- Email: [email protected]
Insulin counteracts the satiating effect of a fat meal in rats
Physiology & Behavior, Vol. 40, pp. 655-659. Copyright©PergamonJournals Ltd., 1987. Printed in the U.S.A.
0031-9384/87 $3.00 + .00
Insulin Counteracts the Satiating Effect of a Fat Meal in Rats M A R K I. F R I E D M A N
AND ISRAEL RAMIREZ
M o n e l l C h e m i c a l S e n s e s C e n t e r , 3500 M a r k e t S t r e e t P h i l a d e l p h i a , P A 19104 R e c e i v e d 16 M a r c h 1987 FRIEDMAN, M. I. AND I. RAMIREZ. Insulin counteracts the satiating effect of a fat meal in rats. PHYSIOL BEHAV 40(5) 655-659, 1987.--The effects of insulin administration of the reduction on food intake which follows a meal of corn oil was examined in normal and streptozotocin-diabetic rats. In the fwst experiment, a single injection of long-acting, protamine zinc insulin (3 IU) curtailed the decrease in 24-hr food intake that occurred in normal and diabetic rats after ingestion of 2.0 ml of oil. In a second experiment, injection of short-acting, regular insulin (0.5 IU) prevented the depression of food intake which occurred 6--24 hr after ingestion of 1.5 ml of corn oil, but not at earlier time intervals. In a third experiment, the short-term suppression of food intake in diabetic rats that occurred within 6 hr after a 1.5 ml meal ofoil was reduced by chronic administration of insulin (3 IU/day) via a subcutaneously implanted osmotic pump. The results indicate that a relatively long-lasting effect of insulin counteracts the satiation from ingested fat and suggest that insulin's role in the control of food intake may depend on the composition of the diet. Food intake
Streptozotocin-diabetes
Appetite
Corn oil
PREVIOUS studies have shown that diabetic rats are more sensitive than normal animals to the satiating effects of fat. Diabetic rats decrease caloric intake from hyperphagic levels when fat is added to their diet, whereas normal animals show no change in caloric intake or may even increase it [3, 5, 9]. Similarily, intake of a low fat chow diet is reduced after a single fat meal in diabetic rats to a greater extent than it is in normal rats [6,15]. We have suggested that diabetic rats show an exaggerated response to fat feeding because their insulin deficiency increases the oxidation o f the ingested fat [5, 6, 9, 11, 15]. The importance of reduced insulin levels in the response of diabetic rats to ingested fat is suggested by studies showing that insulin replacement treatment increases food intake in diabetic rats eating a high-fat diet [4,7]. This response to insulin appears to be specific to the fat component of the diet because when diabetic rats are fed a low-fat diet, insulin treatment has the opposite effect; that is, it decreases food intake (e.g., [4,7]). However, because fat was part of the diet in these experiments, it was not clear whether insulin treatment increased food intake by spurting the intake of fat or by counteracting a suppressive effect of fat on feeding. In the present study we sought to determine more directly whether insulin treatment reverses the satiating effects of fat ingestion. To do so, we used a procedure described previously in which rats are fed a meal of fat and its effect on subsequent chow is monitored. Using this method, it is possible to demonstrate the satiating effect of fat more clearly because the ingestion o f fat is separated in time from its suppressive effect on food intake. In the experiments described below, we examined whether insulin treatment would prevent this decrease in food intake after a fat meal. The results showed that insulin administration can court-
teract the satiating effects of fat without otherwise altering food intake and that this effect of the hormone can be demonstrated in both diabetic and normal rats. GENERAL METHODS Subjects Male CD rats (Charles River; Wilmington, MA) weighing 250-300 g at the start of the experiments were used. Rats were housed individually in standard hanging cages in a room maintained at approximately 22°C with a 12/12 hr day/night lighting schedule. In experiments using a reversed day/night schedule, lights were off at 0900 hr and rats were adapted to the schedule for at least three weeks prior to testing. Purina Laboratory Chow and tap water were available ad lib except when noted otherwise. Experimental Diabetes Rats were randomly assigned to diabetic or normal groups. Diabetes was induced by subcutaneous injection of streptozotocin (50 or 60 mg/kg, in a p H 4.8 citrate buffer). Only rats showing plasma glucose levels greater than 22 mM three days after injection were used. Normal rats were left untreated. Testing began 2--4 weeks after induction of diabetes. Procedures Prior to testing, rats were trained to consume corn oil (1.5 or 2.0 ml, depending on the experiment) which was placed in an inverted cap from a 20 ml scintillation vial attached to the bottom of their cage. Rats typically learned to consume the oil meal voluntarily without any other experimental manipu-
655
656
FRIEDMAN AND RAMIREZ TABLE 1
+2
F O O D I N T A K E (g/24 hr) A F T E R A F A T M E A L A N D I N J E C T I O N O F P R O T A M I N E Z I N C I N S U L I N (3.0 I . U . ) IN N O R M A L A N D DIABETIC RATS
Group
Control
Oil
Normal (N = 11) Saline Insulin
23.3 --- 0.7 25.9 _+ 0.8
20.4 _+ 0.5 24.6 +_ 0.5
Diabetic (N=9) Saline Insulin
39.6 +_ 1.0 42.6 -+ 1.7
34.6 + 1.3 40.5 _+ 1.6
kC-
Normal
"¢1"
"-. E
•-=
-2
o
No corn oil was fed in Control condition. Values are means -+ s.e.m. O) ¢-
Statistical Analyses
::::::::::::::::::' ............ ............
Method During the later part of the daytime period, 9 diabetic and 11 normal rats were injected subcutaneously with either normal saline or 3 IU of protamine zinc insulin (PZI) and then given either 2 ml of corn oil or no oil to drink. Each rat was given each combination of saline or PZI and oil or no oil (control) on different test days in random order. No more than two tests per week were conducted with at least two days between each test, and no more than one PZI injection was given per week. After each rat received each treatment, the entire sequence was repeated, resulting in two tests per condition per rat. Food intake was measured for the 24 hr after injection.
Results Diabetic rats ate more food than normal animals, F(1,18)= 147, p<0.001. Consistent with our previous studies [6,15], rats decreased 24 hr food intake after ingesting the oil meal, see Table 1; F(1,18)=121, p<0.001, and diabetic rats decreased intake more than normal animals, F(1,18)=8.3, p<0.01. Insulin treatment increased food intake by 11-16% F(1,18)=67, p<0.001 and this effect was similar in normal and diabetic rats. However, the effect of insulin on food intake depended on whether rats had consumed the fat meal, F(1,18)=9.9, p<0.01 as rats decreased food intake less after ingesting oil when they were given insulin instead of saline (Fig. 1). The effect of insulin on the satiating effect of fat ingestion was similar in normal and diabetic rats, F(1,18)=l.1, p>0.2. Caloric intakes (assuming chow=3.3 kcal/g) decreased after oil feeding (18.2 kcal) in normal and diabetic rats on average by 9.6 and 16.5 kcal without insulin treatment and 4.6 and 6.9 kcal with insulin.
,
iiiii!iiiii!iiiii!i --4
1
-6
FIG. 1. Change in 24-hr food intake from control (no oil) levels in normal and diabetic rats after a 2.0-ml meal of corn oil and subcutaneous injection of either saline (light bars) or 3.0 IU protamine zinc insulin (dark bars). Values are means_+s.e.m. *p<0.05; **p<0.01.
Data were evaluated by analysis of variance or Student's t-test. Values given are m e a n s - s . e . m . EXPERIMENT 1
Diabetic
0
c-
lation (e.g., deprivation) by the third time it was fed, eating it rapidly upon presentation and finishing it within 20 rain. The oil meal was fed no more than 2-3 times per week during training and testing with at least one day in which no oil was fed between feedings. Food intake, corrected for spillage, was measured to the nearest 0.1 g.
~-
EXPERIMENT 2
The first experiment showed that injection of long-acting insulin reduced the satiating effect of a fat meal over a subsequent 24 hr period. However, from this experiment it was not clear when this effect occurred relative to oil ingestion and hormone treatment because food intakes were measured only after 24 hr and long-acting insulin was used. In addition, although the data analysis indicated that insulin treatment counteracted the response to the fat meal independently from its effect of stimulating feeding, a more convincing demonstration of a specific action of insulin on the satiating effect of fat would require that the increase in food intake after insulin treatment be experimentally dissociated from a reversal of the suppressing effect of fat feeding. To address these issues, another experiment was performed in which rats were given short-acting insulin along with a fat meal and subsequent food intake was measured periodically for the next 24 hr. A dose of short-acting insulin was used which, on the basis of pilot studies, was found to have no appreciable effect on food intake.
Method Normal (N= 12) and diabetic (N= 10) rats that were maintained on a reversed day/night lighting schedule were trained to consume 1.5 ml of corn oil after 1 hr of food deprivation starting at lights out. After training on different days, rats were injected subcutaneously with either saline or 0.5 IU of regular insulin (Actrapid; Squibb) and given either oil or no oil (control) to drink. Each rat was tested once under each of the four conditions according to a partially counterbalanced design. Food was returned 30 rain after rats were fed the oil meal and food intakes were measured 2, 4, 6 and 24 hr later.
Results Diabetic rats ate more than normal animals at all time
I N S U L I N AND SATIETY
657 TABLE 2
TIME COURSE OF FOOD INTAKE (g) AFTER A FAT MEAL AND INJECTION OF REGULAR INSULIN (0.5 I.U.) IN NORMAL AND DIABETIC RATS
Hours Group
0-2
Normal (N= 12) Saline-Control Saline-Oil Insulin-Control Insulin-Oil
8.4 7.8 8.5 8.5
Diabetic (N = 10) Saline-Control Saline-Oil Insulin-Control Insulin-Oil
10.2 ± 8.2 ± 11.2 ± 10.0 ±
2-4
___0.6 ± 0.5 ± 0.5 ± 0.3 0.6 0.7 0.7 0.7
4--6
6-24
3.0 2.2 2.8 2.9
± ± ±
0.4 0.3 0.4 0.5
2.7 4.0 3.9 3.1
± ± ± ±
0.5 0.5 0.3 0.5
17.1 ± 14.9 ± 16.0 ± 15.9 ±
1.0 0.9 0.8 0.6
7.2 5.2 6.8 5.1
± ± ± ±
0.6 0.6 0.6 0.6
6.4 3.7 6.6 4.2
± ± ± ±
0.6 0.8 0.5 0.7
30.0 23.5 27.9 26.9
1.5 1.3 2.0 1.7
± ± ± ±
No corn oil was fed under Control condition. Values are means ± s.e.m.
+2
Normal
0-6 hr
6-24 hr
E ~-2
Diabetic
0-6 hr
6-24 hr
:~:!:~:i:i:i:i:i:;
~-4 !:~:!:i:i:i:i:!:!:
e--
EXPERIMENT 3
g-6 cO
-8
and F(1,20)=7.7, p<0.05, respectively, suggesting a difference in the short- but not long-term response to the satiating effect of fat in these two groups. Insulin treatment increased food intake overall only in the 0-2 hr period, F(1,20)=4.4, p < 0 . 0 5 , but this increase was transient and total (cumulative) food intake was not affected by insulin injection beyond 4 hr. Insulin injection did however significantly reduce the suppression of eating after fat ingestion (Fig. 2), but only in the 6-24 hr interval, F(1,20)=5.5, p<0.05. Insulin administration had no differential effect on the response to fat ingestion in normal and diabetic rats at this later time period, F(1,20) = 1.1, p >0.2. Caloric intakes in the 6--24 hr period were decreased after oil feeding (13.7 kcal) in normal and diabetic rats on average by 7.3 and 21.5 kcal without insulin treatment and 0.3 and 3.3 kcal with insulin.
iiii!iiiii Saline BII Insulin (0.5 I.U.)
FIG. 2. Change in food intake from control (no oil) levels in normal and diabetic rats either 0--6 or 6-24 hr after a 1.5-ml meal of corn oil and subcutaneous injection of either saline (light bars) or 0.5 IU regular insulin (dark bars). Values are means-s.e.m. intervals (ps<0.001). Analysis of total (i.e., cumulative) food intakes (data not shown) showed that oil ingestion significantly reduced eating at each time point measured (p's<0.05, 0.001, 0.001, and 0.001 for 2, 4, 6, and 24 hr respectively) and that this effect was significantly larger in diabetic relative to normal rats 4, 6, and 24 hr after the fat meal (p's<0.05, 0.01, 0.05, respectively). Examination of the time course of food intake (Table 2) revealed that this differential effect in normal and diabetic rats occurred in the 2-4 and 4--6 hr intervals after the fat meal, F(1,20)=6.2, p<0.05
The previous experiment indicated that relatively longterm consequences of insulin treatment are involved in counteracting the satiating effects of a fat meal; the acute (i.e., 0-6 hr) response to fat was unaffected by short-acting insulin. In this experiment we sought to determine whether the short-term suppression of food intake after a fat meal would be reversed in rats undergoing chronic insulin treatment.
Method Normal (N=7) and diabetic (N=7) rats maintained on a reversed light schedule were trained to consume a 1.5 ml oil meal as described in the previous experiment. After training, rats were tested under three conditions according to an ABA design. First, in order to establish a pre-insulin baseline, rats were given in two repeated tests either 1.5 ml of oil or no oil (control) to drink on four different days and food intakes were measured 6 hr later. Then all rats were lightly anesthetized with ether and implanted subcutaneously with 7-day Alzet osmotic rninipumps loaded to deliver 3 IU/day of regular insulin dissolved (at 37°C) in a 0.15 M NaCl saline vehicle containing 1.6% glycerol, 0.7% glutamic acid, 0.5% aspartic acid and 0.2% phenol [2]. Starting the day after minipumps were implanted, feeding tests with and without
658
F R I E D M A N AND R A M I R E Z
Insulin
Pre-insulin
Post-insulin
25
i i i i ti !i i~
F
.......l .......
:,:.:.:,:.:,:.::, :,:+:+::.:.:
20
::::.:::.::::::::: ., .........+.
,,.
!ii !i i!i i i!i l.
.C
E 15 ¢-
0 O L(.
10
Normal
Diabetic
Normal
Diabetic
Normal
Diabetic
FIG. 3. Food intake after a 1.5-ml meal of corn oil (dark bars) in normal and diabetic rat before, during and after chronic insulin administration (3.0 IU/day) via a subcutaneously implanted osmotic minipump. Light bars show food intake during a comparable 6 hr period when no oil was consumed (control). Values are means-s.e.m. *p<0.05; **p<0.01.
oil meals were repeated twice each during the next four days. Finally, to determine the post-insulin response, rats were tested again two times each with and without oil starting 8 days after pumps were implanted. The order of testing with and without oil feeding was counterbalanced in each phase of the experiment. As analysis revealed no effect of order of testing, results from repeated tests during each phase were averaged for each rat. Results As shown in Fig. 3, diabetic rats consumed more food than normal animals in all three phases of the experiment (p's<0.001). In the pre-insulin tests, oil ingestion decreased food intake, F(1,12)=22.7, p<0.001 and did so to a greater extent in diabetic as compared to normal rats, F(1,12) = 12.1, p <0.005. When rats were given insulin, the satiating effect of the fat meal was greatly reduced and oil ingestion no longer reliably decreased food intake across groups or differentially in diabetic rats. In the post-insulin tests, rats again decreased food intake after oil ingestion, F(1,12)=22.1, p<0.001; however, at this time there was no differential effect of the fat meal in normal and diabetic rats, F(1,12)=0.49, p > 0 . 2 . Caloric intakes during the feeding tests were decreased after oil feeding (13.7 kcal) in diabetic rats before, during, and after insulin treatment by 13.5, 6.6 and 13.9 kcal. In normal rats, caloric intakes were decreased after oil feeding in the post-insulin period by 10.2 kcal. DISCUSSION
The results of these experiments show that insulin administration can counteract the satiating effects of a fat meal. This effect did not appear to be dependent on the type of insulin used or its schedule of administration as both longand short-acting insulin given according to three different
regimes produced essentially the same result. Insulin treatment reversed the suppression of food intake by fat feeding without necessarily affecting food intake under control conditions when no fat was consumed. This indicates that the hormone's effect was relatively specific to the satiation produced by the fat meal and points to a mechanism of action based on the metabolic effect of insulin. The present study does not specify the mechanism(s) responsible for the effect of insulin administration reported here. The results do suggest however that the acute effects of the hormone, such a hypoglycemia and increased gastric emptying, are not involved as short-acting insulin did not affect the short-term response to a fat meal. Rather, the data implicate a slowly developing change in metabolism produced by the hormone. The reversal of the satiating effect of the fat meal was observed over a 24-hr period atter injection of long-acting PZI insulin, with a delay of at least 6 hr after administration of short-acting insulin, and in the short-term only against a background of chronic insulin exposure. Insulin is known to induce activity in a number of enzymes, a process which may take several hours, and this may underlie the delayed effect of the hormone shown here. One such enzyme that could be involved is adipose tissue lipoprotein lipase (LPL) as insulin increases L P L activity only after a delay of some hours [10,16]. Evidence suggests that the reduction of food intake after a fat meal is due to the oxidation of the ingested fat [6,15]. Because an increase in adipose LPL activity would foster the uptake of circulating lipids into storage, it is possible that insulin treatment may reduce the satiating effect of fat by redirecting the flux of fat away from oxidative pathways [8,11]. In this regard, it is interesting that normal rats in Experiment 3 became more responsive to the satiating effect of the fat meal after insulin treatment was discontinued because termination of insulin administration has been shown to increase fat oxidation [8,14].
I N S U L I N AND SATIETY
659
Insulin treatment reduced the satiating effect of fat in both diabetic and normal rats, suggesting that this action of the hormone is not a function of diabetic pathology. The effect in normal animals was observed in the first two experiments in which they showed a suppression of food intake after fat ingestion. In the third experiment, fat ingestion did not decrease food intake of normal rats before insulin administration and hormone treatment h a d n o effect on their food intake or response to the fat meal. These observations are consistent with the conclusion made above that insulin is acting specifically to counter the satiating effects of fat ingestion. The present findings are also in keeping with results of previous studies showing that a fat meal has a smaller and less consistent effect on food intake in normal as opposed to diabetic rats [6,15]. It is possible that these inconsistent effects of fat feeding in normal rats are due to their comparatively higher insulin levels and that they do not respond to the fat meal (or to insulin administration) because endogenous insulin already prevents fat from exerting a satiating effect. Endogenous insulin is thought to promote satiety or facilitate the termination of feeding under normal eating conditions (e.g., [1, 12, 13, 18, 19]). The present results suggest that insulin's role in the control of food intake may
depend on what nutrients are consumed. The finding that insulin treatment counteracts the satiating effect of fat without otherwise changing food intake suggests that elevated insulin levels which by themselves do not cause hyperphagia may predispose rats to overeat if the diet is high in fat, This could result because of a primary metabolic shift from oxidation to storage of the ingested fat [8,11]. Such a mechanism would be consistent with the observation that elevated insulin levels increase fat deposition even in the absence of hyperphagia [17]. In a similar vein, high-fat, highcarbohydrate diets may induce overeating because the insulin secreted in response to the carbohydrate portion of the diet counteracts the satiating effect of the dietary fat by directing it into storage and away from oxidative pathways. The overeating of such high-fat, high-carbohydrate food thus may reflect not only the increased palatability of the diet, but also the metabolic fate of the ingested nutrients. ACKNOWLEDGEMENTS This research was supported by NIH grant DK 35014. The authors thank Musa Speranza for her technical assistance, and Michael Tordoff and Rhonda Deems for their helpful comments on the manuscript.
REFERENCES 1. Anika, S. M., T. R. Houpt and K. A. Houpt. Insulin as a satiety hormone. Physiol Behav 25: 21-23, 1980. 2. Bringer, J., T. Held and G. M. Grodsky. Prevention of insulin aggregation by dicarboxylic amino acids during prolonged infusion. Diabetes 30: 83-85, 1981. 3. Edens, N. K. and M. I. Friedman. Response of normal and diabetic rats to increasing dietary medium-chain triglyceride content. J Nutr 114: 565--573, 1984. 4. Friedman, M. I. Insulin-inducedhyperphagia in alloxan-diabetic rats fed a high-fat diet. Physiol Behav 19: 597-599, 1977. 5. Friedman, M. I. Hyperphagia in rats with experimental diabetes mellitus: a response to a decreased supply of utilizable fuels. J Comp Physiol Psychol 92:109--117, 1978. 6. Friedman, M. I., N. K. Edens and I. Ramirez. Differential effects of medium- and long-chain triglycerides on food intake of normal and diabetic rats. Physiol Behav 31: 851-855, 1983. 7. Friedman, M. I., A. L. Emmerich and K. M. Gil. Effects of insulin on food intake and plasma glucose level in fat-fed diabetic rats. Physiol Behav 24: 319-325, 1980. 8. Friedman, M. I. and I. Ramirez. Relationship of fat metabolism to food intake. Am J Clin Nutr 42: 1093--1098, 1985. 9. Friedman, M. I., I. Ramirez, N. K. Edens and J. Granneman. Food intake in diabetic rats: isolation of primary effects of fat feeding. Am J Physiol 249: R44-R51, 1985. 10. Friedman, M. I., I. Ramirez, G. N. Wade, L. I. Siegel and J. Granneman. Metabolic and physiologic effects of a hungerinducing injection of insulin. Physiol Behav 29: 515--518, 1982. 11. Friedman, M. I. and E. M. Stricker. The physiological psychology of hunger: A physiological perspective. Psychol Rev 83: 41)9-431, 1976.
12. Lindberg, N. O., P. C. Coburn and E. M. Stricker. Increased feeding by rats after subdiabetogenic streptozotocin treatment: a role for insulin in satiety. Behav Neurosci 98: 138--145, 1984. 13. Oetting, R. L. and D. A. VanderWeele. Insufin suppresses intake without inducing illness in sham feeding rats. Physiol Behav 34: 557-562, 1985. 14. Ramirez, I. and M. I. Friedman. Metabolic concomitants of hypophagla during recovery from insulin-induced obesity in rats. Am J Physiol 245: E211-E219, 1983. 15. Ramirez, I. and M. I. Friedman. Food intake and blood fuels after oil consumption: Differential effects in normal and diabetic rats. Physiol Behav 31: 847--850, 1983. 16. Salaman, M. R. and D. S. Robinson. Clearing factor lipase in adipose tissue. A medium in which the enzyme activity of tissues from starved rats increases in vitro. Biochem J 99: 640-647, 1966. 17. Torbay, N., E. F. Bracco, A. Geliebter, I. M. Stewart and S. A. Hashim. Insulin increases body fat despite control of food intake and physical activity. Am J Physiol 245: R120--R124, 1985. 18. VanderWeele, D. A., F. X. Pi-Sunyer, D. Novin and M. J. Bush. Chronic insulin infusion suppresses food ingestion and body weight gain in rats. Brain Res Bull 5: Suppl 4, 7-11, 1980. 19. Woods, S. C., D. Porte, Jr., E. Bobbioni, E. Ionescu, J.-F. Sauter, F. Rohner-Jeanrenaud and B. Jeanrenaud. Insulin: its relationship to the central nervous system and to the control of food intake and body weight. Am J Clin Nutr 42: 1063--1071, 1985.
0031-9384/87 $3.00 + .00
Insulin Counteracts the Satiating Effect of a Fat Meal in Rats M A R K I. F R I E D M A N
AND ISRAEL RAMIREZ
M o n e l l C h e m i c a l S e n s e s C e n t e r , 3500 M a r k e t S t r e e t P h i l a d e l p h i a , P A 19104 R e c e i v e d 16 M a r c h 1987 FRIEDMAN, M. I. AND I. RAMIREZ. Insulin counteracts the satiating effect of a fat meal in rats. PHYSIOL BEHAV 40(5) 655-659, 1987.--The effects of insulin administration of the reduction on food intake which follows a meal of corn oil was examined in normal and streptozotocin-diabetic rats. In the fwst experiment, a single injection of long-acting, protamine zinc insulin (3 IU) curtailed the decrease in 24-hr food intake that occurred in normal and diabetic rats after ingestion of 2.0 ml of oil. In a second experiment, injection of short-acting, regular insulin (0.5 IU) prevented the depression of food intake which occurred 6--24 hr after ingestion of 1.5 ml of corn oil, but not at earlier time intervals. In a third experiment, the short-term suppression of food intake in diabetic rats that occurred within 6 hr after a 1.5 ml meal ofoil was reduced by chronic administration of insulin (3 IU/day) via a subcutaneously implanted osmotic pump. The results indicate that a relatively long-lasting effect of insulin counteracts the satiation from ingested fat and suggest that insulin's role in the control of food intake may depend on the composition of the diet. Food intake
Streptozotocin-diabetes
Appetite
Corn oil
PREVIOUS studies have shown that diabetic rats are more sensitive than normal animals to the satiating effects of fat. Diabetic rats decrease caloric intake from hyperphagic levels when fat is added to their diet, whereas normal animals show no change in caloric intake or may even increase it [3, 5, 9]. Similarily, intake of a low fat chow diet is reduced after a single fat meal in diabetic rats to a greater extent than it is in normal rats [6,15]. We have suggested that diabetic rats show an exaggerated response to fat feeding because their insulin deficiency increases the oxidation o f the ingested fat [5, 6, 9, 11, 15]. The importance of reduced insulin levels in the response of diabetic rats to ingested fat is suggested by studies showing that insulin replacement treatment increases food intake in diabetic rats eating a high-fat diet [4,7]. This response to insulin appears to be specific to the fat component of the diet because when diabetic rats are fed a low-fat diet, insulin treatment has the opposite effect; that is, it decreases food intake (e.g., [4,7]). However, because fat was part of the diet in these experiments, it was not clear whether insulin treatment increased food intake by spurting the intake of fat or by counteracting a suppressive effect of fat on feeding. In the present study we sought to determine more directly whether insulin treatment reverses the satiating effects of fat ingestion. To do so, we used a procedure described previously in which rats are fed a meal of fat and its effect on subsequent chow is monitored. Using this method, it is possible to demonstrate the satiating effect of fat more clearly because the ingestion o f fat is separated in time from its suppressive effect on food intake. In the experiments described below, we examined whether insulin treatment would prevent this decrease in food intake after a fat meal. The results showed that insulin administration can court-
teract the satiating effects of fat without otherwise altering food intake and that this effect of the hormone can be demonstrated in both diabetic and normal rats. GENERAL METHODS Subjects Male CD rats (Charles River; Wilmington, MA) weighing 250-300 g at the start of the experiments were used. Rats were housed individually in standard hanging cages in a room maintained at approximately 22°C with a 12/12 hr day/night lighting schedule. In experiments using a reversed day/night schedule, lights were off at 0900 hr and rats were adapted to the schedule for at least three weeks prior to testing. Purina Laboratory Chow and tap water were available ad lib except when noted otherwise. Experimental Diabetes Rats were randomly assigned to diabetic or normal groups. Diabetes was induced by subcutaneous injection of streptozotocin (50 or 60 mg/kg, in a p H 4.8 citrate buffer). Only rats showing plasma glucose levels greater than 22 mM three days after injection were used. Normal rats were left untreated. Testing began 2--4 weeks after induction of diabetes. Procedures Prior to testing, rats were trained to consume corn oil (1.5 or 2.0 ml, depending on the experiment) which was placed in an inverted cap from a 20 ml scintillation vial attached to the bottom of their cage. Rats typically learned to consume the oil meal voluntarily without any other experimental manipu-
655
656
FRIEDMAN AND RAMIREZ TABLE 1
+2
F O O D I N T A K E (g/24 hr) A F T E R A F A T M E A L A N D I N J E C T I O N O F P R O T A M I N E Z I N C I N S U L I N (3.0 I . U . ) IN N O R M A L A N D DIABETIC RATS
Group
Control
Oil
Normal (N = 11) Saline Insulin
23.3 --- 0.7 25.9 _+ 0.8
20.4 _+ 0.5 24.6 +_ 0.5
Diabetic (N=9) Saline Insulin
39.6 +_ 1.0 42.6 -+ 1.7
34.6 + 1.3 40.5 _+ 1.6
kC-
Normal
"¢1"
"-. E
•-=
-2
o
No corn oil was fed in Control condition. Values are means -+ s.e.m. O) ¢-
Statistical Analyses
::::::::::::::::::' ............ ............
Method During the later part of the daytime period, 9 diabetic and 11 normal rats were injected subcutaneously with either normal saline or 3 IU of protamine zinc insulin (PZI) and then given either 2 ml of corn oil or no oil to drink. Each rat was given each combination of saline or PZI and oil or no oil (control) on different test days in random order. No more than two tests per week were conducted with at least two days between each test, and no more than one PZI injection was given per week. After each rat received each treatment, the entire sequence was repeated, resulting in two tests per condition per rat. Food intake was measured for the 24 hr after injection.
Results Diabetic rats ate more food than normal animals, F(1,18)= 147, p<0.001. Consistent with our previous studies [6,15], rats decreased 24 hr food intake after ingesting the oil meal, see Table 1; F(1,18)=121, p<0.001, and diabetic rats decreased intake more than normal animals, F(1,18)=8.3, p<0.01. Insulin treatment increased food intake by 11-16% F(1,18)=67, p<0.001 and this effect was similar in normal and diabetic rats. However, the effect of insulin on food intake depended on whether rats had consumed the fat meal, F(1,18)=9.9, p<0.01 as rats decreased food intake less after ingesting oil when they were given insulin instead of saline (Fig. 1). The effect of insulin on the satiating effect of fat ingestion was similar in normal and diabetic rats, F(1,18)=l.1, p>0.2. Caloric intakes (assuming chow=3.3 kcal/g) decreased after oil feeding (18.2 kcal) in normal and diabetic rats on average by 9.6 and 16.5 kcal without insulin treatment and 4.6 and 6.9 kcal with insulin.
,
iiiii!iiiii!iiiii!i --4
1
-6
FIG. 1. Change in 24-hr food intake from control (no oil) levels in normal and diabetic rats after a 2.0-ml meal of corn oil and subcutaneous injection of either saline (light bars) or 3.0 IU protamine zinc insulin (dark bars). Values are means_+s.e.m. *p<0.05; **p<0.01.
Data were evaluated by analysis of variance or Student's t-test. Values given are m e a n s - s . e . m . EXPERIMENT 1
Diabetic
0
c-
lation (e.g., deprivation) by the third time it was fed, eating it rapidly upon presentation and finishing it within 20 rain. The oil meal was fed no more than 2-3 times per week during training and testing with at least one day in which no oil was fed between feedings. Food intake, corrected for spillage, was measured to the nearest 0.1 g.
~-
EXPERIMENT 2
The first experiment showed that injection of long-acting insulin reduced the satiating effect of a fat meal over a subsequent 24 hr period. However, from this experiment it was not clear when this effect occurred relative to oil ingestion and hormone treatment because food intakes were measured only after 24 hr and long-acting insulin was used. In addition, although the data analysis indicated that insulin treatment counteracted the response to the fat meal independently from its effect of stimulating feeding, a more convincing demonstration of a specific action of insulin on the satiating effect of fat would require that the increase in food intake after insulin treatment be experimentally dissociated from a reversal of the suppressing effect of fat feeding. To address these issues, another experiment was performed in which rats were given short-acting insulin along with a fat meal and subsequent food intake was measured periodically for the next 24 hr. A dose of short-acting insulin was used which, on the basis of pilot studies, was found to have no appreciable effect on food intake.
Method Normal (N= 12) and diabetic (N= 10) rats that were maintained on a reversed day/night lighting schedule were trained to consume 1.5 ml of corn oil after 1 hr of food deprivation starting at lights out. After training on different days, rats were injected subcutaneously with either saline or 0.5 IU of regular insulin (Actrapid; Squibb) and given either oil or no oil (control) to drink. Each rat was tested once under each of the four conditions according to a partially counterbalanced design. Food was returned 30 rain after rats were fed the oil meal and food intakes were measured 2, 4, 6 and 24 hr later.
Results Diabetic rats ate more than normal animals at all time
I N S U L I N AND SATIETY
657 TABLE 2
TIME COURSE OF FOOD INTAKE (g) AFTER A FAT MEAL AND INJECTION OF REGULAR INSULIN (0.5 I.U.) IN NORMAL AND DIABETIC RATS
Hours Group
0-2
Normal (N= 12) Saline-Control Saline-Oil Insulin-Control Insulin-Oil
8.4 7.8 8.5 8.5
Diabetic (N = 10) Saline-Control Saline-Oil Insulin-Control Insulin-Oil
10.2 ± 8.2 ± 11.2 ± 10.0 ±
2-4
___0.6 ± 0.5 ± 0.5 ± 0.3 0.6 0.7 0.7 0.7
4--6
6-24
3.0 2.2 2.8 2.9
± ± ±
0.4 0.3 0.4 0.5
2.7 4.0 3.9 3.1
± ± ± ±
0.5 0.5 0.3 0.5
17.1 ± 14.9 ± 16.0 ± 15.9 ±
1.0 0.9 0.8 0.6
7.2 5.2 6.8 5.1
± ± ± ±
0.6 0.6 0.6 0.6
6.4 3.7 6.6 4.2
± ± ± ±
0.6 0.8 0.5 0.7
30.0 23.5 27.9 26.9
1.5 1.3 2.0 1.7
± ± ± ±
No corn oil was fed under Control condition. Values are means ± s.e.m.
+2
Normal
0-6 hr
6-24 hr
E ~-2
Diabetic
0-6 hr
6-24 hr
:~:!:~:i:i:i:i:i:;
~-4 !:~:!:i:i:i:i:!:!:
e--
EXPERIMENT 3
g-6 cO
-8
and F(1,20)=7.7, p<0.05, respectively, suggesting a difference in the short- but not long-term response to the satiating effect of fat in these two groups. Insulin treatment increased food intake overall only in the 0-2 hr period, F(1,20)=4.4, p < 0 . 0 5 , but this increase was transient and total (cumulative) food intake was not affected by insulin injection beyond 4 hr. Insulin injection did however significantly reduce the suppression of eating after fat ingestion (Fig. 2), but only in the 6-24 hr interval, F(1,20)=5.5, p<0.05. Insulin administration had no differential effect on the response to fat ingestion in normal and diabetic rats at this later time period, F(1,20) = 1.1, p >0.2. Caloric intakes in the 6--24 hr period were decreased after oil feeding (13.7 kcal) in normal and diabetic rats on average by 7.3 and 21.5 kcal without insulin treatment and 0.3 and 3.3 kcal with insulin.
iiii!iiiii Saline BII Insulin (0.5 I.U.)
FIG. 2. Change in food intake from control (no oil) levels in normal and diabetic rats either 0--6 or 6-24 hr after a 1.5-ml meal of corn oil and subcutaneous injection of either saline (light bars) or 0.5 IU regular insulin (dark bars). Values are means-s.e.m. intervals (ps<0.001). Analysis of total (i.e., cumulative) food intakes (data not shown) showed that oil ingestion significantly reduced eating at each time point measured (p's<0.05, 0.001, 0.001, and 0.001 for 2, 4, 6, and 24 hr respectively) and that this effect was significantly larger in diabetic relative to normal rats 4, 6, and 24 hr after the fat meal (p's<0.05, 0.01, 0.05, respectively). Examination of the time course of food intake (Table 2) revealed that this differential effect in normal and diabetic rats occurred in the 2-4 and 4--6 hr intervals after the fat meal, F(1,20)=6.2, p<0.05
The previous experiment indicated that relatively longterm consequences of insulin treatment are involved in counteracting the satiating effects of a fat meal; the acute (i.e., 0-6 hr) response to fat was unaffected by short-acting insulin. In this experiment we sought to determine whether the short-term suppression of food intake after a fat meal would be reversed in rats undergoing chronic insulin treatment.
Method Normal (N=7) and diabetic (N=7) rats maintained on a reversed light schedule were trained to consume a 1.5 ml oil meal as described in the previous experiment. After training, rats were tested under three conditions according to an ABA design. First, in order to establish a pre-insulin baseline, rats were given in two repeated tests either 1.5 ml of oil or no oil (control) to drink on four different days and food intakes were measured 6 hr later. Then all rats were lightly anesthetized with ether and implanted subcutaneously with 7-day Alzet osmotic rninipumps loaded to deliver 3 IU/day of regular insulin dissolved (at 37°C) in a 0.15 M NaCl saline vehicle containing 1.6% glycerol, 0.7% glutamic acid, 0.5% aspartic acid and 0.2% phenol [2]. Starting the day after minipumps were implanted, feeding tests with and without
658
F R I E D M A N AND R A M I R E Z
Insulin
Pre-insulin
Post-insulin
25
i i i i ti !i i~
F
.......l .......
:,:.:.:,:.:,:.::, :,:+:+::.:.:
20
::::.:::.::::::::: ., .........+.
,,.
!ii !i i!i i i!i l.
.C
E 15 ¢-
0 O L(.
10
Normal
Diabetic
Normal
Diabetic
Normal
Diabetic
FIG. 3. Food intake after a 1.5-ml meal of corn oil (dark bars) in normal and diabetic rat before, during and after chronic insulin administration (3.0 IU/day) via a subcutaneously implanted osmotic minipump. Light bars show food intake during a comparable 6 hr period when no oil was consumed (control). Values are means-s.e.m. *p<0.05; **p<0.01.
oil meals were repeated twice each during the next four days. Finally, to determine the post-insulin response, rats were tested again two times each with and without oil starting 8 days after pumps were implanted. The order of testing with and without oil feeding was counterbalanced in each phase of the experiment. As analysis revealed no effect of order of testing, results from repeated tests during each phase were averaged for each rat. Results As shown in Fig. 3, diabetic rats consumed more food than normal animals in all three phases of the experiment (p's<0.001). In the pre-insulin tests, oil ingestion decreased food intake, F(1,12)=22.7, p<0.001 and did so to a greater extent in diabetic as compared to normal rats, F(1,12) = 12.1, p <0.005. When rats were given insulin, the satiating effect of the fat meal was greatly reduced and oil ingestion no longer reliably decreased food intake across groups or differentially in diabetic rats. In the post-insulin tests, rats again decreased food intake after oil ingestion, F(1,12)=22.1, p<0.001; however, at this time there was no differential effect of the fat meal in normal and diabetic rats, F(1,12)=0.49, p > 0 . 2 . Caloric intakes during the feeding tests were decreased after oil feeding (13.7 kcal) in diabetic rats before, during, and after insulin treatment by 13.5, 6.6 and 13.9 kcal. In normal rats, caloric intakes were decreased after oil feeding in the post-insulin period by 10.2 kcal. DISCUSSION
The results of these experiments show that insulin administration can counteract the satiating effects of a fat meal. This effect did not appear to be dependent on the type of insulin used or its schedule of administration as both longand short-acting insulin given according to three different
regimes produced essentially the same result. Insulin treatment reversed the suppression of food intake by fat feeding without necessarily affecting food intake under control conditions when no fat was consumed. This indicates that the hormone's effect was relatively specific to the satiation produced by the fat meal and points to a mechanism of action based on the metabolic effect of insulin. The present study does not specify the mechanism(s) responsible for the effect of insulin administration reported here. The results do suggest however that the acute effects of the hormone, such a hypoglycemia and increased gastric emptying, are not involved as short-acting insulin did not affect the short-term response to a fat meal. Rather, the data implicate a slowly developing change in metabolism produced by the hormone. The reversal of the satiating effect of the fat meal was observed over a 24-hr period atter injection of long-acting PZI insulin, with a delay of at least 6 hr after administration of short-acting insulin, and in the short-term only against a background of chronic insulin exposure. Insulin is known to induce activity in a number of enzymes, a process which may take several hours, and this may underlie the delayed effect of the hormone shown here. One such enzyme that could be involved is adipose tissue lipoprotein lipase (LPL) as insulin increases L P L activity only after a delay of some hours [10,16]. Evidence suggests that the reduction of food intake after a fat meal is due to the oxidation of the ingested fat [6,15]. Because an increase in adipose LPL activity would foster the uptake of circulating lipids into storage, it is possible that insulin treatment may reduce the satiating effect of fat by redirecting the flux of fat away from oxidative pathways [8,11]. In this regard, it is interesting that normal rats in Experiment 3 became more responsive to the satiating effect of the fat meal after insulin treatment was discontinued because termination of insulin administration has been shown to increase fat oxidation [8,14].
I N S U L I N AND SATIETY
659
Insulin treatment reduced the satiating effect of fat in both diabetic and normal rats, suggesting that this action of the hormone is not a function of diabetic pathology. The effect in normal animals was observed in the first two experiments in which they showed a suppression of food intake after fat ingestion. In the third experiment, fat ingestion did not decrease food intake of normal rats before insulin administration and hormone treatment h a d n o effect on their food intake or response to the fat meal. These observations are consistent with the conclusion made above that insulin is acting specifically to counter the satiating effects of fat ingestion. The present findings are also in keeping with results of previous studies showing that a fat meal has a smaller and less consistent effect on food intake in normal as opposed to diabetic rats [6,15]. It is possible that these inconsistent effects of fat feeding in normal rats are due to their comparatively higher insulin levels and that they do not respond to the fat meal (or to insulin administration) because endogenous insulin already prevents fat from exerting a satiating effect. Endogenous insulin is thought to promote satiety or facilitate the termination of feeding under normal eating conditions (e.g., [1, 12, 13, 18, 19]). The present results suggest that insulin's role in the control of food intake may
depend on what nutrients are consumed. The finding that insulin treatment counteracts the satiating effect of fat without otherwise changing food intake suggests that elevated insulin levels which by themselves do not cause hyperphagia may predispose rats to overeat if the diet is high in fat, This could result because of a primary metabolic shift from oxidation to storage of the ingested fat [8,11]. Such a mechanism would be consistent with the observation that elevated insulin levels increase fat deposition even in the absence of hyperphagia [17]. In a similar vein, high-fat, highcarbohydrate diets may induce overeating because the insulin secreted in response to the carbohydrate portion of the diet counteracts the satiating effect of the dietary fat by directing it into storage and away from oxidative pathways. The overeating of such high-fat, high-carbohydrate food thus may reflect not only the increased palatability of the diet, but also the metabolic fate of the ingested nutrients. ACKNOWLEDGEMENTS This research was supported by NIH grant DK 35014. The authors thank Musa Speranza for her technical assistance, and Michael Tordoff and Rhonda Deems for their helpful comments on the manuscript.
REFERENCES 1. Anika, S. M., T. R. Houpt and K. A. Houpt. Insulin as a satiety hormone. Physiol Behav 25: 21-23, 1980. 2. Bringer, J., T. Held and G. M. Grodsky. Prevention of insulin aggregation by dicarboxylic amino acids during prolonged infusion. Diabetes 30: 83-85, 1981. 3. Edens, N. K. and M. I. Friedman. Response of normal and diabetic rats to increasing dietary medium-chain triglyceride content. J Nutr 114: 565--573, 1984. 4. Friedman, M. I. Insulin-inducedhyperphagia in alloxan-diabetic rats fed a high-fat diet. Physiol Behav 19: 597-599, 1977. 5. Friedman, M. I. Hyperphagia in rats with experimental diabetes mellitus: a response to a decreased supply of utilizable fuels. J Comp Physiol Psychol 92:109--117, 1978. 6. Friedman, M. I., N. K. Edens and I. Ramirez. Differential effects of medium- and long-chain triglycerides on food intake of normal and diabetic rats. Physiol Behav 31: 851-855, 1983. 7. Friedman, M. I., A. L. Emmerich and K. M. Gil. Effects of insulin on food intake and plasma glucose level in fat-fed diabetic rats. Physiol Behav 24: 319-325, 1980. 8. Friedman, M. I. and I. Ramirez. Relationship of fat metabolism to food intake. Am J Clin Nutr 42: 1093--1098, 1985. 9. Friedman, M. I., I. Ramirez, N. K. Edens and J. Granneman. Food intake in diabetic rats: isolation of primary effects of fat feeding. Am J Physiol 249: R44-R51, 1985. 10. Friedman, M. I., I. Ramirez, G. N. Wade, L. I. Siegel and J. Granneman. Metabolic and physiologic effects of a hungerinducing injection of insulin. Physiol Behav 29: 515--518, 1982. 11. Friedman, M. I. and E. M. Stricker. The physiological psychology of hunger: A physiological perspective. Psychol Rev 83: 41)9-431, 1976.
12. Lindberg, N. O., P. C. Coburn and E. M. Stricker. Increased feeding by rats after subdiabetogenic streptozotocin treatment: a role for insulin in satiety. Behav Neurosci 98: 138--145, 1984. 13. Oetting, R. L. and D. A. VanderWeele. Insufin suppresses intake without inducing illness in sham feeding rats. Physiol Behav 34: 557-562, 1985. 14. Ramirez, I. and M. I. Friedman. Metabolic concomitants of hypophagla during recovery from insulin-induced obesity in rats. Am J Physiol 245: E211-E219, 1983. 15. Ramirez, I. and M. I. Friedman. Food intake and blood fuels after oil consumption: Differential effects in normal and diabetic rats. Physiol Behav 31: 847--850, 1983. 16. Salaman, M. R. and D. S. Robinson. Clearing factor lipase in adipose tissue. A medium in which the enzyme activity of tissues from starved rats increases in vitro. Biochem J 99: 640-647, 1966. 17. Torbay, N., E. F. Bracco, A. Geliebter, I. M. Stewart and S. A. Hashim. Insulin increases body fat despite control of food intake and physical activity. Am J Physiol 245: R120--R124, 1985. 18. VanderWeele, D. A., F. X. Pi-Sunyer, D. Novin and M. J. Bush. Chronic insulin infusion suppresses food ingestion and body weight gain in rats. Brain Res Bull 5: Suppl 4, 7-11, 1980. 19. Woods, S. C., D. Porte, Jr., E. Bobbioni, E. Ionescu, J.-F. Sauter, F. Rohner-Jeanrenaud and B. Jeanrenaud. Insulin: its relationship to the central nervous system and to the control of food intake and body weight. Am J Clin Nutr 42: 1063--1071, 1985.