
I
10
i
i
12
14
12
14
Food Water
30
20 10
8
tO T i m e o f day
( hours )
FIG. 4. Effects of water (upper), food-pellets (middle) or both (lower) supply on prandial fall of plasma corticosterone levels in rats subjected to meal feeding. The hormone value is expressed with the mean (n=8) and standard error (vertical bar). The time of meal is indicated by a rectangle column. Open circles with dotted lines indicate the hormone levels under meal feeding (Experiment 1). Asterisks indicate a statistically significant difference between meal feeding and a respective feeding condition (*p<0.05, **p<0.01).
tained under the 2 week water restriction (9.5+ 1.1/xg/dl vs. 7.5_+0.7/zg/dl at 10:00 hr). The hormone levels under prolonged food restriction were not different from those under prolonged meal feeding. On the other hand, the prandial hormone levels under prolonged water restriction were significantly lower than those under prolonged meal feeding. The prandial hormone levels at 11:00 and 12:00 hr were also lower under prolonged food restriction than those under prolonged meal feeding.
Experiment 3 Procedure. Three groups of rats were subjected to the same feeding schedule as Experiment 1 (meal available from 10:00 to 12:00 hr) for 2 weeks. The first two groups consisted of 80 rats (20 cages) and the third group of 88 rats (22 cages). After 2 weeks, only water was supplied to the first group at the regular meal time (water ingestion). To the second group,
794
HON M A/¢1 ~~/+. TABLE 1
FOOD AND W A T E R I N T A K E S (g/100g B.W.) UNDER D I F F E R E N T FEEDING CONDITIONS
Food-pellets Control (ad lib) Meal feeding Food restriction Water restriction (2 h) Water restriction (30 min) Water ingestion Food ingestion Water and food ingestion
T: R: R: T: R: T: R: T: R: R: R:
6.86 ± 0.13 5.31 _+ 0.14 4.05 ± 0.09" -1.44 ± 0.13 5.39 _+ 0.17b 1.81 -+ 0.10 5.90 _+ 0.20h -1.92 _+ 0.06" 4.32 ± 0.13;'
Water 6.63 + 0.34 5.72 _+ 0.22 3.39 _+ 0.32 11.94 ± 0.54h 5.14 + 0.28 4.51 -+ 0.24~ 3.59 _+ 0.13~' 5.40 +_ 0.27
Food and water intakes were measured during the restricted period (R) and a 24 hr period (T). The amount is expressed with the mean and standard error (8 cages). Alphabets at the shoulder of figure indicate statistically significant difference (p <0.01) against the value of meal feeding (a) or the controls under ad lib feeding lb).
food-pellets were supplied without water at the meal (food ingestion). To the third group, water was supplied from the beginning of the meal time and food-pellets were provided one hour later (water and food ingestion). Blood corticosterone levels were determined before and after the meal at 15 to 60 min intervals. Food and water intakes were measured in 8 cages. R e s u l t s . Figure 4 illustrates effects of water or foodpellets supply on the prandial fall of prefeeding corticosterone peak under restricted daily feeding. Neither water nor food-pellets supply was effective by itself in reducing the hormone level. When water was supplied as usual but foodpellets were given one hour later, the high corticosterone level which had been unaffected by water decreased rapidly after food-pellets supply. Table l summarizes the food and water intakes of rats under the 8 different conditions mentioned above (those of prolonged restriction were not included). Food and water consumptions were measured per rat cage. Under meal feeding, the food and water intakes decreased slightly but significantly as compared with the controls under ad lib feeding. When supply of food-pellets was restricted with free-access to water (food restriction), the food intake decreased more than that under meal feeding, whereas water intake almost doubled the control level. In contrast, both water and food intakes were reduced significantly when water supply was restricted. Under water restrictions, one fourth of the daily food was consumed during the time of water supply. Under meal feeding, food intake was diminished significantly when rats were not allowed to drink simultaneously (food ingestion). Water intake was also diminished when it was supplied without food-pellets (water ingestion). It was, however, not different from that under food restrictions. DISCUSSION
As clearly demonstrated in Fig. 2, the prefeeding corticosterone peak was established by restriction of food-pellets alone with free-access to water for 2 weeks. In contrast, restriction of water with free-access to food-pellets for the
same period failed to induce the prefeeding hormone peak to an extent comparable in amplitude to that under meal feeding. Neither shortening of an available time for watet+ (30 min) nor prolongation of the period of restriction schedule 14 weeks) was effective in increasing the prefeeding hormone level. These results indicate the predominant role of foodpellets over water in the formation of the prefeeding corticosterone peak under restricted daily feeding. A slight but significant elevation in the prefeeding hormone level was observed under both schedules of water restriction (Fig. 1), which may suggest some role of water restriction in the peak formation. But as demonstrated in Table 1, even under water restriction one fourth of the daily food intake occurred during the time of water supply. Therefore it is possible to interpret that a slight elevation of the prefeeding hormone level under water restriction was a result not of the water restriction but of an unusual feeding time associated with the water restriction. The present findings may also exclude an involvement of vasopressin as a main physiological factor in the formation of the prefeeding corticosterone peak, which has been previously argued [10]. An attenuated hormone peak by water restriction is consistent with the finding of Mistlberger and Rechtschaffen [15] but contradictory to those in other behavioral [3,17] as well as endocrinological studies [4,11]. The reason for this discrepancy is unknown but seems to be related to differences in sex, body weight and contents of meal supplied. Because these factors are well-known to influence the general metabolic state which has been suggested to be involved in the formation of the prefeeding hormone peak [9]. In the present study, rats were housed in groups, and group housing has been reported to elevate the base-line level of plasma corticosterone [1]. We compared the amplitude and duration of the prefeeding corticosterone peak between rats in group and individual housings, and did not find any meaningful difference between the two different modes of habituation (unpublished observation). The important role of food ingestion in the control of plasma corticosterone was also demonstrated in the prandial fall of the hormone level. Neither water nor food-pellets supply succeeded in decreasing the high prefeeding hormone level by itself (Fig. 4). However as indicated in Table I, food consumption was considerably reduced when food-pellets were supplied without water, whereas water intake was comparable to that in food restriction when water was supplied without food-pellets. The result indicates that rats had difficulties in ingesting food-pellets without water. In this regard, the above experiment was not properly designed to see effects of food ingestion on the plasma hormone level. On the other hand, when water was supplied beforehand, which had no effect on the hormone level by itself, the hormone level decreased subsequently to food-pellets supply. Under these conditions, rats could ingest food-pellets as much as under food restriction, in which the prandial fall of the hormone level was observed. From these results we conclude that food but not water ingestion is the main trigger to induce the prandial fall of plasma corticosterone level. We do not know the mechanism which is involved in the reduction of the hormone level. Previously Doell et al, [4] excluded possibilities of the changes in the distribution volume of the hormone and in the clearance of the hormone after meal feeding. Heybach and Vernikos-Danellis [6] observed a prandial fall of plasma ACTH concentration and a concomitant increase in pituitary ACTH level under restricted feeding. They also found that ACTH response to an acute stress
FOOD A N D P L A S M A C O R T I C O S T E R O N E
795
was inhibited when it was applied a few min after meal presentation, and suggested that the prandial fall of plasma A C T H and corticosterone levels involved active inhibitory mechanisms on A C T H secretion. On the other hand, Wilkinson et al. [19] suggested that decrease in plasma corticosterone occurring after drinking in water-restricted rats was not dependent on changes in plasma A C T H concentration. Recently the jejunal resection was reported to modify the plasma corticosterone level in rats [13]. The subdiaphragmatic vagotomy had no effect on the prefeeding hormone peak [18] but prevents a rapid prandial fall of the hormone level (unpublished observation). Therefore it might be possible to speculate that a signal mediating the information of food in-
gestion arises from the digestive system and goes to the central nervous system which in turn inhibits the release of corticosterone from the adrenal cortex. In conclusion, food ingestion is more important than water intake in establishing the prefeeding corticosterone peak and in inducing the prandial fall of the plasma corticosterone level under restricted daily feeding. ACKNOWLEDGEMENTS This work was supported in part by grants from Ministry of Education, Science, and Culture of Japan (No. 59224001 and 60216001).
REFERENCES 1. Barrett, A. M. and M. A. Stockham. The effect of housing conditions and simple experimental procedures upon the corticostetone level in the plasma of rats, J Endocrinol 26: 97-105, 1963. 2. Boulos, Z. and M. Terman. Food availability and daily biological rhythms. Neurosci Biobehav Rev 4: 119-131, 1980. 3. Dhume, R. A. and M. G. Gogate. Water as entrainer of circadian running activity in rat. Physiol Behav 28: 431-436, 1982. 4. Doell, R. G., M. F. Dallman, R. B. Clayton, G. D. Gray and S. Levine. Dissociation of adrenal corticosteroid production from ACTH in water-restricted female rats. Am J Physiol 241: R21R24, 1981. 5. Heybach, J. P. and J. Vernikos-Danellis. Inhibition of adrenocorticotrophin secretion during deprivation-induced eating and drinking in rats. Neuroendocrinology 28: 329--338, 1979. 6. Heybach, J. P. and J. Vernikos-Danellis. Inhibition of the pituitary-adrenal response to stress during deprivation-induced feeding. Endocrinology 104: 967-973, 1979. 7. Honma, K., C. von Goetz and J. Aschoff. Effects of restricted daily feeding on freerunning circadian rhythms in rats. Physiol Behav 30: 905-913, 1983. 8. Honma, K., S. Honma and T. Hiroshige. Critical role of food amount for prefeeding corticosterone peak in rats. Am J Physh;l 245: R339-R344, 1983. 9. Honma, K., S. Honma and T. Hiroshige. Feeding-associated corticosterone peak in rats under various feeding cycles. Am J Physiol 246: R721-R726, 1984. 10. Itoh, S., G. Katsuura and R.' Hirota. Conditioned circadian rhythm of plasma corticosterone in the rat induced by food restriction. Jpn J Physiol 30: 365-375, 1980.
11. Johnson, J. T. and S. Levine. Influence of water deprivation on adrenocortical rhythms. Neuroendocrinology 11: 268-273, 1973. 12. Kato, H., M. Saito and M. Suda. Effects of starvation on the circadian adrenocortical rhythm in rats. Endocrinology 106: 918-921, 1978. 13. Kato, H., M. Saito and T. Shimazu. Attenuated blood corticosterone rhythm in rats with jejunal resection. Life Sci 34: 331335, 1984. 14. Krieger, D. T. Food and water restriction shifts corticosterone, temperature, activity and brain amine periodicity. Endocrinology 95: 1195-1201, 1974. 15. Mistlberger, R. M. and A. Rechtschaffen. Periodic water availability is not a potent zeitgeber for entrainment of circadian locomotor rhythms in rats. Physiol Behav 34: 17-22, 1985. 16. Moberg, G. P,, L. L. Bellinger and V. E. Mendel. Effect of meal feeding on daily rhythm of plasma corticosterone and growth hormone in the rat. Neuroendocrinology 19: 160-169, 1975, 17. Moore, R. Y. Suprachiasmatic nucleus, secondary synchronizing stimuli and the central neural control of circadian rhythms. Brain Res 183: 13-28, 1980. 18. Moreira, A. C. and D. T. Krieger. The effects of subdiaphragmatic vagotomy on circadian corticosterone rhythmicity in rats with continuous or restricted food access. Physiol Behav 28: 787-790, 1982. 19. Wilkinson, C. W., J. Shinsako and M. F. Dallman. Rapid decreases in adrenal and plasma corticosterone concentrations after drinking are not mediated by changes in plasma adrenocorticotropin concentration. Endocrinology 110: 15991606, 1982.
0031-9384/86$3.00 + .00
Food Ingestion is More Important to Plasma Corticosterone Dynamics Than Water Intake in Rats Under Restricted Daily Feeding KEN-ICHI HONMA, SATO HONMA, TAKAYUKI HIRAI, YUMIKO KATSUNO AND TSUTOMU HIROSHIGE
D e p a r t m e n t o f Physiology, H o k k a i d o University School o f Medicine, Sapporo 060 Japan R e c e i v e d 25 S e p t e m b e r 1985 HONMA, K., S. HONMA, T. HIRAI, Y. KATSUNO AND T. HIROSHIGE. Food ingestion is more important to plasma corticosterone dynamics than water intake in rats under restricted daily feeding. PHYSIOL BEHAV 37(5) 791-795, 1986.--The roles of food and/or water ingestion in the regulation of plasma corticosterone level were examined in rats under restricted daily feeding. When the time of food-pellets and water supply was restricted to 2 hours in the early light period (meal feeding) for 2 weeks, the corticosterone level increased prior to meal (prefeeding peak). A similar prefeeding hormone peak was observed when supply of food-pellets was restricted to 2 hours with free-access to water (food restriction). In contrast, when water supply was restricted to 2 hours with free-access to food-pellets (water restriction), the hormone level before water supply did not increase as much as that under meal feeding or food restriction. Shortening of an available time for water under water restriction or prolongation of the restriction schedule failed to elevate the hormone level furthermore. On the other hand, the high prefeeding corticosterone level before meal decreased subsequently to meal feeding (prandial fall), which was not observed when rats were kept fasting during the meal time. This prandial fall of the hormone level was not observed by water intake alone, and closely related to food-pellets ingestion. It is concluded that food ingestion is more important than water intake to the formation of the prefeeding corticosterone peak and to the prandial fall of the hormone level under restricted daily feeding. Restricted daily feeding
Prefeeding corticosterone peak
Food ingestion
ioral study is confirmed in other variables such as plasma corticosterone, the interpretation would be more plausible that the prefeeding events are specifically related to food ingestion and are not generally associated with any motivationally relevant signal [15]. In the present study, we examined the effects of food and water ingestions separately on plasma corticosterone level in rats in attempting to know which factor contributes mainly to the changes in the corticosterone levels associated with restricted daily feeding.
IT is well established that plasma corticosterone levels in rats increase prior to the time of meal supply (prefeeding peak) and decrease subsequently to meal feeding (prandial fall), when animals are subjected to restricted feeding schedule, in which daily meal is restricted to a fixed time of the day [14,16]. Similar changes in plasma adrenocorticotropic hormone were observed [5,6]. Two different processes seem to be involved in the changes of the plasma corticosterone level under restricted feeding. One is a process related to the prefeeding corticosterone peak which has properties of oscillation [7,12], and the other is responsible for the prandial fall of the hormone level which is a direct result of meal intake [8]. Although precise mechanisms of these hormonal changes are not known, we have previously suggested an involvement of some metabolic process in establishing the prefeeding hormone peak [9]. The prefeeding corticosterone peak was reported to appear not only by daily restriction of food-pellets and water, but also by restriction of either of them [4, 8, 10, 11]. Recently, the restriction of water intake alone was shown to be ineffective in inducing the rat prefeeding activity peak [15] which seems to be regulated by a common mechanism to the prefeeding hormone peak [2,7]. If the finding in the behav-
METHOD
Animals and Housing Male rats of the Wistar strain were bred and reared in our animal quarter where environmental conditions are kept constant (temperature, 22_ + I°C; humidity, 60_+5%; lights-on from 06:00 to 18:00 hr). Light was supplied by fluorescent tubes and the light intensity at the surface of the rat cages was about 200 Ix. Four rats were housed together in a plastic cage (36x30x 17 cm) and fed commercial chow and tap water ad lib, unless otherwise stated. At the beginning of each experiment, rats were about 3 months old and their body weight were 300-400 g.
791
7~2
Q
H(~NM,\ / / Fo*~d
30
30
[
.
C~
..
if
20
20
i.
10 T. pg/dl
8
10
12
14
A
8
10
12
T i m e of day ( hours )
FIG. 1. Effects of meal feeding or starvation on plasma corticosterone levels in rats under restricted daily feeding (meal feeding). The hormone value is expressed with the mean (n = 8) and standard error (vertical bar). The meal was supplied from 10:00to 12:00 hr (rectangle column). Open circles with dotted lines indicate the hormone levels on the day of meal feeding, and closed circles with solid lines indicate those on the day of starvation. Asterisks indicate a statistically significant difference between meal feeding and starvation (*p<0.05, **p <0.01).
Water
30 .
1
20
2 10
t
I
8
I
i
10
i
12
Water
Blood Sampling and Plasma Corticosterone Determination Blood samples were obtained from freely moving rats. A small incision was made with a razor blade at the tip of the tail, and ca 30 tzl of blood were collected in a heparinized capillary tube. Eight rats (in 2 cages) were used to obtain the hormone level at one time point. Sampling from each cage was finished within 2 min from the initial contact with the cage, so that the total sampling time for one point was less than 5 rain. In one experiment (Experiment 1), blood was sampled from the same animal sequentially with intervals at least 4 hr apart from the previous sampling. To obtain the hormonal fluctuation in a shorter period than 4 hr, we subjected several groups of rats (each consists of 8 rats) to an identical experimental design and sampled one by one at the interval designated. Plasma corticosterone was determined by a competitive protein binding assay with 10/xl of specimen [8].
Statistical Analysis One-way or two-way analysis of variance was used for comparison of two mean corticosterone values. EXPERIMENTAL PROCEDURESAND RESULTS
Experiment I Procedures. Eighty rats (20 cages) were subjected to restricted daily feeding for 2 weeks. Rats were allowed to access to food-pellets and water from 10:00 to 12:00 hr each day (meal feeding). On the 14th day of meal feeding, plasma corticosterone levels were determined from 08:00 to 14:00 hr at 15 to 60 rain intervals. The feeding schedule was continued thereafter. Two days later, the hormone levels were determined again without meal supply. The food and water intakes were measured in 8 cages under ad lib feeding and 2 weeks after the beginning of meal feeding. The base-line control of plasma corticosterone was obtained at 10:00 hr from rats (n=8) fed ad lib. Results. Figure 1 illustrates effects of meal feeding and of starvation at the regular meal time on plasma corticosterone levels. Plasma corticosterone level prior to meal increased
30 20 10
I
8
I
I
I
I
tO
|
12
Time of day ( hours )
FIG. 2. Effects of food-pellets (food restriction; upper) or water restriction (2 hr water; middle, 30 min water; lower) on plasma corticosterone levels in rats. The hormone value is expressed with the mean (n=8) and standard error (vertical bar). The time of food or water supply is indicated by a rectangle column. Open circles with dotted lines in each panel indicate the hormone levels under meal feeding (Experiment 1). Asterisks indicate a statistically significant difference between meal feeding and the experimental group (*p<0.05, **p<0.01).
under meal feeding as compared with the base-line control (3.3_+0.6 p,g/dl). When the regular meal was supplied, the high prefeeding hormone level decreased rapidly almost to the basal morning level. In contrast, when the meal was omitted, the high hormone level persisted during the whole meal time. But the hormone level decreased gradually thereafter. The corticosterone levels under meal feeding were used as the meal-feeding control in the following experiments.
Experiment 2 Procedures. Three groups of rats were subjected to feeding schedules of the following designs for 2 weeks. Each was consisted of 48 rats (12 cages). In the first group rats had free-access to food-pellets from 10:00 to 12:00 hr (food restriction). Water was supplied continuously. In the second group rats had free-access to water from 10:00 to 12:00 hr
FOOD AND PLASMA CORTICOSTERONE
793
supply
Water
30
30
i ~
i
I
20
20
.u
""
10
'i
' I m
E
+
i
8
i
I
10
4 I
I
14
12
o'3
n.,
8
10
Food
30
12
i
Time of day ( hours )
FIG. 3. Effects of prolonged meal feeding (open circles with dotted lines), food restriction (closed circles with solid lines), and water restriction (closed circles with broken lines) on plasma corticosterone levels. The time of respective diet supply is indicated by a rectangle column. Asterisks indicate a statistically significant difference between prolonged meal feeding and prolonged water or food restriction (*p<0.05, **p<0.01).
e...
20
o
10
I
o ,,9 i
E
I
8
i
l
i
a" (water restriction, 2 hr). In the third group free-access to water was restricted to 30 min per day from 10:00 to 10:30 hr (water restriction, 30 min). To the latter two groups foodpellets were supplied continuously. Plasma corticosterone levels were determined from 08:00 to 12:00 hr at 30 to 60 min intervals. Food and water intakes were measured in 8 cages for each group. An additional experiment with the same design as mentioned above was performed extending the period of restriction schedule to 4 weeks. Each group was designated by adding a prefix, "prolonged." A group of water restriction (30 min) was excluded. The prolonged meal-feeding control was also obtained. Results. Figure 2 illustrates results of daily food or water restriction on the plasma corticosterone level. When foodpellets supply was restricted with free-access to water (food restriction; upper panel), a prefeeding hormone peak similar to that under meal feeding was established, except for the hormone value at 08:00 hr which was significantly lower under food restriction. The rapid prandial decrease in the hormone level was also detected under food restriction. On the other hand, when water supply was restricted with freeaccess to food-pellets (water restriction, 2 hr; middle panel), the hormone levels prior to water supply were significantly lower than those under meal feeding or food restriction. However, the prefeeding hormone level at 10:00 hr was high under the water restriction as compared with the base-line control under ad lib feeding (7.5_+0.9/~g/dl vs. 3.3_+0.6/zg/dl). Shortening of the access-time to 30 min under water restriction did not elevate the prefeeding hormone level furthermore (7.5_+0.9/xg/dl vs. 8.1_+1.7 /zg/dl at 10:00 hr). A decrease in the plasma hormone level 30 min after water intake was slight but statistically significant in both cases (water restriction, 2 hr, 7.5_+0.9/zg/dl vs. 4.9_+0.7 tzg/dl; water restriction, 30 rain, 8.1-4--1.7/xg/dl vs. 4.6_+0.3/zg/dl). Figure 3 illustrates the effects of prolonged food or water restriction on plasma corticosterone levels. The prefeeding hormone levels under prolonged water restriction were significantly lower at 9:00 and 10:00 hr than those under prolonged meal feeding, and were not different from those ob-
I
10
i
i
12
14
12
14
Food Water
30
20 10
8
tO T i m e o f day
( hours )
FIG. 4. Effects of water (upper), food-pellets (middle) or both (lower) supply on prandial fall of plasma corticosterone levels in rats subjected to meal feeding. The hormone value is expressed with the mean (n=8) and standard error (vertical bar). The time of meal is indicated by a rectangle column. Open circles with dotted lines indicate the hormone levels under meal feeding (Experiment 1). Asterisks indicate a statistically significant difference between meal feeding and a respective feeding condition (*p<0.05, **p<0.01).
tained under the 2 week water restriction (9.5+ 1.1/xg/dl vs. 7.5_+0.7/zg/dl at 10:00 hr). The hormone levels under prolonged food restriction were not different from those under prolonged meal feeding. On the other hand, the prandial hormone levels under prolonged water restriction were significantly lower than those under prolonged meal feeding. The prandial hormone levels at 11:00 and 12:00 hr were also lower under prolonged food restriction than those under prolonged meal feeding.
Experiment 3 Procedure. Three groups of rats were subjected to the same feeding schedule as Experiment 1 (meal available from 10:00 to 12:00 hr) for 2 weeks. The first two groups consisted of 80 rats (20 cages) and the third group of 88 rats (22 cages). After 2 weeks, only water was supplied to the first group at the regular meal time (water ingestion). To the second group,
794
HON M A/¢1 ~~/+. TABLE 1
FOOD AND W A T E R I N T A K E S (g/100g B.W.) UNDER D I F F E R E N T FEEDING CONDITIONS
Food-pellets Control (ad lib) Meal feeding Food restriction Water restriction (2 h) Water restriction (30 min) Water ingestion Food ingestion Water and food ingestion
T: R: R: T: R: T: R: T: R: R: R:
6.86 ± 0.13 5.31 _+ 0.14 4.05 ± 0.09" -1.44 ± 0.13 5.39 _+ 0.17b 1.81 -+ 0.10 5.90 _+ 0.20h -1.92 _+ 0.06" 4.32 ± 0.13;'
Water 6.63 + 0.34 5.72 _+ 0.22 3.39 _+ 0.32 11.94 ± 0.54h 5.14 + 0.28 4.51 -+ 0.24~ 3.59 _+ 0.13~' 5.40 +_ 0.27
Food and water intakes were measured during the restricted period (R) and a 24 hr period (T). The amount is expressed with the mean and standard error (8 cages). Alphabets at the shoulder of figure indicate statistically significant difference (p <0.01) against the value of meal feeding (a) or the controls under ad lib feeding lb).
food-pellets were supplied without water at the meal (food ingestion). To the third group, water was supplied from the beginning of the meal time and food-pellets were provided one hour later (water and food ingestion). Blood corticosterone levels were determined before and after the meal at 15 to 60 min intervals. Food and water intakes were measured in 8 cages. R e s u l t s . Figure 4 illustrates effects of water or foodpellets supply on the prandial fall of prefeeding corticosterone peak under restricted daily feeding. Neither water nor food-pellets supply was effective by itself in reducing the hormone level. When water was supplied as usual but foodpellets were given one hour later, the high corticosterone level which had been unaffected by water decreased rapidly after food-pellets supply. Table l summarizes the food and water intakes of rats under the 8 different conditions mentioned above (those of prolonged restriction were not included). Food and water consumptions were measured per rat cage. Under meal feeding, the food and water intakes decreased slightly but significantly as compared with the controls under ad lib feeding. When supply of food-pellets was restricted with free-access to water (food restriction), the food intake decreased more than that under meal feeding, whereas water intake almost doubled the control level. In contrast, both water and food intakes were reduced significantly when water supply was restricted. Under water restrictions, one fourth of the daily food was consumed during the time of water supply. Under meal feeding, food intake was diminished significantly when rats were not allowed to drink simultaneously (food ingestion). Water intake was also diminished when it was supplied without food-pellets (water ingestion). It was, however, not different from that under food restrictions. DISCUSSION
As clearly demonstrated in Fig. 2, the prefeeding corticosterone peak was established by restriction of food-pellets alone with free-access to water for 2 weeks. In contrast, restriction of water with free-access to food-pellets for the
same period failed to induce the prefeeding hormone peak to an extent comparable in amplitude to that under meal feeding. Neither shortening of an available time for watet+ (30 min) nor prolongation of the period of restriction schedule 14 weeks) was effective in increasing the prefeeding hormone level. These results indicate the predominant role of foodpellets over water in the formation of the prefeeding corticosterone peak under restricted daily feeding. A slight but significant elevation in the prefeeding hormone level was observed under both schedules of water restriction (Fig. 1), which may suggest some role of water restriction in the peak formation. But as demonstrated in Table 1, even under water restriction one fourth of the daily food intake occurred during the time of water supply. Therefore it is possible to interpret that a slight elevation of the prefeeding hormone level under water restriction was a result not of the water restriction but of an unusual feeding time associated with the water restriction. The present findings may also exclude an involvement of vasopressin as a main physiological factor in the formation of the prefeeding corticosterone peak, which has been previously argued [10]. An attenuated hormone peak by water restriction is consistent with the finding of Mistlberger and Rechtschaffen [15] but contradictory to those in other behavioral [3,17] as well as endocrinological studies [4,11]. The reason for this discrepancy is unknown but seems to be related to differences in sex, body weight and contents of meal supplied. Because these factors are well-known to influence the general metabolic state which has been suggested to be involved in the formation of the prefeeding hormone peak [9]. In the present study, rats were housed in groups, and group housing has been reported to elevate the base-line level of plasma corticosterone [1]. We compared the amplitude and duration of the prefeeding corticosterone peak between rats in group and individual housings, and did not find any meaningful difference between the two different modes of habituation (unpublished observation). The important role of food ingestion in the control of plasma corticosterone was also demonstrated in the prandial fall of the hormone level. Neither water nor food-pellets supply succeeded in decreasing the high prefeeding hormone level by itself (Fig. 4). However as indicated in Table I, food consumption was considerably reduced when food-pellets were supplied without water, whereas water intake was comparable to that in food restriction when water was supplied without food-pellets. The result indicates that rats had difficulties in ingesting food-pellets without water. In this regard, the above experiment was not properly designed to see effects of food ingestion on the plasma hormone level. On the other hand, when water was supplied beforehand, which had no effect on the hormone level by itself, the hormone level decreased subsequently to food-pellets supply. Under these conditions, rats could ingest food-pellets as much as under food restriction, in which the prandial fall of the hormone level was observed. From these results we conclude that food but not water ingestion is the main trigger to induce the prandial fall of plasma corticosterone level. We do not know the mechanism which is involved in the reduction of the hormone level. Previously Doell et al, [4] excluded possibilities of the changes in the distribution volume of the hormone and in the clearance of the hormone after meal feeding. Heybach and Vernikos-Danellis [6] observed a prandial fall of plasma ACTH concentration and a concomitant increase in pituitary ACTH level under restricted feeding. They also found that ACTH response to an acute stress
FOOD A N D P L A S M A C O R T I C O S T E R O N E
795
was inhibited when it was applied a few min after meal presentation, and suggested that the prandial fall of plasma A C T H and corticosterone levels involved active inhibitory mechanisms on A C T H secretion. On the other hand, Wilkinson et al. [19] suggested that decrease in plasma corticosterone occurring after drinking in water-restricted rats was not dependent on changes in plasma A C T H concentration. Recently the jejunal resection was reported to modify the plasma corticosterone level in rats [13]. The subdiaphragmatic vagotomy had no effect on the prefeeding hormone peak [18] but prevents a rapid prandial fall of the hormone level (unpublished observation). Therefore it might be possible to speculate that a signal mediating the information of food in-
gestion arises from the digestive system and goes to the central nervous system which in turn inhibits the release of corticosterone from the adrenal cortex. In conclusion, food ingestion is more important than water intake in establishing the prefeeding corticosterone peak and in inducing the prandial fall of the plasma corticosterone level under restricted daily feeding. ACKNOWLEDGEMENTS This work was supported in part by grants from Ministry of Education, Science, and Culture of Japan (No. 59224001 and 60216001).
REFERENCES 1. Barrett, A. M. and M. A. Stockham. The effect of housing conditions and simple experimental procedures upon the corticostetone level in the plasma of rats, J Endocrinol 26: 97-105, 1963. 2. Boulos, Z. and M. Terman. Food availability and daily biological rhythms. Neurosci Biobehav Rev 4: 119-131, 1980. 3. Dhume, R. A. and M. G. Gogate. Water as entrainer of circadian running activity in rat. Physiol Behav 28: 431-436, 1982. 4. Doell, R. G., M. F. Dallman, R. B. Clayton, G. D. Gray and S. Levine. Dissociation of adrenal corticosteroid production from ACTH in water-restricted female rats. Am J Physiol 241: R21R24, 1981. 5. Heybach, J. P. and J. Vernikos-Danellis. Inhibition of adrenocorticotrophin secretion during deprivation-induced eating and drinking in rats. Neuroendocrinology 28: 329--338, 1979. 6. Heybach, J. P. and J. Vernikos-Danellis. Inhibition of the pituitary-adrenal response to stress during deprivation-induced feeding. Endocrinology 104: 967-973, 1979. 7. Honma, K., C. von Goetz and J. Aschoff. Effects of restricted daily feeding on freerunning circadian rhythms in rats. Physiol Behav 30: 905-913, 1983. 8. Honma, K., S. Honma and T. Hiroshige. Critical role of food amount for prefeeding corticosterone peak in rats. Am J Physh;l 245: R339-R344, 1983. 9. Honma, K., S. Honma and T. Hiroshige. Feeding-associated corticosterone peak in rats under various feeding cycles. Am J Physiol 246: R721-R726, 1984. 10. Itoh, S., G. Katsuura and R.' Hirota. Conditioned circadian rhythm of plasma corticosterone in the rat induced by food restriction. Jpn J Physiol 30: 365-375, 1980.
11. Johnson, J. T. and S. Levine. Influence of water deprivation on adrenocortical rhythms. Neuroendocrinology 11: 268-273, 1973. 12. Kato, H., M. Saito and M. Suda. Effects of starvation on the circadian adrenocortical rhythm in rats. Endocrinology 106: 918-921, 1978. 13. Kato, H., M. Saito and T. Shimazu. Attenuated blood corticosterone rhythm in rats with jejunal resection. Life Sci 34: 331335, 1984. 14. Krieger, D. T. Food and water restriction shifts corticosterone, temperature, activity and brain amine periodicity. Endocrinology 95: 1195-1201, 1974. 15. Mistlberger, R. M. and A. Rechtschaffen. Periodic water availability is not a potent zeitgeber for entrainment of circadian locomotor rhythms in rats. Physiol Behav 34: 17-22, 1985. 16. Moberg, G. P,, L. L. Bellinger and V. E. Mendel. Effect of meal feeding on daily rhythm of plasma corticosterone and growth hormone in the rat. Neuroendocrinology 19: 160-169, 1975, 17. Moore, R. Y. Suprachiasmatic nucleus, secondary synchronizing stimuli and the central neural control of circadian rhythms. Brain Res 183: 13-28, 1980. 18. Moreira, A. C. and D. T. Krieger. The effects of subdiaphragmatic vagotomy on circadian corticosterone rhythmicity in rats with continuous or restricted food access. Physiol Behav 28: 787-790, 1982. 19. Wilkinson, C. W., J. Shinsako and M. F. Dallman. Rapid decreases in adrenal and plasma corticosterone concentrations after drinking are not mediated by changes in plasma adrenocorticotropin concentration. Endocrinology 110: 15991606, 1982.