- Email: [email protected]
Sham-feeding of corn oil by rats: Sensory and postingestive factors
Physiology & Behavior, Vol. 47, pp. 77%781. © Pergamon Press plc, 1990, Printed in the U.S.A.
0031-9384/90 $3.00 + .00
BRIEF COMMUNICATION
Sham-Feeding of Corn Oil by Rats: Sensory and Postingestive Factors D A N I E L L E R. R E E D l
Yale University and Monell Chemical Senses Center M I C H A E L G . T O R D O F F A N D M A R K I. F R I E D M A N
Monell Chemical Senses Center R e c e i v e d 25 S e p t e m b e r 1989
REED, D. R., M. G. TORDOFF AND M. I. FRIEDMAN. Sham-feeding of corn oil by rats: Sensory and postingestive factors. PHYSIOL BEHAV 47(4) 779--781, 1990.--Previous research indicates that rats fed a high-fat (HF) diet increase their intake and preference for oil compared with rats fed a high-carbohydrate (HC) diet. To assess whether this increased intake was due to the sensory or postingestive properties of oil, rats were adapted to either the HF or HC diet and then allowed to sham-feed pure corn oil daily for 30 rain. During the first 4 trials, rats fed the HF diet sham-fed more oil than did rats fed the HC diet; however, this difference diminished with repeated testing and was absent after 8 trials. In both diet groups, 4-5 calories (approximately 25%) of sham-fed oil could not be recovered and may have escaped to the intestine. These results suggest that, compared with rats fed a HC diet, rats fed a HF diet are initially attracted to the sensory properties of oil, but that the differential oil intakes of rats fed the HF or HC diet are maintained by postingestive, rather than sensory factors. High-fat diet
Food preferences
Oil intake
Fat acceptance
METHOD
RATS fed a high-fat (HF) diet show an increased acceptance of oil compared with rats fed a high-carbohydrate (HC) diet (7,10). Potentially, diet composition might affect the rats' response to either the sensory or postingestive effects of oil and produce the change in oil intake. The sham-feeding preparation is one way to distinguish between these two possibilities. During sham-feeding, food drains out of the esophagus or stomach through a fistula and its postingestive effects are minimized (6, 9, 14). Earlier work shows that rats sham-feed corn oil and mineral oil (5,13), suggesting that the orosensory stimulation produced by ingesting oil is reinforcing. To test whether the sensory or postingestive effects of oil consumption are important in the maintenance of differential oil intake by rats fed different diets, we studied sham-feeding of oil by rats fed HF or HC diets. If the sensory stimulation associated with oil is critical for increased oil intake by rats fed a HF diet, then they should sham-feed more oil than rats fed the HC diet. If, on the other hand, the postingestive effect of oil is important then rats fed the HF and HC diet should sham-feed the same amount of oil.
Subjects Eighteen male Sprague-Dawley rats (Charles River, Wilmington, MA), initially weighing 200--225 g, were used as subjects. They were housed individually in wire mesh hanging cages except during the first I0 days after surgery, when they were housed in solid-bottom cages with wood shavings to prevent the newly implanted cannulas from being pulled out on the wire mesh. Rats were maintained on a 12:12 hr lighting cycle with lights off at 1330 hr. The room temperature was 12-23°C. The protocol was approved by the Animal Care and Use Committee.
Diet Composition, Food Intake and Body Weight For the duration of the experiment, rats were fed nutritionally complete, semisynthetic diets that were either high in fat (HF; N = 9) or carbohydrate content (HC; N = 9; Table 1). The diets were
IRequests for reprints should be addressed to Danielle R. Reed, Monell Chemical Senses Center, 3500 Market St., Philadelphia, PA 19104.
779
780
REED, TORDOFF AND FRIEDMAN
TABLE 1
TABLE 2
COMPOSITION OF DIETS
PERCENT OF INGESTED OIL RECOVERED AND NUMBER OF OIL CALORIES UNACCOUNTEDFOR AFTER SHAM-FEEDING
Ingredients (g/100 g)
High-Fat (HF)
High-Carbohydrate (HC)
Casein Cornstarch Corn Oil Alphacel AIN Mineral Mix AIN Vitamin Mix Choline Bitartrate DL Methionine
20.0 10.4 23.1 41.5 3.5 1.0 0.2 0.3
20.0 52.0 4.6 18.4 3.5 1.0 0.2 0.3
kcal/g % kcal fat
3.3 63.0
3.3 13.0
Diets were prepared and pelleted by ICN Nutritional Biochemicals (Cleveland, OH).
provided in glass jars attached to the rats' cages by springs. Food intake was measured to the nearest 0.1 g (corrected for spillage) daily for the first 17 days of sham-feeding. Tap water was freely available except during the sham-feeding trials. Body weights were measured on the day before the first sham-feeding trial.
% Recovered kcal
High-Fat
High-Carbohydrate
76 _ 5 5.0 --+ 0.3
80 - 5 4.0 _+ 0.2
Differences between diet groups were not statistically significant. Values are means _ SEM.
Data Analysis The difference between the two dietary groups in body weight was analyzed using a t-test. Food intake during the sham-feeding period was analyzed using a mixed design ANOVA with Diet as the between-group factor and Trial (1-17) and Test Condition (Oil vs. No Oil) as the within-group factors. Oil intake during the sham-feeding period was analyzed using a mixed design ANOVA with Diet as the between-group factor and Trial (1-17) as the within-group factor. Planned comparisons (t-tests) were used to confirm differences in oil intakes on individual trials. The probability cut-off for all tests was p < 0 . 0 5 . All values are reported as mean ± SEM.
Surgical and Testing Procedure Rats were adapted to the diets for six days and were then food deprived for 24 hr before implantation of gastric cannula. They were anesthetized with a mixture of ketamine (90 mg/kg) and acepromazine (1 mg/kg), and cannulas were implanted using a method described previously (14). Rats were allowed to recover from surgery for 24 days before sham-feeding trials began. Rats were tested in groups of 9 (4 from one diet group and 5 from the other) on alternate days. One hour before the onset of the dark period on days when rats were not tested, food intake from the previous day was measured and food cups were replenished if necessary. On test days, the same procedure was followed, but the food was not returned until after the sham-feeding trial. For testing, rats were food deprived for the first hour of the dark period and then their stomachs were rinsed with warm water. A drainage tube was attached to the cannula and the rats were placed in a Plexiglas box, and were separated by opaque dividers. The drainage tube, which was fitted through a slot in the bottom of the box, allowed the stomach contents to flow into a preweighed plastic tub. Rats were then given pure corn oil to drink for 30 min (Mazola, Best Foods, Englewood Cliffs, NJ) in 50-ml plastic centrifuge tubes. Oil intake was measured by weighing the tubes (to the nearest 0.1 g) before and after the trial. At the end of the trial, the cannula screw caps were replaced, and the rats were returned to their home cages.
RESULTS
Food Intake and Body Weight There was no difference in body weight between rats fed the HF or HC diets on the day before the beginning of sham-feeding trials (HF = 348 ± 6 g vs. HC = 342 - 7 g). Sham-feeding oil did not affect subsequent food intake of either group (p>0.05), but all rats ate more food over time, F ( 6 , 9 6 ) = 7 . 6 7 , p < 0 . 0 5 . The average daily food intake for rats fed the HF and HC diet was 86.8 ± 6.5 kcals and 91.4--_ 4.3, respectively.
Oil Intake During Sham-Feeding Rats fed the HF diet sham-drank more oil than did those fed the HC diet on Trials I - 4 and Trial 10 (p<0.05; Fig. 1); however, the two diet groups did not differ in oil intake over the entire 17 trials
7 6 0 cq
5 4
Recovery of Ingested Oil
(1)
3
In order to determine the quantity of oil absorbed during sham-feeding, the fluid and oil that drained from the fistulas was collected and each rat's drainage tube and stomach were rinsed at the end of each trial with warm water, which was also collected. The material was dried (60°C) to constant weight. The final weight was subtracted from the initial weight, and this difference was an estimate of the amount of oil recovered. The percentage of oil recovered after sham-feeding was calculated for each trial by dividing the amount of oil recovered by the amount consumed during the trial and multiplying by 100.
E
2 1 0 1
2
3 4 5 6 7
8
9 10 11 12 13 14 15 16 17
Trials
FIG. 1. Corn oil intakes (g/30 min) of sham-feeding rats fed diets high in fat or carbohydrate. *Denotes a significant difference in oil intake between the two dietary groups on that trial. Values are means __.SEM.
SHAM-FEEDING OIL
781
(p>0.05). Both groups sham-drank more oil over successive trials, F(15,240) = 7.59, p<0.05.
Oil Recovery Reliable measurements of oil recovery were obtained in 298 out of 306 trials. Of these measurements, 12 were from trials in which rats did not drink oil. These were not used for the calculation of oil recovery, but did show that the estimated contribution of dried gastric mucous to the total weight of the recovered oil was small (0.11 ~ 0.02 g). Because the quantity of gastric mucous may vary with the amount of oil sham-fed, this value was not used in any further calculations. There was no correlation between the amount of oil ingested during sham-feeding and the percentage recovered (r= .06). The amount of oil unaccounted for (i.e., amount ingested subtracted from the amount recovered) was approximately 4-5 calories per trial but there was no difference in the number of unrecovered calories between the two diet groups (Table 2). DISCUSSION During the first few trials, rats fed the HF diet sham-fed significantly more oil than did those fed the HC diet. With repeated trials, however, the intake of oil by rats in the HC diet group increased, so that after four trials it was similar to that of the HF diet group. This pattern of results differs from that seen in rats that drink oil under normal circumstances (i.e., " r e a l " feeding). When the normal postingestive consequences of the oil are allowed, oil intake of rats fed the HF diet increases with repeated testing but the oil intake of rats fed the HC diet stays low (7). This could be due to postingestive effects of oil which promote oil intake in rats fed the HF diet or suppress oil intake in rats fed the HC diet. When these postingestional effects are prevented by sham-feeding, oil intakes are the same for both diet groups. Why rats fed a HF diet should initially sham-drink more corn oil than those fed an equicaloric HC diet is unclear. Because the HF diet contained corn oil, familiarity with corn oil may have attracted rats fed the HF diet to the oil. This explanation seems
unlikely because rats that have been fed both the HF and HC diets, and are equally familiar with the taste of corn oil, only drink more oil when they are currently eating the HF diet (7). Also, rats fed a corn oil-based HF diet eat more of a variety of fats than do rats fed a HC diet (7,10). We believe the increased acceptance of several forms of fat, along with the present results, suggests that some metabolic or physiological consequence of eating the HF diet predisposes rats to prefer the sensory properties of oil. However, because the effect of diet dissipated with repeated sham-feeding experience, the sensory properties of oil alone do not appear to be sufficient to maintain the difference in oil intake seen when rats "real-feed" oil. The postingestive effects of oil are required as well. This interpretation is consistent with previous studies (1--4, 8, 10-12) which indicate that differences in the capacities to derive calories from oil, or in the metabolic handling of oil, underlie the differential oil intake of rats fed the HF or HC diets. All the sham-ingested oil could not be recovered; up to 20% (4--5 calories) was probably absorbed. Recently, Sclafani and Nissenbaum (9) found that blood glucose concentrations increased after rats with gastric cannulas sham-drank 32% Polycose for 30 min. This implies that the gastric cannula preparation does not completely prevent absorption of nutrients. This might suggest that hungry rats may choose to sham-feed corn oil rather than mineral oil (5) because they receive some caloric benefit. In the present experiment, rats did not reduce their intake of maintenance diet on days when they sham-fed oil relative to days when they did not. This finding suggests that the rats absorbed a behaviorally inconsequential amount of oil. Certainly, the amount absorbed was not sufficient to maintain differential intakes of oil by rats fed the HF and HC diets. Nevertheless, the gastric sham-feeding preparation is imperfect, and the actual contribution of oral and postingestive benefits derived during sham-feeding requires cautious interpretation. ACKNOWLEDGEMENTS This work was supported by the Howard Heinz Endowment. The technical assistance of Frederic Sandier and the comments of Israel Ramirez on this manuscript are gratefully acknowledged.
REFERENCES 1. Arbour, K. J.; Wilkie, D. M. Rodents' (Rattus, Mesocricetur, and Meriones) use of learned caloric information in diet selection. J. Comp. Psychol. 102:177-181; 1988. 2. Bolles, R. C.; Hayward, L.; Crandall, C. Conditioned taste preferences based on caloric density. J. Exp. Psychol. [Anita. Behav.] 7:59-69; 1981. 3. Friedman, M. I.; Ramirez, I.; Edens, N .K.; Granneman, J. Food intake in diabetic rats: isolation of primary metabolic effects of fat feeding. Am. J. Physiol. 249:R44-R51; 1985. 4. Kanarek, R.; Ho, L. Patterns of nutrient selection in rats with streptozotocin-induced diabetes. Physiol. Behav. 32:639-645; 1984. 5. Mindell, S.; Smith, G. P. Rats sham feed corn oil and mineral oil but prefer corn oil. Soc. Neurosci. Abstr. 13:462; 1987. 6. Mook, D. G. Oral and postingestional determinants of the intake of various solutions in rats with esophageal fistulas. J. Comp. Physiol. Psychol. 56:645-659; 1963. 7. Reed, D. R.; Friedman, M. I. Diet composition alters the preference for fat in rats. Appetite, in press; 1989. 8. Richter, C. P.; Schmidt, E. Increased fat and decreased carbohydrate appetite of pancreatectomized rats. Endocrinology 28:179-192; 1941.
9. Sclafani, A.; Nissenhaum, J. W. Oral versus postingestive origin of polysaccharide appetite in the rat. Neurosci. Biobehav. Rev. 11: 169-172; 1987. 10. Tepper, B. J.; Friedman, M. I. Diabetes and a high-fat/low carbohydrate diet enhance the acceptability of oil emulsions to rats, Physiol. Behav. 45:717-721; 1989. ll. Tepper, B. J.; Kanarek, R. B. Selection of protein and fat by diabetic rats following separate dilution of the dietary sources. Physiol. Behav. 45:49-61; 1989. 12. Tordoff, M. G.; Tepper, B. J.; Friedman, M. I. Food flavor preferences produced by drinking glucose and oil in normal and diabetic rats: Evidence for conditioning based on fuel oxidation. Physiol. Behav. 41:481-487; 1987. 13. Weatherford, S. C.; Smith, G. P.; Melville, L. D. D-1 and D-2 receptor antagonists decrease corn oil sham feeding in rats. Physiol. Behav. 44:569-572; 1988. 14. Young, R. C.; Gibbs, J.; Antin, J.; Holt, J.; Smith, G. P. Absence of satiety during sham-feeding in the rat. J. Comp. Physiol. Psychol. 87:795-800; 1974.
0031-9384/90 $3.00 + .00
BRIEF COMMUNICATION
Sham-Feeding of Corn Oil by Rats: Sensory and Postingestive Factors D A N I E L L E R. R E E D l
Yale University and Monell Chemical Senses Center M I C H A E L G . T O R D O F F A N D M A R K I. F R I E D M A N
Monell Chemical Senses Center R e c e i v e d 25 S e p t e m b e r 1989
REED, D. R., M. G. TORDOFF AND M. I. FRIEDMAN. Sham-feeding of corn oil by rats: Sensory and postingestive factors. PHYSIOL BEHAV 47(4) 779--781, 1990.--Previous research indicates that rats fed a high-fat (HF) diet increase their intake and preference for oil compared with rats fed a high-carbohydrate (HC) diet. To assess whether this increased intake was due to the sensory or postingestive properties of oil, rats were adapted to either the HF or HC diet and then allowed to sham-feed pure corn oil daily for 30 rain. During the first 4 trials, rats fed the HF diet sham-fed more oil than did rats fed the HC diet; however, this difference diminished with repeated testing and was absent after 8 trials. In both diet groups, 4-5 calories (approximately 25%) of sham-fed oil could not be recovered and may have escaped to the intestine. These results suggest that, compared with rats fed a HC diet, rats fed a HF diet are initially attracted to the sensory properties of oil, but that the differential oil intakes of rats fed the HF or HC diet are maintained by postingestive, rather than sensory factors. High-fat diet
Food preferences
Oil intake
Fat acceptance
METHOD
RATS fed a high-fat (HF) diet show an increased acceptance of oil compared with rats fed a high-carbohydrate (HC) diet (7,10). Potentially, diet composition might affect the rats' response to either the sensory or postingestive effects of oil and produce the change in oil intake. The sham-feeding preparation is one way to distinguish between these two possibilities. During sham-feeding, food drains out of the esophagus or stomach through a fistula and its postingestive effects are minimized (6, 9, 14). Earlier work shows that rats sham-feed corn oil and mineral oil (5,13), suggesting that the orosensory stimulation produced by ingesting oil is reinforcing. To test whether the sensory or postingestive effects of oil consumption are important in the maintenance of differential oil intake by rats fed different diets, we studied sham-feeding of oil by rats fed HF or HC diets. If the sensory stimulation associated with oil is critical for increased oil intake by rats fed a HF diet, then they should sham-feed more oil than rats fed the HC diet. If, on the other hand, the postingestive effect of oil is important then rats fed the HF and HC diet should sham-feed the same amount of oil.
Subjects Eighteen male Sprague-Dawley rats (Charles River, Wilmington, MA), initially weighing 200--225 g, were used as subjects. They were housed individually in wire mesh hanging cages except during the first I0 days after surgery, when they were housed in solid-bottom cages with wood shavings to prevent the newly implanted cannulas from being pulled out on the wire mesh. Rats were maintained on a 12:12 hr lighting cycle with lights off at 1330 hr. The room temperature was 12-23°C. The protocol was approved by the Animal Care and Use Committee.
Diet Composition, Food Intake and Body Weight For the duration of the experiment, rats were fed nutritionally complete, semisynthetic diets that were either high in fat (HF; N = 9) or carbohydrate content (HC; N = 9; Table 1). The diets were
IRequests for reprints should be addressed to Danielle R. Reed, Monell Chemical Senses Center, 3500 Market St., Philadelphia, PA 19104.
779
780
REED, TORDOFF AND FRIEDMAN
TABLE 1
TABLE 2
COMPOSITION OF DIETS
PERCENT OF INGESTED OIL RECOVERED AND NUMBER OF OIL CALORIES UNACCOUNTEDFOR AFTER SHAM-FEEDING
Ingredients (g/100 g)
High-Fat (HF)
High-Carbohydrate (HC)
Casein Cornstarch Corn Oil Alphacel AIN Mineral Mix AIN Vitamin Mix Choline Bitartrate DL Methionine
20.0 10.4 23.1 41.5 3.5 1.0 0.2 0.3
20.0 52.0 4.6 18.4 3.5 1.0 0.2 0.3
kcal/g % kcal fat
3.3 63.0
3.3 13.0
Diets were prepared and pelleted by ICN Nutritional Biochemicals (Cleveland, OH).
provided in glass jars attached to the rats' cages by springs. Food intake was measured to the nearest 0.1 g (corrected for spillage) daily for the first 17 days of sham-feeding. Tap water was freely available except during the sham-feeding trials. Body weights were measured on the day before the first sham-feeding trial.
% Recovered kcal
High-Fat
High-Carbohydrate
76 _ 5 5.0 --+ 0.3
80 - 5 4.0 _+ 0.2
Differences between diet groups were not statistically significant. Values are means _ SEM.
Data Analysis The difference between the two dietary groups in body weight was analyzed using a t-test. Food intake during the sham-feeding period was analyzed using a mixed design ANOVA with Diet as the between-group factor and Trial (1-17) and Test Condition (Oil vs. No Oil) as the within-group factors. Oil intake during the sham-feeding period was analyzed using a mixed design ANOVA with Diet as the between-group factor and Trial (1-17) as the within-group factor. Planned comparisons (t-tests) were used to confirm differences in oil intakes on individual trials. The probability cut-off for all tests was p < 0 . 0 5 . All values are reported as mean ± SEM.
Surgical and Testing Procedure Rats were adapted to the diets for six days and were then food deprived for 24 hr before implantation of gastric cannula. They were anesthetized with a mixture of ketamine (90 mg/kg) and acepromazine (1 mg/kg), and cannulas were implanted using a method described previously (14). Rats were allowed to recover from surgery for 24 days before sham-feeding trials began. Rats were tested in groups of 9 (4 from one diet group and 5 from the other) on alternate days. One hour before the onset of the dark period on days when rats were not tested, food intake from the previous day was measured and food cups were replenished if necessary. On test days, the same procedure was followed, but the food was not returned until after the sham-feeding trial. For testing, rats were food deprived for the first hour of the dark period and then their stomachs were rinsed with warm water. A drainage tube was attached to the cannula and the rats were placed in a Plexiglas box, and were separated by opaque dividers. The drainage tube, which was fitted through a slot in the bottom of the box, allowed the stomach contents to flow into a preweighed plastic tub. Rats were then given pure corn oil to drink for 30 min (Mazola, Best Foods, Englewood Cliffs, NJ) in 50-ml plastic centrifuge tubes. Oil intake was measured by weighing the tubes (to the nearest 0.1 g) before and after the trial. At the end of the trial, the cannula screw caps were replaced, and the rats were returned to their home cages.
RESULTS
Food Intake and Body Weight There was no difference in body weight between rats fed the HF or HC diets on the day before the beginning of sham-feeding trials (HF = 348 ± 6 g vs. HC = 342 - 7 g). Sham-feeding oil did not affect subsequent food intake of either group (p>0.05), but all rats ate more food over time, F ( 6 , 9 6 ) = 7 . 6 7 , p < 0 . 0 5 . The average daily food intake for rats fed the HF and HC diet was 86.8 ± 6.5 kcals and 91.4--_ 4.3, respectively.
Oil Intake During Sham-Feeding Rats fed the HF diet sham-drank more oil than did those fed the HC diet on Trials I - 4 and Trial 10 (p<0.05; Fig. 1); however, the two diet groups did not differ in oil intake over the entire 17 trials
7 6 0 cq
5 4
Recovery of Ingested Oil
(1)
3
In order to determine the quantity of oil absorbed during sham-feeding, the fluid and oil that drained from the fistulas was collected and each rat's drainage tube and stomach were rinsed at the end of each trial with warm water, which was also collected. The material was dried (60°C) to constant weight. The final weight was subtracted from the initial weight, and this difference was an estimate of the amount of oil recovered. The percentage of oil recovered after sham-feeding was calculated for each trial by dividing the amount of oil recovered by the amount consumed during the trial and multiplying by 100.
E
2 1 0 1
2
3 4 5 6 7
8
9 10 11 12 13 14 15 16 17
Trials
FIG. 1. Corn oil intakes (g/30 min) of sham-feeding rats fed diets high in fat or carbohydrate. *Denotes a significant difference in oil intake between the two dietary groups on that trial. Values are means __.SEM.
SHAM-FEEDING OIL
781
(p>0.05). Both groups sham-drank more oil over successive trials, F(15,240) = 7.59, p<0.05.
Oil Recovery Reliable measurements of oil recovery were obtained in 298 out of 306 trials. Of these measurements, 12 were from trials in which rats did not drink oil. These were not used for the calculation of oil recovery, but did show that the estimated contribution of dried gastric mucous to the total weight of the recovered oil was small (0.11 ~ 0.02 g). Because the quantity of gastric mucous may vary with the amount of oil sham-fed, this value was not used in any further calculations. There was no correlation between the amount of oil ingested during sham-feeding and the percentage recovered (r= .06). The amount of oil unaccounted for (i.e., amount ingested subtracted from the amount recovered) was approximately 4-5 calories per trial but there was no difference in the number of unrecovered calories between the two diet groups (Table 2). DISCUSSION During the first few trials, rats fed the HF diet sham-fed significantly more oil than did those fed the HC diet. With repeated trials, however, the intake of oil by rats in the HC diet group increased, so that after four trials it was similar to that of the HF diet group. This pattern of results differs from that seen in rats that drink oil under normal circumstances (i.e., " r e a l " feeding). When the normal postingestive consequences of the oil are allowed, oil intake of rats fed the HF diet increases with repeated testing but the oil intake of rats fed the HC diet stays low (7). This could be due to postingestive effects of oil which promote oil intake in rats fed the HF diet or suppress oil intake in rats fed the HC diet. When these postingestional effects are prevented by sham-feeding, oil intakes are the same for both diet groups. Why rats fed a HF diet should initially sham-drink more corn oil than those fed an equicaloric HC diet is unclear. Because the HF diet contained corn oil, familiarity with corn oil may have attracted rats fed the HF diet to the oil. This explanation seems
unlikely because rats that have been fed both the HF and HC diets, and are equally familiar with the taste of corn oil, only drink more oil when they are currently eating the HF diet (7). Also, rats fed a corn oil-based HF diet eat more of a variety of fats than do rats fed a HC diet (7,10). We believe the increased acceptance of several forms of fat, along with the present results, suggests that some metabolic or physiological consequence of eating the HF diet predisposes rats to prefer the sensory properties of oil. However, because the effect of diet dissipated with repeated sham-feeding experience, the sensory properties of oil alone do not appear to be sufficient to maintain the difference in oil intake seen when rats "real-feed" oil. The postingestive effects of oil are required as well. This interpretation is consistent with previous studies (1--4, 8, 10-12) which indicate that differences in the capacities to derive calories from oil, or in the metabolic handling of oil, underlie the differential oil intake of rats fed the HF or HC diets. All the sham-ingested oil could not be recovered; up to 20% (4--5 calories) was probably absorbed. Recently, Sclafani and Nissenbaum (9) found that blood glucose concentrations increased after rats with gastric cannulas sham-drank 32% Polycose for 30 min. This implies that the gastric cannula preparation does not completely prevent absorption of nutrients. This might suggest that hungry rats may choose to sham-feed corn oil rather than mineral oil (5) because they receive some caloric benefit. In the present experiment, rats did not reduce their intake of maintenance diet on days when they sham-fed oil relative to days when they did not. This finding suggests that the rats absorbed a behaviorally inconsequential amount of oil. Certainly, the amount absorbed was not sufficient to maintain differential intakes of oil by rats fed the HF and HC diets. Nevertheless, the gastric sham-feeding preparation is imperfect, and the actual contribution of oral and postingestive benefits derived during sham-feeding requires cautious interpretation. ACKNOWLEDGEMENTS This work was supported by the Howard Heinz Endowment. The technical assistance of Frederic Sandier and the comments of Israel Ramirez on this manuscript are gratefully acknowledged.
REFERENCES 1. Arbour, K. J.; Wilkie, D. M. Rodents' (Rattus, Mesocricetur, and Meriones) use of learned caloric information in diet selection. J. Comp. Psychol. 102:177-181; 1988. 2. Bolles, R. C.; Hayward, L.; Crandall, C. Conditioned taste preferences based on caloric density. J. Exp. Psychol. [Anita. Behav.] 7:59-69; 1981. 3. Friedman, M. I.; Ramirez, I.; Edens, N .K.; Granneman, J. Food intake in diabetic rats: isolation of primary metabolic effects of fat feeding. Am. J. Physiol. 249:R44-R51; 1985. 4. Kanarek, R.; Ho, L. Patterns of nutrient selection in rats with streptozotocin-induced diabetes. Physiol. Behav. 32:639-645; 1984. 5. Mindell, S.; Smith, G. P. Rats sham feed corn oil and mineral oil but prefer corn oil. Soc. Neurosci. Abstr. 13:462; 1987. 6. Mook, D. G. Oral and postingestional determinants of the intake of various solutions in rats with esophageal fistulas. J. Comp. Physiol. Psychol. 56:645-659; 1963. 7. Reed, D. R.; Friedman, M. I. Diet composition alters the preference for fat in rats. Appetite, in press; 1989. 8. Richter, C. P.; Schmidt, E. Increased fat and decreased carbohydrate appetite of pancreatectomized rats. Endocrinology 28:179-192; 1941.
9. Sclafani, A.; Nissenhaum, J. W. Oral versus postingestive origin of polysaccharide appetite in the rat. Neurosci. Biobehav. Rev. 11: 169-172; 1987. 10. Tepper, B. J.; Friedman, M. I. Diabetes and a high-fat/low carbohydrate diet enhance the acceptability of oil emulsions to rats, Physiol. Behav. 45:717-721; 1989. ll. Tepper, B. J.; Kanarek, R. B. Selection of protein and fat by diabetic rats following separate dilution of the dietary sources. Physiol. Behav. 45:49-61; 1989. 12. Tordoff, M. G.; Tepper, B. J.; Friedman, M. I. Food flavor preferences produced by drinking glucose and oil in normal and diabetic rats: Evidence for conditioning based on fuel oxidation. Physiol. Behav. 41:481-487; 1987. 13. Weatherford, S. C.; Smith, G. P.; Melville, L. D. D-1 and D-2 receptor antagonists decrease corn oil sham feeding in rats. Physiol. Behav. 44:569-572; 1988. 14. Young, R. C.; Gibbs, J.; Antin, J.; Holt, J.; Smith, G. P. Absence of satiety during sham-feeding in the rat. J. Comp. Physiol. Psychol. 87:795-800; 1974.